Terroir climate change

3Historic and future climate variability

and climate change: effects on vocation, stress and new vine areas

Cra viticoltura_libro 1_capitolo 3.indd 1 03/06/10 15:48

CLIMATE, GRAPES, AND WINE: STRUCTURE AND SUITABILITY IN A VARIABLE AND CHANGING CLIMATE

G.V. Jones(1) (1) Department of Environmental Studies

Southern Oregon University 1250 Siskiyou Blvd

Ashland, Oregon Email: [email protected]

ABSTRACT Climate is a pervasive factor in the success of all agricultural systems, influencing whether a

crop is suitable to a given region, largely controlling crop production and quality, and ultimately driving economic sustainability. Climate’s influence on agribusiness is never more evident than with viticulture and wine production where climate is arguably the most critical aspect in ripening fruit to optimum characteristics to produce a given wine style. Any assessment of climate for wine production must examine a multitude of factors that operate over many temporal and spatial scales. Namely climate influences must be considered at the macroscale (synoptic climate) to the mesoscale (regional climate) to the toposcale (site climate) to the microscale (vine row and canopy climate). In addition, climate influences come from both broad structural conditions and singular weather events manifested through many temperature, precipitation, and moisture parameters. To understand climate’s role in growing winegrapes and wine production one must consider 1) the weather and climate structure necessary for optimum quality and production characteristics, 2) the climate suitability to different winegrape cultivars, 3) the climate’s variability in wine producing regions, and 4) the influence of climate change on the structure, suitability, and variability of climate.

KEYWORDClimate – grapes – wine – temperature – climate change – climate variability

INTRODUCTIONAs a component of terroir, climate arguably exerts the most profound effect on the ability of a

region or site to produce quality grapes and therefore wine. Worldwide, the average climatic conditions of wine regions determine to a large degree the grape varieties that can be grown there, while wine production and quality are chiefly influenced by site-specific factors, husbandry decisions, and short-term climate variability (Jones and Hellman, 2003). Historically there have been numerous temperature-based metrics (e.g., heat summation, degree-days, mean temperature of the warmest month, average growing season temperatures, etc.) that have been used for establishing optimum climates for the range of winegrape cultivars (Gladstones, 1992; (Tonietto and Carbonneau, 2004; Jones, 2006). This is not to say that precipitation or any other weather/climate factor is not important, but that temperature is the most influential factor in overall growth and productivity of winegrapes. At the global scale the general bounds on climate


CLIMATE, GRAPES, AND WINE: STRUCTURE AND SUITABILITY IN A VARIABLE AND CHANGING CLIMATE

G.V. Jones(1) (1) Department of Environmental Studies

Southern Oregon University 1250 Siskiyou Blvd

Ashland, Oregon Email: [email protected]

ABSTRACT Climate is a pervasive factor in the success of all agricultural systems, influencing whether a

crop is suitable to a given region, largely controlling crop production and quality, and ultimately driving economic sustainability. Climate’s influence on agribusiness is never more evident than with viticulture and wine production where climate is arguably the most critical aspect in ripening fruit to optimum characteristics to produce a given wine style. Any assessment of climate for wine production must examine a multitude of factors that operate over many temporal and spatial scales. Namely climate influences must be considered at the macroscale (synoptic climate) to the mesoscale (regional climate) to the toposcale (site climate) to the microscale (vine row and canopy climate). In addition, climate influences come from both broad structural conditions and singular weather events manifested through many temperature, precipitation, and moisture parameters. To understand climate’s role in growing winegrapes and wine production one must consider 1) the weather and climate structure necessary for optimum quality and production characteristics, 2) the climate suitability to different winegrape cultivars, 3) the climate’s variability in wine producing regions, and 4) the influence of climate change on the structure, suitability, and variability of climate.

KEYWORDClimate – grapes – wine – temperature – climate change – climate variability

INTRODUCTIONAs a component of terroir, climate arguably exerts the most profound effect on the ability of a

region or site to produce quality grapes and therefore wine. Worldwide, the average climatic conditions of wine regions determine to a large degree the grape varieties that can be grown there, while wine production and quality are chiefly influenced by site-specific factors, husbandry decisions, and short-term climate variability (Jones and Hellman, 2003). Historically there have been numerous temperature-based metrics (e.g., heat summation, degree-days, mean temperature of the warmest month, average growing season temperatures, etc.) that have been used for establishing optimum climates for the range of winegrape cultivars (Gladstones, 1992; (Tonietto and Carbonneau, 2004; Jones, 2006). This is not to say that precipitation or any other weather/climate factor is not important, but that temperature is the most influential factor in overall growth and productivity of winegrapes. At the global scale the general bounds on climate

3 - 3

VIII INTERNATIONAL TERROIR CONGRESS


suitability for viticulture are found between 12-22°C for the growing season in each hemisphere (Jones, 2007; Fig. 1). As seen in Fig. 1 the 12-22°C climate bounds depict a largely midlatitude suitability for winegrape production, however many sub-tropical to tropical areas at higher elevations also fall within these climate zones. Furthermore, any general depiction of average temperatures will also include large areas that have not been typically associated with winegrape production. This is evident in Fig. 1 where large areas of eastern Europe, western Asia, China, the mid-western and eastern United States, southeastern Argentina, southeastern South Africa, and southern Australia fall within the 12-22°C thresholds. While many of these regions may have the growing season temperatures conducive to growing winegrapes, other limiting factors such as winter minimum temperatures, spring and fall frosts, short growing seasons, and water availability would limit much of the areas mapped to the average conditions. Furthermore, while the vast majority of the world’s wine regions are found within these average growing season climate zones, there are some exceptions. For example, there are defined winegrape growing areas in the United States (Texas, Oklahoma, and the Mississippi delta region), Brazil (São Francisco Valley), and South Africa (Lower Orange River in the Northern Cape) that are warmer than 22°C during their respective growing seasons. However, these regions have different climate risks, have developed viticultural practices to deal with the warmer climates (e.g., two crops per year, irrigation, etc.), or produce table grapes or raisons, and do not necessarily represent the average wine region.

Figure 1: Global wine regions, general climate zones, climate variability mechanisms, and their areas of known influences as described in the text. The wine regions are derived from governmentally defined boundaries (e.g., American Viticultural Areas in the United States, Geographic Indicators in Australia and Brazil, and Wine of Origin Wards in South Africa) or areas under winegrape cultivation identified with remote sensing (e.g., Corine Land Cover for Europe) or aerial imagery (e.g., Canada, Chile, Argentina, and New Zealand). The general climate zones are given by the 12-22°C growing season (Apr-Oct in the Northern Hemisphere and Oct-Apr in the Southern Hemisphere) average temperatures derived from the WorldClim database (Hijmans et al. 2005). ENSO – El Niño Southern Oscillation, PDO – Pacific Decadal Oscillation, NAO – North Atlantic Oscillation, IOD – Indian Ocean Dipole, AO – Arctic Oscillation, AAO – Antarctic Oscillation, SST – Sea Surface Temperatures.

Further refining the climate suitability for many of the world’s most recognizable cultivars, Jones (2006) shows that high quality wine production is more realistically limited to 13-21°C average growing season temperatures (Fig. 2). The climate-maturity zoning in Fig. 2 was developed based upon both climate and plant growth for many cultivars grown in cool to hot regions throughout the world’s benchmark areas for those winegrapes. While many of these cultivars are grown and produce wines outside of the bounds depicted in Fig. 2, these are more bulk wine (high yielding) for the lower end market and do not typically attain the typicity or quality for those same cultivars in their ideal climate. Furthermore, growing season average temperatures below 13°C are typically limited to hybrids or very early ripening cultivars that do not necessarily have large-scale commercial appeal. At the upper limits of climate, some production can also be found with growing season average temperatures greater than 21°C, although it is mostly limited to fortified wines, table grapes and raisons (up to 24°C).

Figure 2: Climate-maturity groupings based on relationships between phenological requirements and growing season average temperatures for high to premium quality wine production in the world's benchmark regions for many of the world’s most common cultivars. The dashed line at the end of the bars indicates that some adjustments may occur as more data become available, but changes of more than +/- 0.2-0.5°C are highly unlikely (Jones, 2006).

An example of cool climate suitability is found with the widely recognized Pinot Noir variety, which is typically grown in regions that span from cool to lower intermediate climates with growing seasons that range from roughly 14.0-16.0°C (e.g., Burgundy or Northern Oregon). The coolest of these is the Tamar Valley of Tasmania, while the warmest is the Russian River Valley of California. Across this 2°C climate niche, Pinot Noir produces the broad style for which is it known with the cooler zones producing lighter, elegant wines and the warmer zones producing more full-bodied, fruit-driven wines. While Pinot Noir can be grown outside the 14.0-16.0°C

3 - 4



suitability for viticulture are found between 12-22°C for the growing season in each hemisphere (Jones, 2007; Fig. 1). As seen in Fig. 1 the 12-22°C climate bounds depict a largely midlatitude suitability for winegrape production, however many sub-tropical to tropical areas at higher elevations also fall within these climate zones. Furthermore, any general depiction of average temperatures will also include large areas that have not been typically associated with winegrape production. This is evident in Fig. 1 where large areas of eastern Europe, western Asia, China, the mid-western and eastern United States, southeastern Argentina, southeastern South Africa, and southern Australia fall within the 12-22°C thresholds. While many of these regions may have the growing season temperatures conducive to growing winegrapes, other limiting factors such as winter minimum temperatures, spring and fall frosts, short growing seasons, and water availability would limit much of the areas mapped to the average conditions. Furthermore, while the vast majority of the world’s wine regions are found within these average growing season climate zones, there are some exceptions. For example, there are defined winegrape growing areas in the United States (Texas, Oklahoma, and the Mississippi delta region), Brazil (São Francisco Valley), and South Africa (Lower Orange River in the Northern Cape) that are warmer than 22°C during their respective growing seasons. However, these regions have different climate risks, have developed viticultural practices to deal with the warmer climates (e.g., two crops per year, irrigation, etc.), or produce table grapes or raisons, and do not necessarily represent the average wine region.

Figure 1: Global wine regions, general climate zones, climate variability mechanisms, and their areas of known influences as described in the text. The wine regions are derived from governmentally defined boundaries (e.g., American Viticultural Areas in the United States, Geographic Indicators in Australia and Brazil, and Wine of Origin Wards in South Africa) or areas under winegrape cultivation identified with remote sensing (e.g., Corine Land Cover for Europe) or aerial imagery (e.g., Canada, Chile, Argentina, and New Zealand). The general climate zones are given by the 12-22°C growing season (Apr-Oct in the Northern Hemisphere and Oct-Apr in the Southern Hemisphere) average temperatures derived from the WorldClim database (Hijmans et al. 2005). ENSO – El Niño Southern Oscillation, PDO – Pacific Decadal Oscillation, NAO – North Atlantic Oscillation, IOD – Indian Ocean Dipole, AO – Arctic Oscillation, AAO – Antarctic Oscillation, SST – Sea Surface Temperatures.

Further refining the climate suitability for many of the world’s most recognizable cultivars, Jones (2006) shows that high quality wine production is more realistically limited to 13-21°C average growing season temperatures (Fig. 2). The climate-maturity zoning in Fig. 2 was developed based upon both climate and plant growth for many cultivars grown in cool to hot regions throughout the world’s benchmark areas for those winegrapes. While many of these cultivars are grown and produce wines outside of the bounds depicted in Fig. 2, these are more bulk wine (high yielding) for the lower end market and do not typically attain the typicity or quality for those same cultivars in their ideal climate. Furthermore, growing season average temperatures below 13°C are typically limited to hybrids or very early ripening cultivars that do not necessarily have large-scale commercial appeal. At the upper limits of climate, some production can also be found with growing season average temperatures greater than 21°C, although it is mostly limited to fortified wines, table grapes and raisons (up to 24°C).

Figure 2: Climate-maturity groupings based on relationships between phenological requirements and growing season average temperatures for high to premium quality wine production in the world's benchmark regions for many of the world’s most common cultivars. The dashed line at the end of the bars indicates that some adjustments may occur as more data become available, but changes of more than +/- 0.2-0.5°C are highly unlikely (Jones, 2006).

An example of cool climate suitability is found with the widely recognized Pinot Noir variety, which is typically grown in regions that span from cool to lower intermediate climates with growing seasons that range from roughly 14.0-16.0°C (e.g., Burgundy or Northern Oregon). The coolest of these is the Tamar Valley of Tasmania, while the warmest is the Russian River Valley of California. Across this 2°C climate niche, Pinot Noir produces the broad style for which is it known with the cooler zones producing lighter, elegant wines and the warmer zones producing more full-bodied, fruit-driven wines. While Pinot Noir can be grown outside the 14.0-16.0°C

3 - 5



growing season average temperature bounds it typically is unripe or overripe and readily loses its typicity. As examples of intermediate to warmer climate cultivars, the noble winegrapes Cabernet Franc and Cabernet Sauvignon, are clearly two of the most widely recognized in the world. The spread of these two cultivars worldwide has produced an assortment of wine styles from quite diverse regions. Fig. 2 shows this wide diversity with both Cabernet Franc and Cabernet Sauvignon having roughly 3.5°C climate ranges, nearly double that of Pinot Noir. Cabernet Franc can be grown in intermediate to warm climates (15.4-19.8°C) as evidenced by its quality production in the Loire Valley of France. Cabernet Sauvignon on the other hand is grown in regions that span from intermediate to hot climates with growing seasons that range from roughly 16.8-20.2°C (e.g., Bordeaux or Napa). The lower climate limit for Cabernet Sauvignon suitability is found in Hawke’s Bay, New Zealand while the upper climate limit is found in Robertson, South Africa.

While the average climate structure in a region determines the broad suitability of winegrape cultivars, climate variability influences issues of production and quality risk associated with how equitable the climate is vintage to vintage. Climate variability in wine regions influences grape and wine production through cold temperature extremes during the winter in some regions, frost frequency and severity during the spring and fall, high temperature events during the summer, extreme rain or hail events, and broad spatial and temporal drought conditions. Climate variability mechanisms that influence wine regions are tied to large scale atmospheric and oceanic interactions that operate at different spatial and temporal scales. The most prominent of these is the large scale Pacific sector El Niño-Southern Oscillation (ENSO), which has broad influences on wine region climates in North America, Australia and New Zealand, South Africa, South America, and Europe (Jones, 2010; Fig. 1). While each of the known climate variability mechanisms reveals some temporal periodicity, increases in climate variability for many wine regions have been observed. Increases in climate variability in a given region would bring about greater risk associated with climate extremes, which in turn would strain the economic viability of wine production in any region.

Both observations and models indicate that climates experience changes in both the mean and the variability of temperatures in wine regions and elsewhere (Jones, 2007). For example, if the change response of a warming climate was only in the mean, then there would be less cold weather and more hot and record hot weather. On the other hand, increases in the temperature variance alone would result in more cold and hot weather and record conditions. However, evidence points to increases in both the mean and variance which would bring about less change for cold weather events and much more hot weather and record hot weather (IPCC, 2007). For example, Schär et al. (2004) demonstrate that the European summer climate might experience a pronounced increase in year-to-year variability in response to greenhouse-gas forcing. Such an increase in variability might be able to explain the unusual European summer 2003, and would strongly affect the incidence of heat waves and droughts in the future. Evidence of changing climate variability in wine regions was also found by Jones (2005) and Jones et al. (2005a) where the coefficient of variability in the growing season climates throughout the western US and many other wine regions globally has increased over the last 50 years. Jones et al. (2005a) also found that model projections through to 2050 show a continued increase in the coefficient of variability in 20 of 27 wine regions globally.

From the discussion above on the climate structure, suitability, and variability associated with regional to global wine production it is clear that viticultural regions are located in relatively

narrow geographical and climatic ranges. In addition, winegrapes have relatively large cultivar differences in climate suitability further limiting some winegrapes to even smaller areas that are appropriate for their cultivation. These narrow niches for optimum quality and production put the cultivation of winegrapes at greater risk from both short-term climate variability and long-term climate changes than more other more broad acre crops (Jones, 2007).

At the global scale, trends in wine region climates have resulted in warmer growing season climates that have allowed many regions to produce better wine, while future climate projections indicate more benefits for some regions and challenges for others. The observed growing season warming rates for numerous wine regions across the globe during 1950-2000 averaged 1.3°C (Jones et al. 2005a), with the warming driven mostly by changes in minimum temperatures, with greater heat accumulation, a decline in frost frequency that is most significant in the dormant period and spring, earlier last spring frosts, later first fall frosts, and longer frost-free periods (Jones, 2005). However, climate model projections by 2050 for the same wine regions predict growing season warming of an additional 1.5-2.5°F on average with spatial analyses showing the potential for relatively large latitudinal shifts in viable viticulture zones with increasing area on the poleward fringe in the Northern Hemisphere (NH) and decreasing area in the Southern Hemisphere (SH) due to the lack of land mass (Jones, 2007). Within regions, spatial shifts are projected to be toward the coast, up in elevation, and to the north (NH) or south (SH). Furthermore, climate variability analyses have shown evidence of increased frequency of extreme events in many regions, while climate models predict a continued increase in variability globally. In addition, phenological changes observed over the last 50 years for numerous locations and varieties globally indicate that grapevines have responded to the observed warming with earlier events (bud break, bloom, véraison, and harvest) and shorter intervals between events that range from 6-17 days depending on variety and location (Jones et al. 2005b).

To place viticulture and wine production in the context of climate suitability and the potential impacts from climate change, Fig. 2 provides the framework for examining today’s climate-maturity ripening potential for premium quality wine varieties grown in cool, intermediate, warm, and hot climates (Jones, 2006). From the general bounds that cool to hot climate suitability places on high quality wine production, it is clear that the impacts of climate change are not likely to be uniform across all varieties and regions, but are more likely to be related to climatic thresholds whereby any continued warming would push a region outside the ability to produce quality wine with existing varieties. For example, if a region has an average growing season temperature of 15°C and the climate warms by 1°C, then that region is climatically more conducive to ripening some varieties, while potentially less for others. If the magnitude of the warming is 2°C or larger, then a region may potentially shift into another climate maturity type (e.g., from intermediate to warm). While the range of potential varieties that a region can ripen will expand in many cases, if a region is a hot climate maturity type and warms beyond what is considered viable, then grape growing becomes challenging and maybe even impossible.

CONCLUSIONS Overall, winegrapes are a climatically sensitive crop whereby quality production is achieved

across a fairly narrow geographic range. In addition, winegrapes are grown largely in mid-latitude regions that are prone to high climatic variability that drive relatively large vintage differences in quality and productivity. However, while understanding the climate structure and variability in

3 - 6



growing season average temperature bounds it typically is unripe or overripe and readily loses its typicity. As examples of intermediate to warmer climate cultivars, the noble winegrapes Cabernet Franc and Cabernet Sauvignon, are clearly two of the most widely recognized in the world. The spread of these two cultivars worldwide has produced an assortment of wine styles from quite diverse regions. Fig. 2 shows this wide diversity with both Cabernet Franc and Cabernet Sauvignon having roughly 3.5°C climate ranges, nearly double that of Pinot Noir. Cabernet Franc can be grown in intermediate to warm climates (15.4-19.8°C) as evidenced by its quality production in the Loire Valley of France. Cabernet Sauvignon on the other hand is grown in regions that span from intermediate to hot climates with growing seasons that range from roughly 16.8-20.2°C (e.g., Bordeaux or Napa). The lower climate limit for Cabernet Sauvignon suitability is found in Hawke’s Bay, New Zealand while the upper climate limit is found in Robertson, South Africa.

While the average climate structure in a region determines the broad suitability of winegrape cultivars, climate variability influences issues of production and quality risk associated with how equitable the climate is vintage to vintage. Climate variability in wine regions influences grape and wine production through cold temperature extremes during the winter in some regions, frost frequency and severity during the spring and fall, high temperature events during the summer, extreme rain or hail events, and broad spatial and temporal drought conditions. Climate variability mechanisms that influence wine regions are tied to large scale atmospheric and oceanic interactions that operate at different spatial and temporal scales. The most prominent of these is the large scale Pacific sector El Niño-Southern Oscillation (ENSO), which has broad influences on wine region climates in North America, Australia and New Zealand, South Africa, South America, and Europe (Jones, 2010; Fig. 1). While each of the known climate variability mechanisms reveals some temporal periodicity, increases in climate variability for many wine regions have been observed. Increases in climate variability in a given region would bring about greater risk associated with climate extremes, which in turn would strain the economic viability of wine production in any region.

Both observations and models indicate that climates experience changes in both the mean and the variability of temperatures in wine regions and elsewhere (Jones, 2007). For example, if the change response of a warming climate was only in the mean, then there would be less cold weather and more hot and record hot weather. On the other hand, increases in the temperature variance alone would result in more cold and hot weather and record conditions. However, evidence points to increases in both the mean and variance which would bring about less change for cold weather events and much more hot weather and record hot weather (IPCC, 2007). For example, Schär et al. (2004) demonstrate that the European summer climate might experience a pronounced increase in year-to-year variability in response to greenhouse-gas forcing. Such an increase in variability might be able to explain the unusual European summer 2003, and would strongly affect the incidence of heat waves and droughts in the future. Evidence of changing climate variability in wine regions was also found by Jones (2005) and Jones et al. (2005a) where the coefficient of variability in the growing season climates throughout the western US and many other wine regions globally has increased over the last 50 years. Jones et al. (2005a) also found that model projections through to 2050 show a continued increase in the coefficient of variability in 20 of 27 wine regions globally.

From the discussion above on the climate structure, suitability, and variability associated with regional to global wine production it is clear that viticultural regions are located in relatively

narrow geographical and climatic ranges. In addition, winegrapes have relatively large cultivar differences in climate suitability further limiting some winegrapes to even smaller areas that are appropriate for their cultivation. These narrow niches for optimum quality and production put the cultivation of winegrapes at greater risk from both short-term climate variability and long-term climate changes than more other more broad acre crops (Jones, 2007).

At the global scale, trends in wine region climates have resulted in warmer growing season climates that have allowed many regions to produce better wine, while future climate projections indicate more benefits for some regions and challenges for others. The observed growing season warming rates for numerous wine regions across the globe during 1950-2000 averaged 1.3°C (Jones et al. 2005a), with the warming driven mostly by changes in minimum temperatures, with greater heat accumulation, a decline in frost frequency that is most significant in the dormant period and spring, earlier last spring frosts, later first fall frosts, and longer frost-free periods (Jones, 2005). However, climate model projections by 2050 for the same wine regions predict growing season warming of an additional 1.5-2.5°F on average with spatial analyses showing the potential for relatively large latitudinal shifts in viable viticulture zones with increasing area on the poleward fringe in the Northern Hemisphere (NH) and decreasing area in the Southern Hemisphere (SH) due to the lack of land mass (Jones, 2007). Within regions, spatial shifts are projected to be toward the coast, up in elevation, and to the north (NH) or south (SH). Furthermore, climate variability analyses have shown evidence of increased frequency of extreme events in many regions, while climate models predict a continued increase in variability globally. In addition, phenological changes observed over the last 50 years for numerous locations and varieties globally indicate that grapevines have responded to the observed warming with earlier events (bud break, bloom, véraison, and harvest) and shorter intervals between events that range from 6-17 days depending on variety and location (Jones et al. 2005b).

To place viticulture and wine production in the context of climate suitability and the potential impacts from climate change, Fig. 2 provides the framework for examining today’s climate-maturity ripening potential for premium quality wine varieties grown in cool, intermediate, warm, and hot climates (Jones, 2006). From the general bounds that cool to hot climate suitability places on high quality wine production, it is clear that the impacts of climate change are not likely to be uniform across all varieties and regions, but are more likely to be related to climatic thresholds whereby any continued warming would push a region outside the ability to produce quality wine with existing varieties. For example, if a region has an average growing season temperature of 15°C and the climate warms by 1°C, then that region is climatically more conducive to ripening some varieties, while potentially less for others. If the magnitude of the warming is 2°C or larger, then a region may potentially shift into another climate maturity type (e.g., from intermediate to warm). While the range of potential varieties that a region can ripen will expand in many cases, if a region is a hot climate maturity type and warms beyond what is considered viable, then grape growing becomes challenging and maybe even impossible.

CONCLUSIONS Overall, winegrapes are a climatically sensitive crop whereby quality production is achieved

across a fairly narrow geographic range. In addition, winegrapes are grown largely in mid-latitude regions that are prone to high climatic variability that drive relatively large vintage differences in quality and productivity. However, while understanding the climate structure and variability in

3 - 7



wine regions worldwide provides more exacting varietal selections and vintage to vintage production management, the projected rate and magnitude of future climate change will likely bring about numerous potential impacts for the wine industry, including – added pressure on increasingly scarce water supplies, additional changes in grapevine phenological timing, further disruption or alteration of balanced composition in grapes and wine, regionally-specific needs to change the types of varieties grown, necessary shifts in regional wine styles, and spatial changes in viable grape growing regions. While uncertainty exists in the exact rate and magnitude of climate change in the future, it would be advantageous for the wine industry to be proactive in assessing the impacts, invest in appropriate plant breeding and genetic research, be ready to adopt suitable adaptation strategies, be willing to alter varieties and management practices or controls, or mitigate wine quality differences by developing new technologies.

BIBLIOGRAPHY

Gladstones, J., 1992. Viticulture and Environment: Winetitles, Adelaide, Australia. 310 pp. Hijmans, R. J., S. E. Cameron, J. L. Parra, P. G. Jones and A. Jarvis. 2005. Very high resolution

interpolated climate surfaces for global land areas. International Journal of Climatology 25: 1965-1978.

IPCC 2007. Alley R. et al. Climate Change 2007: The Physical Science Basis. Summary for Policymakers. Contribution of the Working Group I to the Fourth Assessment of the Intergovernmental Panel on Climate Change. IPCC Secretariat (http://www.ipcc.ch/).

Jones, G.V. and E. Hellman. 2003. Site Assessment: in “Oregon Viticulture” Hellman, E. (ed.), 5th Edition, Oregon State University Press, Corvallis, Oregon, p 44-50.

Jones, G.V., 2005. Climate change in the western United States grape growing regions. Acta Horticulturae (ISHS), 689:41-60.

Jones, G.V., M.A. White, O.R. Cooper, and K. Storchmann. 2005a. Climate Change and Global Wine Quality. Climatic Change, 73(3): 319-343.

Jones, G.V., E. Duchêne, D. Tomasi, J. Yuste, O. Braslavksa, H. Schultz, C. Martinez, S. Boso, F. Langellier, C. Perruchot, and G. Guimberteau. 2005b. Changes in European Winegrape Phenology and Relationships with Climate, GESCO Proceedings, August 2005.

Jones, G.V. 2006. Climate and Terroir: Impacts of Climate Variability and Change on Wine. In Fine Wine and Terroir - The Geoscience Perspective. Macqueen, R. W., and L. D. Meinert, (eds.), Geoscience Canada Reprint Series Number 9, Geological Association of Canada, St. John's, Newfoundland, 247 pages.

Jones, G.V. 2007. Climate Change and the global wine industry. Proceedings from the 13th

Australian Wine Industry Technical Conference, Adelaide, Australia. Jones, G.V., Ried, R., and A. Vilks 2010. A Climate for Wine. In “The Geography of Wine”,

edited by P. Dougherty. Springer Press, (in press). Schär, C., P.L. Vidale, D. Lüthi, C. Frei, C. Häberli, M.A. Liniger, and C. Appenzeller. 2004. The

role of increasing temperature variability for European summer heat waves. Nature, 427, 332-336.

Tonietto, J. and A. Carbonneau. 2004. A multicriteria climatic classification system for grape-growing regions worldwide. Agric. Forest Meteorol. 124, 81-97.

THE IMPACT OF GLOBAL WARMING ON ONTARIO’S ICEWINEINDUSTRY

D.Cyr(1) and T.B. Shaw(2)

(1) Department of Finance, Operations and Information Systems & Cool Climate Oenology andViticulture Institute, Brock University

St. Catharines Ontario, Canada, L2S [email protected]

(2) Department of Geography &Cool Climate Oenology and Viticulture InstituteBrock University, St. Catharines Ontario, Canada, L2S 3A1

Corresponding author: [email protected], Ext. 3866

ABSTRACTOntario’s wine regions lie at the climatic margins of commercial viticulture owing to their

cold winters and short cool growing season. The gradual warming of northern latitudesprojected under a human-induced climate change scenario could bring mixed benefits tothese wine regions. On the one hand, climate change could moderate the severity of wintertemperatures and extend the growing season and on the other, it could be jeopardize theproduction of internationally renowned icewines for which Canada is famous. This paperexamines the trends in winter temperatures over the last forty years for the NiagaraPeninsula wine region in Ontario. The study analyzes the occurrences of temperatures ≤ -8o

C in the months of November, December, January and February in which the frozen grapesare normally picked. The results of trend analysis showed a high degree of variability alongwith a weak declining trend in the number of picking days. Two major risks to icewinegrapes are prolonged warm and wet conditions that could lead to rot and secondly,destruction of the crop by bird predators. The study also discussed the potential use ofweather contracts to mitigate these risks.

Key Words: climate change, Ontario, icewine, impacts, weather contracts

INTRODUCTIONOntario’s main wine regions comprise the Niagara Peninsula and the adjacent regions of

Lake Erie Northshore, Pelee Island and Prince Edward County. Although the Great Lakesmoderate their climates throughout the year, these areas are often incorrectly perceived asbeing on the climatic limits of successful commercial viticulture, owing to their snowywinters and short cool growing seasons. Endowed with a favourable range of mesoclimates,topographies and soils, wine production has evolved slowly under a scrupulous system ofsite selection matched by suitable cold-tolerant international grape varieties. Still andsparkling wines of quality and distinction are produced in a wide range of styles mainly forthe Canadian market. However, it is the icewine that uniquely combines the regions’climatic, viticulture and oenological attributes that first brought international recognition tothe Canadian wine industry. Although cold winters are the norm for this semi-continentalclimate, global warming could threaten the stability of icewine production. On one hand,these regions are likely to benefit from a warmer climate that could extend the growingseason and moderate the severity of winter temperatures. On the other, the production of theinternationally renowned icewines could be jeopardized by unpredictably warmer spells

3 - 8



wine regions worldwide provides more exacting varietal selections and vintage to vintage production management, the projected rate and magnitude of future climate change will likely bring about numerous potential impacts for the wine industry, including – added pressure on increasingly scarce water supplies, additional changes in grapevine phenological timing, further disruption or alteration of balanced composition in grapes and wine, regionally-specific needs to change the types of varieties grown, necessary shifts in regional wine styles, and spatial changes in viable grape growing regions. While uncertainty exists in the exact rate and magnitude of climate change in the future, it would be advantageous for the wine industry to be proactive in assessing the impacts, invest in appropriate plant breeding and genetic research, be ready to adopt suitable adaptation strategies, be willing to alter varieties and management practices or controls, or mitigate wine quality differences by developing new technologies.

BIBLIOGRAPHY

Gladstones, J., 1992. Viticulture and Environment: Winetitles, Adelaide, Australia. 310 pp. Hijmans, R. J., S. E. Cameron, J. L. Parra, P. G. Jones and A. Jarvis. 2005. Very high resolution

interpolated climate surfaces for global land areas. International Journal of Climatology 25: 1965-1978.

IPCC 2007. Alley R. et al. Climate Change 2007: The Physical Science Basis. Summary for Policymakers. Contribution of the Working Group I to the Fourth Assessment of the Intergovernmental Panel on Climate Change. IPCC Secretariat (http://www.ipcc.ch/).

Jones, G.V. and E. Hellman. 2003. Site Assessment: in “Oregon Viticulture” Hellman, E. (ed.), 5th Edition, Oregon State University Press, Corvallis, Oregon, p 44-50.

Jones, G.V., 2005. Climate change in the western United States grape growing regions. Acta Horticulturae (ISHS), 689:41-60.

Jones, G.V., M.A. White, O.R. Cooper, and K. Storchmann. 2005a. Climate Change and Global Wine Quality. Climatic Change, 73(3): 319-343.

Jones, G.V., E. Duchêne, D. Tomasi, J. Yuste, O. Braslavksa, H. Schultz, C. Martinez, S. Boso, F. Langellier, C. Perruchot, and G. Guimberteau. 2005b. Changes in European Winegrape Phenology and Relationships with Climate, GESCO Proceedings, August 2005.

Jones, G.V. 2006. Climate and Terroir: Impacts of Climate Variability and Change on Wine. In Fine Wine and Terroir - The Geoscience Perspective. Macqueen, R. W., and L. D. Meinert, (eds.), Geoscience Canada Reprint Series Number 9, Geological Association of Canada, St. John's, Newfoundland, 247 pages.

Jones, G.V. 2007. Climate Change and the global wine industry. Proceedings from the 13th

Australian Wine Industry Technical Conference, Adelaide, Australia. Jones, G.V., Ried, R., and A. Vilks 2010. A Climate for Wine. In “The Geography of Wine”,

edited by P. Dougherty. Springer Press, (in press). Schär, C., P.L. Vidale, D. Lüthi, C. Frei, C. Häberli, M.A. Liniger, and C. Appenzeller. 2004. The

role of increasing temperature variability for European summer heat waves. Nature, 427, 332-336.

Tonietto, J. and A. Carbonneau. 2004. A multicriteria climatic classification system for grape-growing regions worldwide. Agric. Forest Meteorol. 124, 81-97.

THE IMPACT OF GLOBAL WARMING ON ONTARIO’S ICEWINEINDUSTRY

D.Cyr(1) and T.B. Shaw(2)

(1) Department of Finance, Operations and Information Systems & Cool Climate Oenology andViticulture Institute, Brock University

St. Catharines Ontario, Canada, L2S [email protected]

(2) Department of Geography &Cool Climate Oenology and Viticulture InstituteBrock University, St. Catharines Ontario, Canada, L2S 3A1

Corresponding author: [email protected], Ext. 3866

ABSTRACTOntario’s wine regions lie at the climatic margins of commercial viticulture owing to their

cold winters and short cool growing season. The gradual warming of northern latitudesprojected under a human-induced climate change scenario could bring mixed benefits tothese wine regions. On the one hand, climate change could moderate the severity of wintertemperatures and extend the growing season and on the other, it could be jeopardize theproduction of internationally renowned icewines for which Canada is famous. This paperexamines the trends in winter temperatures over the last forty years for the NiagaraPeninsula wine region in Ontario. The study analyzes the occurrences of temperatures ≤ -8o

C in the months of November, December, January and February in which the frozen grapesare normally picked. The results of trend analysis showed a high degree of variability alongwith a weak declining trend in the number of picking days. Two major risks to icewinegrapes are prolonged warm and wet conditions that could lead to rot and secondly,destruction of the crop by bird predators. The study also discussed the potential use ofweather contracts to mitigate these risks.

Key Words: climate change, Ontario, icewine, impacts, weather contracts

INTRODUCTIONOntario’s main wine regions comprise the Niagara Peninsula and the adjacent regions of

Lake Erie Northshore, Pelee Island and Prince Edward County. Although the Great Lakesmoderate their climates throughout the year, these areas are often incorrectly perceived asbeing on the climatic limits of successful commercial viticulture, owing to their snowywinters and short cool growing seasons. Endowed with a favourable range of mesoclimates,topographies and soils, wine production has evolved slowly under a scrupulous system ofsite selection matched by suitable cold-tolerant international grape varieties. Still andsparkling wines of quality and distinction are produced in a wide range of styles mainly forthe Canadian market. However, it is the icewine that uniquely combines the regions’climatic, viticulture and oenological attributes that first brought international recognition tothe Canadian wine industry. Although cold winters are the norm for this semi-continentalclimate, global warming could threaten the stability of icewine production. On one hand,these regions are likely to benefit from a warmer climate that could extend the growingseason and moderate the severity of winter temperatures. On the other, the production of theinternationally renowned icewines could be jeopardized by unpredictably warmer spells

3 - 9



2

during the first half of winter, when frozen grapes are typically harvested, after havingexperienced a number of freeze-thaw cycles.

Climate Change and ViticultureThe Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report in

2007 stated that there has been anthropogenic warming over the last 50 years averaged overeach continent. The temperature increase is widespread over the globe, but is greater athigher northern latitudes. Average Arctic temperatures have increased at almost twice theglobal average rate in the past 100 years. Observed decreases in snow and ice extent arealso consistent with warming. All of North America is projected to warm during thiscentury, and the annual mean warming is likely to exceed the global mean warming in mostareas. In northern regions, warming is likely to be largest in winter. It is very likely that colddays, cold nights and frosts could become less frequent over most land areas, while hot daysand hot nights could become more frequent. Also predicted are is a very likely decrease insnow season length and snow depth in most of Europe and North America. (IPCC, 2007).

There exits now several studies that have analyzed the impacts of climate change andvariability on viticulture (Kenny and Harrison, 1992; Stock et al., 2004; Jones andGoodrich, 2008), while others have examined temperatures trends (including the growingseason average, maximum and minimum temperatures) the climatic characteristics ofparticular wine regions, grapevine phenology, grape composition and yield, and theresulting wine quality (Jones and Davis, 2000; Bindi et al.,1996). Several studies haveprojected a continued warming trend in the growing season, more frequent extremetemperatures, early budburst and an increase in the frost-free- period (Easterling et. al.,2000; Kramer, 1994; Duchene and Schneider, 2004).

In Canada, most northern agricultural regions are expected to experience warmerconditions, longer frost-free seasons and increased evapotranspiration. The national averagetemperature for the winter 2009/2010 was 4.0°C above normal, based on preliminary data,which makes this the warmest winter on record since nationwide records began in 1948.The previous record was 2005/2006 which was 3.9°C above normal. The climate isgradually becoming wetter and warmer in southern Canada throughout the twentiethcentury and in all of Canada during the latter half of the century. In southern Canada, springtemperatures have increased, greatly shortening the period of freezing temperatures suitablefor snowfall while daily minimum temperatures, indicators of night-time temperatures, haveincreased significantly over the past century (Zhang et.al, 2002).

The severity of winters has determined the distribution of perennial fruit and vine crops,but warmer winters are not altogether beneficial. Winter damage could actually increase ineastern Canada, due to reduced cold hardening during the fall, an increase in the frequencyof winter freeze-thaw events, and a decrease in protective snow cover. Conversely, vine andorchard crops are expected to benefit from a decreased risk of winter damage. However,milder winter temperatures would reduce cold stress, while a decrease in late spring frostswould lower the risk of bud damage in many regions. Nonetheless, an increase in winterfree-thaw events would decrease the hardiness of the trees, and increase their sensitivity tocold temperatures in late winter (Natural Resources Canada, 2002). Also damaging toperennial crops are temperature fluctuations within the winter season (repeated freeze-thawcycles) and the annual variability that are characteristic of the climate of the Great LakeRegion. Fig. 1 shows winter departures from the normal temperature with an increasing buthighly fluctuating trend.

3

Winter Departures and Long-Term Trends for the Great Lakes Region1948-2010

-3

-2

-1

0

1

2

3

4

5

1948

1950

1952

1954

1956

1958

1960

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

Year

Deg

ree

C

Figure. 1: Winter departures and long term trends in the Great Lakes region 1948 -2010

IcewineThe first icewine (Eiswein) is believed to have been made in Germany around 1794 in

Franconia with the advent of early freezing temperatures before the grape crop could beharvested. Nonetheless, winemakers harvested and pressed the frozen grapes and fermentedthe juice to produce a sweet wine. Germany continued to produce icewine in an unregulatedfashion until 1982 when it was formally granted its own quality category in the GermanWine Law. Unlike the Canadian winters, the already moderate European winters nowbecoming increasingly warmer with climate change make icewine production in Germanyand Austria highly risky.

In Ontario, icewines are made principally from the Vidal, Riesling, Cabernet Franc andGewürztraminer grapes that must harvested in a frozen state at temperatures ≤ -8o C afterNovember 15. The temperatures at picking time are limited from -8o C to -14o C, which alsodetermines the maximum and minimum amount of juice that can be extracted. The optimumharvesting temperature is between -10o and -12o C. The ideal parameters for the pressedjuice are between 38o and 42o Brix, between 120 and 150 L per ton according to the variety,titratable acidity of 10-12 g/l tartaric acid and pH between 3.1 and 3.3 (Ziraldo and Kaiser,2007). According to the Vintners Quality Alliance Ontario (VQA), the finished wine musthave a Brix of 35o or more, residual sugar of 125 g/l and a minimum alcohol content of 7percent but not exceeding 14.9 percent by volume. Riesling and Vidal icewines exhibittypically rich aromas and flavours that are characteristic of ripe tropical fruits such aslychee, papaya and pineapple. The sweet but firm acidity of icewines, attributed to malicacid dominance, make them perfectly balanced.

This paper examines the trends in winter temperatures over the last forty years for theNiagara Peninsula Wine Region of Ontario that is also Canada’s major producer of

3 - 10



2

during the first half of winter, when frozen grapes are typically harvested, after havingexperienced a number of freeze-thaw cycles.

Climate Change and ViticultureThe Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report in

2007 stated that there has been anthropogenic warming over the last 50 years averaged overeach continent. The temperature increase is widespread over the globe, but is greater athigher northern latitudes. Average Arctic temperatures have increased at almost twice theglobal average rate in the past 100 years. Observed decreases in snow and ice extent arealso consistent with warming. All of North America is projected to warm during thiscentury, and the annual mean warming is likely to exceed the global mean warming in mostareas. In northern regions, warming is likely to be largest in winter. It is very likely that colddays, cold nights and frosts could become less frequent over most land areas, while hot daysand hot nights could become more frequent. Also predicted are is a very likely decrease insnow season length and snow depth in most of Europe and North America. (IPCC, 2007).

There exits now several studies that have analyzed the impacts of climate change andvariability on viticulture (Kenny and Harrison, 1992; Stock et al., 2004; Jones andGoodrich, 2008), while others have examined temperatures trends (including the growingseason average, maximum and minimum temperatures) the climatic characteristics ofparticular wine regions, grapevine phenology, grape composition and yield, and theresulting wine quality (Jones and Davis, 2000; Bindi et al.,1996). Several studies haveprojected a continued warming trend in the growing season, more frequent extremetemperatures, early budburst and an increase in the frost-free- period (Easterling et. al.,2000; Kramer, 1994; Duchene and Schneider, 2004).

In Canada, most northern agricultural regions are expected to experience warmerconditions, longer frost-free seasons and increased evapotranspiration. The national averagetemperature for the winter 2009/2010 was 4.0°C above normal, based on preliminary data,which makes this the warmest winter on record since nationwide records began in 1948.The previous record was 2005/2006 which was 3.9°C above normal. The climate isgradually becoming wetter and warmer in southern Canada throughout the twentiethcentury and in all of Canada during the latter half of the century. In southern Canada, springtemperatures have increased, greatly shortening the period of freezing temperatures suitablefor snowfall while daily minimum temperatures, indicators of night-time temperatures, haveincreased significantly over the past century (Zhang et.al, 2002).

The severity of winters has determined the distribution of perennial fruit and vine crops,but warmer winters are not altogether beneficial. Winter damage could actually increase ineastern Canada, due to reduced cold hardening during the fall, an increase in the frequencyof winter freeze-thaw events, and a decrease in protective snow cover. Conversely, vine andorchard crops are expected to benefit from a decreased risk of winter damage. However,milder winter temperatures would reduce cold stress, while a decrease in late spring frostswould lower the risk of bud damage in many regions. Nonetheless, an increase in winterfree-thaw events would decrease the hardiness of the trees, and increase their sensitivity tocold temperatures in late winter (Natural Resources Canada, 2002). Also damaging toperennial crops are temperature fluctuations within the winter season (repeated freeze-thawcycles) and the annual variability that are characteristic of the climate of the Great LakeRegion. Fig. 1 shows winter departures from the normal temperature with an increasing buthighly fluctuating trend.

3

Winter Departures and Long-Term Trends for the Great Lakes Region1948-2010

-3

-2

-1

0

1

2

3

4

5

1948

1950

1952

1954

1956

1958

1960

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

Year

Deg

ree

C

Figure. 1: Winter departures and long term trends in the Great Lakes region 1948 -2010

IcewineThe first icewine (Eiswein) is believed to have been made in Germany around 1794 in

Franconia with the advent of early freezing temperatures before the grape crop could beharvested. Nonetheless, winemakers harvested and pressed the frozen grapes and fermentedthe juice to produce a sweet wine. Germany continued to produce icewine in an unregulatedfashion until 1982 when it was formally granted its own quality category in the GermanWine Law. Unlike the Canadian winters, the already moderate European winters nowbecoming increasingly warmer with climate change make icewine production in Germanyand Austria highly risky.

In Ontario, icewines are made principally from the Vidal, Riesling, Cabernet Franc andGewürztraminer grapes that must harvested in a frozen state at temperatures ≤ -8o C afterNovember 15. The temperatures at picking time are limited from -8o C to -14o C, which alsodetermines the maximum and minimum amount of juice that can be extracted. The optimumharvesting temperature is between -10o and -12o C. The ideal parameters for the pressedjuice are between 38o and 42o Brix, between 120 and 150 L per ton according to the variety,titratable acidity of 10-12 g/l tartaric acid and pH between 3.1 and 3.3 (Ziraldo and Kaiser,2007). According to the Vintners Quality Alliance Ontario (VQA), the finished wine musthave a Brix of 35o or more, residual sugar of 125 g/l and a minimum alcohol content of 7percent but not exceeding 14.9 percent by volume. Riesling and Vidal icewines exhibittypically rich aromas and flavours that are characteristic of ripe tropical fruits such aslychee, papaya and pineapple. The sweet but firm acidity of icewines, attributed to malicacid dominance, make them perfectly balanced.

This paper examines the trends in winter temperatures over the last forty years for theNiagara Peninsula Wine Region of Ontario that is also Canada’s major producer of

3 - 11



4

icewines. The study analyzes the occurrences of temperatures ≤ -8o C in the months ofNovember, December, January and February in which the frozen grapes are normallypicked. Even though the grapes may be partially frozen on several occasions at higherminimum temperatures, the temperature at harvest must fall to -8oC in order to be certifiedas icewine according to the Vintners Quality Alliance (VQA) standards. The idealconditions should include dry and moderately cold conditions over a two week period withpartial freezing, followed by a real freeze with daytime temperatures between -10o C to -12oC in the months of December and January.

DATA AND METHODSTo achieve the above objectives, the study examined daily climatic data for a

representative climatic station located at Vineland in the Niagara Peninsula wine region ofOntario. This climatic station has a long history of reliable data going back to the 1880s andis located in an area that has remained essentially rural. This study utilized data for the 1977to 2008 period retrieved from the Environment Canada’s Online Climate Database. Datafrom this official source typically undergo rigorous quality checks before being released forpublic use. This period of analysis contains no missing records. The study examined thedaily minimum temperatures for the months of December, January and February and thelatter half of November. It included any day with minimum temperature ≤ -8o C, the legallyrecognized lower threshold temperature that ensures that the grape is fully frozen on thevine for icewine.

The analysis of the occurrences of days with the threshold minimum temperatures forharvesting icewine grapes does not account for the length of the freeze event. For example,an event with ≤ -8o C could have lasted under an hour while another event could have lastedfor several hours. Records of hourly values suitable for icewine only began about fifteenyears ago for the Niagara Peninsula wine region and less than ten years for the Lake ErieNorth Shore. In absence of long-term hourly values and for the purpose of this study, wedefined an icewine event as one in which there were two consecutive days with minimumtemperatures ≤ -8o C. To determine any evidence of long-term trend and variability in thedata, we analyzed the time series using simple linear regression. We analyzed the totalnumber of events for individual months and the combined total for each winter season.

RESULTS AND DISCUSSIONOptimal Harvesting Period

November 15 marks the start of the official harvesting date of grapes for icewines.However, the record over the last forty years showed a total of twelve days experiencedminimum temperatures ≤ -8o C. Only one year (1989) had four consecutive days with ≤ -8o

C that were suitable for harvesting frozen grapes. An early harvest date means higher yields,since loss to predators, a longer hanging time and spoilage due to warmer temperatures canreduce yields and quality. Fig. 2 and 3 show the most suitable period for harvesting frozengrapes. It begins roughly from the beginning of the second week in January and extends toapproximately the middle of February. This period has a slightly >50% chance of theoccurrence of freezing temperatures ≤ -8o C.

5

Figure 2. Average daily minimum temperatures for December to February for theperiod 1970-2007

Figure 3. Probability of daily minimum temperatures ≤ -8o C for the period 1970-2007

Trends in Freezing TemperaturesSome winegrowers will harvest at -8o C, while others may await lower temperatures over

a longer. An analysis of the frequency of days (Fig. 4) with freezing temperatures ≤ -8o Cfor the December to February period shows a high degree of inter-annual variability with aweak declining trend for the 1970-2007 period. Values range from a high of 50 in 1978 to alow of 11 in 2002 duration.

3 - 12



4

icewines. The study analyzes the occurrences of temperatures ≤ -8o C in the months ofNovember, December, January and February in which the frozen grapes are normallypicked. Even though the grapes may be partially frozen on several occasions at higherminimum temperatures, the temperature at harvest must fall to -8oC in order to be certifiedas icewine according to the Vintners Quality Alliance (VQA) standards. The idealconditions should include dry and moderately cold conditions over a two week period withpartial freezing, followed by a real freeze with daytime temperatures between -10o C to -12oC in the months of December and January.

DATA AND METHODSTo achieve the above objectives, the study examined daily climatic data for a

representative climatic station located at Vineland in the Niagara Peninsula wine region ofOntario. This climatic station has a long history of reliable data going back to the 1880s andis located in an area that has remained essentially rural. This study utilized data for the 1977to 2008 period retrieved from the Environment Canada’s Online Climate Database. Datafrom this official source typically undergo rigorous quality checks before being released forpublic use. This period of analysis contains no missing records. The study examined thedaily minimum temperatures for the months of December, January and February and thelatter half of November. It included any day with minimum temperature ≤ -8o C, the legallyrecognized lower threshold temperature that ensures that the grape is fully frozen on thevine for icewine.

The analysis of the occurrences of days with the threshold minimum temperatures forharvesting icewine grapes does not account for the length of the freeze event. For example,an event with ≤ -8o C could have lasted under an hour while another event could have lastedfor several hours. Records of hourly values suitable for icewine only began about fifteenyears ago for the Niagara Peninsula wine region and less than ten years for the Lake ErieNorth Shore. In absence of long-term hourly values and for the purpose of this study, wedefined an icewine event as one in which there were two consecutive days with minimumtemperatures ≤ -8o C. To determine any evidence of long-term trend and variability in thedata, we analyzed the time series using simple linear regression. We analyzed the totalnumber of events for individual months and the combined total for each winter season.

RESULTS AND DISCUSSIONOptimal Harvesting Period

November 15 marks the start of the official harvesting date of grapes for icewines.However, the record over the last forty years showed a total of twelve days experiencedminimum temperatures ≤ -8o C. Only one year (1989) had four consecutive days with ≤ -8o

C that were suitable for harvesting frozen grapes. An early harvest date means higher yields,since loss to predators, a longer hanging time and spoilage due to warmer temperatures canreduce yields and quality. Fig. 2 and 3 show the most suitable period for harvesting frozengrapes. It begins roughly from the beginning of the second week in January and extends toapproximately the middle of February. This period has a slightly >50% chance of theoccurrence of freezing temperatures ≤ -8o C.

5

Figure 2. Average daily minimum temperatures for December to February for theperiod 1970-2007

Figure 3. Probability of daily minimum temperatures ≤ -8o C for the period 1970-2007

Trends in Freezing TemperaturesSome winegrowers will harvest at -8o C, while others may await lower temperatures over

a longer. An analysis of the frequency of days (Fig. 4) with freezing temperatures ≤ -8o Cfor the December to February period shows a high degree of inter-annual variability with aweak declining trend for the 1970-2007 period. Values range from a high of 50 in 1978 to alow of 11 in 2002 duration.

3 - 13



6

Figure 4. Frequency of days with ≤ -8oC between December and February for the1970-2007 period

Trends in Icewine EventsWe defined an icewine event as one with at least two consecutive days with ≤ -8o C.

Trend analysis in Fig. 5 shows high inter-annual variability along with a declining trend forthe three coldest months. The sharpest decline is observed in February followed by Januaryand December, respectively.

Figure 5. Temporal distribution of icewine events (picking days) for the months ofDecember, January and February for the 1970-2007 period

7

MANAGING THE RISKS OF GLOBAL WARMING

In addition to the declining trend observed in the number of picking days, Fig. 5 alsoappears to provide some indication of increased inter-year volatility, particularly in themonths of January and February, from 1990 onward. This is consistent with the results ofCyr and Kusy (2007) who, in a study of estimated annual hours of icewine harvesting time,find some evidence of increased volatility from the early 1990’s onward.

Although adaptation strategies in terms of viticulture, to mitigate crop loss due to theimpact of increasingly warm and wet conditions may be developed, the increased volatilitycan create uncertainty in terms of the year-to-year expenditures associated with suchmethods. Cyr and Kusy (2007) considered the potential use of weather contracts for hedgingsuch economic risks, particularly with respect to icewine harvesting. Although weatherderivative contracts first began trading in the mid 1990s the availability of such contractsfor hedging specialized weather risks has only developed substantially in recent years. Thisgrowth is partly due to the increased awareness of the risks resulting from global warmingand the potential role of weather contracts in mitigating some of them (Chicago MercantileExchange, 2009). Many issues still remain problematic in terms of the use of weathercontracts however, including the identification of appropriate statistically models forestimating future weather variability as well as other issues critical to the pricing of suchcontracts. In addition the practical application of such contracts requires an assessment ofthe correlation of specific weather events faced by a producer to those associated with anearby weather station employed as the basis of the contracts.

CONCLUSIONS

Like many agricultural sectors, the viticulture industry is highly sensitive to the weather.Major risks at the production level are attributed to occurrences of extreme events andrandom variability in key weather variables. The study analyzes the occurrences oftemperatures ≤ -8o C in the months of November, December, January and February inwhich the frozen grapes are normally picked. The results of trend analysis showed a highdegree of inter-annual variability along with a weak declining trend over the last forty yearsin the number of days suitable for harvesting the frozen grape. Two major risks to icewinegrapes are firstly, prolonged warm and wet conditions that could lead to rot and secondly,the destruction of the crop by bird predators. In the short-term, producers can hedge theirrisks by buying weather contracts, while modelling long-term changes in the regionalclimate could help to determine appropriate adaptive strategies.

References

BINDI, M., FIBBI, L., GOZZINI, B., ORLANDINI, S., AND MIGLIETTA, F. Modellingthe impact of future climate change scenarios on yield and yield variability of grapevine,Climate Research, 7, 213-224.

BRKLACICH, M., BRYANT, C., VEENHOF, B AND BEAUCHESNE, A. (1998):Implications of global climatic change for Canadian agriculture: a review and appraisal ofresearch from 1984 to 1997; in Responding to Global Climate Change: National SectoralIssue, (ed.) G. Koshida and W. Avis, Environment Canada, Canada Country Study: ClimateImpacts and Adaptation, v. VII, p. 219-256.

3 - 14



6

Figure 4. Frequency of days with ≤ -8oC between December and February for the1970-2007 period

Trends in Icewine EventsWe defined an icewine event as one with at least two consecutive days with ≤ -8o C.

Trend analysis in Fig. 5 shows high inter-annual variability along with a declining trend forthe three coldest months. The sharpest decline is observed in February followed by Januaryand December, respectively.

Figure 5. Temporal distribution of icewine events (picking days) for the months ofDecember, January and February for the 1970-2007 period

7

MANAGING THE RISKS OF GLOBAL WARMING

In addition to the declining trend observed in the number of picking days, Fig. 5 alsoappears to provide some indication of increased inter-year volatility, particularly in themonths of January and February, from 1990 onward. This is consistent with the results ofCyr and Kusy (2007) who, in a study of estimated annual hours of icewine harvesting time,find some evidence of increased volatility from the early 1990’s onward.

Although adaptation strategies in terms of viticulture, to mitigate crop loss due to theimpact of increasingly warm and wet conditions may be developed, the increased volatilitycan create uncertainty in terms of the year-to-year expenditures associated with suchmethods. Cyr and Kusy (2007) considered the potential use of weather contracts for hedgingsuch economic risks, particularly with respect to icewine harvesting. Although weatherderivative contracts first began trading in the mid 1990s the availability of such contractsfor hedging specialized weather risks has only developed substantially in recent years. Thisgrowth is partly due to the increased awareness of the risks resulting from global warmingand the potential role of weather contracts in mitigating some of them (Chicago MercantileExchange, 2009). Many issues still remain problematic in terms of the use of weathercontracts however, including the identification of appropriate statistically models forestimating future weather variability as well as other issues critical to the pricing of suchcontracts. In addition the practical application of such contracts requires an assessment ofthe correlation of specific weather events faced by a producer to those associated with anearby weather station employed as the basis of the contracts.

CONCLUSIONS

Like many agricultural sectors, the viticulture industry is highly sensitive to the weather.Major risks at the production level are attributed to occurrences of extreme events andrandom variability in key weather variables. The study analyzes the occurrences oftemperatures ≤ -8o C in the months of November, December, January and February inwhich the frozen grapes are normally picked. The results of trend analysis showed a highdegree of inter-annual variability along with a weak declining trend over the last forty yearsin the number of days suitable for harvesting the frozen grape. Two major risks to icewinegrapes are firstly, prolonged warm and wet conditions that could lead to rot and secondly,the destruction of the crop by bird predators. In the short-term, producers can hedge theirrisks by buying weather contracts, while modelling long-term changes in the regionalclimate could help to determine appropriate adaptive strategies.

References

BINDI, M., FIBBI, L., GOZZINI, B., ORLANDINI, S., AND MIGLIETTA, F. Modellingthe impact of future climate change scenarios on yield and yield variability of grapevine,Climate Research, 7, 213-224.

BRKLACICH, M., BRYANT, C., VEENHOF, B AND BEAUCHESNE, A. (1998):Implications of global climatic change for Canadian agriculture: a review and appraisal ofresearch from 1984 to 1997; in Responding to Global Climate Change: National SectoralIssue, (ed.) G. Koshida and W. Avis, Environment Canada, Canada Country Study: ClimateImpacts and Adaptation, v. VII, p. 219-256.

3 - 15



8

CHICAGO MERCANTILE EXCHANGE GROUP/STORM EXCHANGE. 2009. BestPractices for the Agricultural Community (Webniar Series). Accessed on September 19th

2009 at http://www.cmegroup.com/education/events/forms/storm_webinar_series.html.

CYR, D. AND KUSY, M. 2007. Canadian Icewine production: A case for the use ofweather derivatives, Journal of Wine Economics, 2, 1-23

DUCHENE, E and SCHNEIDER, C. 2005. Grapevine and climatic changes: a glance at thesituation on Alsace, Agronomy and Sustainable Development, 25, 93-99.

EASTERLING D.R. et. al. 2000. Observed variability and trends in extreme climateevents: A brief review, Bulletin of American Meteorological Society, 81, 417-425.

INTERNATIONAL PANEL ON CLIMATE CHANGE (IPCC) 2007. IPCC FourthAssessment Report. http://www.ipcc.chHAEBERLI, W., FRAUENFELDER, R. , KAAB, A and WAGNER, S .2004.Characteristics and potential climatic significance of miniature ice caps, Journal ofGlaciology, 50 (168) 129-136.

JONES, G.V. and DAVIS, R.E. 2000. Climate influences on grapevine phenology , grapecomposition, and wine quality for Bordeaux, France, American Journal of Enology andViticulture, 51, 249-261

JONES, G.V. and GOODRICH, G.B. 2008. Influence of climate variabilityon wine regionsin the Western United States and on wine quality in the Napa Valley, Climate Research, 35,241-254.KENNY, G.J. and HARRISON, P.A. 1992. The effects of climate variability and change ongrape suitability in Europe, Journal of Wine Research, 3, 163-183.

KRAMER, K. 1994. A modelling analysis of the effects of climatic warming on theprobability of spring frost damage to tree species in the Netherlands and Germany, Plant ,Cell and Environment, 17, 367-377.

NATURAL RESOURCES CANADA. 2002. Climate Change Impacts and Adaptation: ACanadian Perspective : Impacts on Agriculture, Ottawa, Ontario.

STOCK, M., GERSTENGARBE, F., KARTSCHALL, T. and WERNER, P. 2004.Reliability of Climate Change Impact Assessments for ViticultureInternational Scoiety forHorticulture Science, 68 (9).

ZHANG, X., VINCENT, L.A, HOGG, W.D. NIITSOO, A. 2002. Temperature andprecipitation trends in Canada during the 20th century, Atmosphere-Ocean, 38, 395-429.

VINTNERS QUALITY ALLIANCE ONTARIO: www.vqaontario.com

ZIRALDO, D and KAISER, K. 2007. Ice Wine: Extreme Wine Making, Key Porter BooksLimited, Toronto, Ontario.

L’EFFET DU CLIMAT VITICOLE SUR LA TYPICITÉ DES VINS ROUGES

Caractérisation au Niveau des Régions Viticoles Ibéro-Americaines

J. Tonietto(1), V. Sotés(2), M.C. Zanus(1), C. Montes(3), E.M. Uliarte(4), L. Antelo(5), P. Clímaco(6)

A. Peña(7), C.C. Guerra(1), C.D. Catania(4), E. Kohlberg(8), G. E. Pereira(1), J.R. da Silva(9), J.V. Ragoût(10), L.V. Navarro(10), O. Laureano(9), R. de Castro(9), R.F. del Monte(4), S.A. del Monte(4), V.D. Gómez-Miguel(2),

A.Carbonneau(11).(1)EMBRAPA Uva e Vinho, Rua Livramento, 515 - 95700-000 - Bento Gonçalves, Brésil,

[email protected] ; (2) UPM - Universidad Politécnica de Madri, Espagne ; (3) CEAZA - Centro de Estudios Avanzados en Zonas Áridas, Chili ; (4)INTA - EEA Mendoza, Argentine ; (5) PFCUVS-FAUTAPO, Desarrollo de Mercados, Bolivie ; (6)INIA/INRB, Estação Vitivinícola Nacional, Portugal ; (7)Universidad de Chile ; (8) Expert Oenologue, Bolivie ; (9)ISA-UTL - Instituto Superior de Agronomia, Portugal ; (10) Expert

Oenologue, Espagne ; (11) AGRO Montpellier, France.

RÉSUMÉIl n’existe presque pas d’études qui caractérisent l’effet du climat viticole sur la typicité des

vins en considérant les différents types de climats à l’échelle mondiale. Cette étude fait partie d’un projet CYTED de zonage vitivinicole. L’objectif a été de caractériser l’effet du climat viticole sur la typicité des vins sur une macro région viticole du monde. La méthodologie a été appliquée à un ensemble de 45 régions viticoles situées sur 6 pays Ibéro-Américains : Argentine, Bolivie, Brésil, Chili, Espagne et Portugal. Le climat viticole de chaque région viticole a été caractérisé para les 3 indices climatiques viticoles du Système CCM Géoviticole : IH (Indice Héliothermique de Huglin), (IF) Indice de Fraîcheur des Nuits) et IS (Indice de Sécheresse). Les principales caractéristiques sensorielles observées de façon fréquente sur des vins rouges représentatifs élaborés avec des raisins-de-cuve de chacune des ces 45 régions viticoles ont été décrites pour des œnologues de chaque pays, an utilisant la méthodologie proposée par Zanus & Tonietto (2007). L’évaluation sensorielle réalisée concerne l’intensité de perception de la Couleur (Cou), de l’Arôme Total (Ar), de l’Arôme – fruit mûr (Ar-Fm), de la Concentration (Con), de l’Alcool (Al), des Tanins (Tan), de l’Acidité (Ac) et la Longueur en bouche (Lon). Les données ont été soumises à l’analyse des corrélations pour l’ensemble des variables et à l’ACP. L’étude indique qu’une partie de la typicité des vins est déterminée par le climat viticole des régions et que les indices du Système CCM Géoviticole sont pertinents pour relier aux caractéristiques sensorielles des vins. Le déterminisme de l’IH, de l’IS et de l’IF à été mis en évidence.

MOTS CLÉS : climat viticole, indice climatique, Système CCM, vin, typicité.

ABSTRACTTHE EFFECT OF VITICULTURAL CLIMATE ON RED WINE TYPICITY

A Characterization on Iberoamerican Grape-Growing Wine Regions

There are many studies in the world that characterize the effect of the climate on grape composition and wine typicity concerning particular viticultural regions and climates. However, there are not studies, in a worldwide scale, that characterize this effect considering different climate types. This study is part of a CYTED project in vitivinicultural zoning. The objective was to characterize the effect of viticultural climate on the wine typicity on a macro viticultural region of the world. The methodology employed in this investigation used 45 grape-growing regions in 6 Iberoamerican countries: Argentina, Bolivia, Brazil, Chile, Portugal and Spain. The viticultural climate of each region was characterized by the 3

13 - 16



8

CHICAGO MERCANTILE EXCHANGE GROUP/STORM EXCHANGE. 2009. BestPractices for the Agricultural Community (Webniar Series). Accessed on September 19th

2009 at http://www.cmegroup.com/education/events/forms/storm_webinar_series.html.

CYR, D. AND KUSY, M. 2007. Canadian Icewine production: A case for the use ofweather derivatives, Journal of Wine Economics, 2, 1-23

DUCHENE, E and SCHNEIDER, C. 2005. Grapevine and climatic changes: a glance at thesituation on Alsace, Agronomy and Sustainable Development, 25, 93-99.

EASTERLING D.R. et. al. 2000. Observed variability and trends in extreme climateevents: A brief review, Bulletin of American Meteorological Society, 81, 417-425.

INTERNATIONAL PANEL ON CLIMATE CHANGE (IPCC) 2007. IPCC FourthAssessment Report. http://www.ipcc.chHAEBERLI, W., FRAUENFELDER, R. , KAAB, A and WAGNER, S .2004.Characteristics and potential climatic significance of miniature ice caps, Journal ofGlaciology, 50 (168) 129-136.

JONES, G.V. and DAVIS, R.E. 2000. Climate influences on grapevine phenology , grapecomposition, and wine quality for Bordeaux, France, American Journal of Enology andViticulture, 51, 249-261

JONES, G.V. and GOODRICH, G.B. 2008. Influence of climate variabilityon wine regionsin the Western United States and on wine quality in the Napa Valley, Climate Research, 35,241-254.KENNY, G.J. and HARRISON, P.A. 1992. The effects of climate variability and change ongrape suitability in Europe, Journal of Wine Research, 3, 163-183.

KRAMER, K. 1994. A modelling analysis of the effects of climatic warming on theprobability of spring frost damage to tree species in the Netherlands and Germany, Plant ,Cell and Environment, 17, 367-377.

NATURAL RESOURCES CANADA. 2002. Climate Change Impacts and Adaptation: ACanadian Perspective : Impacts on Agriculture, Ottawa, Ontario.

STOCK, M., GERSTENGARBE, F., KARTSCHALL, T. and WERNER, P. 2004.Reliability of Climate Change Impact Assessments for ViticultureInternational Scoiety forHorticulture Science, 68 (9).

ZHANG, X., VINCENT, L.A, HOGG, W.D. NIITSOO, A. 2002. Temperature andprecipitation trends in Canada during the 20th century, Atmosphere-Ocean, 38, 395-429.

VINTNERS QUALITY ALLIANCE ONTARIO: www.vqaontario.com

ZIRALDO, D and KAISER, K. 2007. Ice Wine: Extreme Wine Making, Key Porter BooksLimited, Toronto, Ontario.

L’EFFET DU CLIMAT VITICOLE SUR LA TYPICITÉ DES VINSROUGES

Caractérisation au Niveau des Régions Viticoles Ibéro-Americaines

J. Tonietto(1), V. Sotés(2), M.C. Zanus(1), C. Montes(3), E.M. Uliarte(4), L. Antelo(5), P. Clímaco(6)

A. Peña(7), C.C. Guerra(1), C.D. Catania(4), E. Kohlberg(8), G. E. Pereira(1), J.R. da Silva(9), J.V. Ragoût(10),L.V. Navarro(10), O. Laureano(9), R. de Castro(9), R.F. del Monte(4), S.A. del Monte(4), V.D. Gómez-Miguel(2),

A.Carbonneau(11).(1)EMBRAPA Uva e Vinho, Rua Livramento, 515 - 95700-000 - Bento Gonçalves, Brésil,

[email protected] ; (2) UPM - Universidad Politécnica de Madri, Espagne ; (3) CEAZA - Centro deEstudios Avanzados en Zonas Áridas, Chili ; (4)INTA - EEA Mendoza, Argentine ; (5) PFCUVS-FAUTAPO, Desarrollo de Mercados, Bolivie ; (6)INIA/INRB, Estação Vitivinícola Nacional, Portugal ; (7)Universidad de Chile ; (8) Expert Oenologue, Bolivie ; (9)ISA-UTL - Instituto Superior de Agronomia, Portugal ; (10) Expert

Oenologue, Espagne ; (11) AGRO Montpellier, France.

RÉSUMÉIl n’existe presque pas d’études qui caractérisent l’effet du climat viticole sur la typicité des

vins en considérant les différents types de climats à l’échelle mondiale. Cette étude fait partie d’un projet CYTED de zonage vitivinicole. L’objectif a été de caractériser l’effet du climat viticole sur la typicité des vins sur une macro région viticole du monde. La méthodologie a été appliquée à un ensemble de 45 régions viticoles situées sur 6 pays Ibéro-Américains : Argentine, Bolivie, Brésil, Chili, Espagne et Portugal. Le climat viticole de chaque région viticole a été caractérisé para les 3 indices climatiques viticoles du Système CCM Géoviticole : IH (Indice Héliothermique de Huglin), (IF) Indice de Fraîcheur des Nuits) et IS (Indice de Sécheresse). Les principales caractéristiques sensorielles observées de façon fréquente sur des vins rouges représentatifs élaborés avec des raisins-de-cuve de chacune des ces 45 régions viticoles ont été décrites pour des œnologues de chaque pays, an utilisant la méthodologie proposée par Zanus & Tonietto (2007). L’évaluation sensorielle réalisée concerne l’intensité de perception de la Couleur (Cou), de l’Arôme Total (Ar), de l’Arôme – fruit mûr (Ar-Fm), de la Concentration (Con), de l’Alcool (Al), des Tanins (Tan), de l’Acidité (Ac) et la Longueur en bouche (Lon). Les données ont été soumises à l’analyse des corrélations pour l’ensemble des variables et à l’ACP. L’étude indique qu’une partie de la typicité des vins est déterminée par le climat viticole des régions et que les indices du Système CCM Géoviticole sont pertinents pour relier aux caractéristiques sensorielles des vins. Le déterminisme de l’IH, de l’IS et de l’IF à été mis en évidence.

MOTS CLÉS : climat viticole, indice climatique, Système CCM, vin, typicité.

ABSTRACTTHE EFFECT OF VITICULTURAL CLIMATE ON RED WINE TYPICITY

A Characterization on Iberoamerican Grape-Growing Wine Regions

There are many studies in the world that characterize the effect of the climate on grape composition and wine typicity concerning particular viticultural regions and climates. However, there are not studies, in a worldwide scale, that characterize this effect considering different climate types. This study is part of a CYTED project in vitivinicultural zoning. The objective was to characterize the effect of viticultural climate on the wine typicity on a macro viticultural region of the world. The methodology employed in this investigation used 45 grape-growing regions in 6 Iberoamerican countries: Argentina, Bolivia, Brazil, Chile, Portugal and Spain. The viticultural climate of each region was characterized by the 3

13 - 17



viticultural climate index of the Géoviticulture MCC System (Tonietto & Carbonneau, 2004): HI (Heliothermal index), CI (Cool night index) and DI (Dryness index). The main sensory characteristics observed frequently in representative red wines produced with grapes of each of these 45 grape-growing regions were described by enologists in the respective countries, using the methodology of Zanus & Tonietto (2008). The sensory evaluation concerned to the intensity of perception of Color (Cou), Total Aroma (Ar), Aroma - ripe fruit (Ar-Rf), Body – palate concentration (Con), Alcohol (Al), Tannins (Tan) and Acidity (Ac). The Persistence in mouth (Lon) was also evaluated. The data were submitted to a correlation matrix for the variables and to a Principal Component Analysis (PCA). The results showed significant correlation effect for: HI – positive with Al and negative with Ac; DI – positive with Ac and negative with Al and Ar-Rf; CI – negative with Cou, Tan, Lon, Ar and Con. The results confirm the effect of the temperatures on increasing alcohol and reducing acidity perception of red wines. The soil water availability shows that higher values of DI contributes to rise the acidity perception and to diminish alcohol and aroma (ripe fruit) perception. The effect of nycto-temperatures during ripening was confirmed influencing several sensory characteristics of the wines: the cooler the night temperatures during maturation (lower CI values) the higher is the perception of color, aroma, palate concentration, tannins and the persistence in mouth. Part of the wine typicity of the regions was determined by the viticultural climate. Others are related with varieties, viticultural and wine making processes, among others in each region.

KEYWORDS : viticultural climate, climatic index, MCC System, wine, typicity.

INTRODUCTIONIl existe plusieurs études dans le monde qui caractérisent l’effet du climat sur la composition

physique et chimique du raisin-de-cuve et sur la typicité des vins dans des régions et climats viticoles particuliers. Mais il n’existent pas d’études à l’échelle mondiale qui caractérisent cet effet en considérant les différents types de climats mondiaux. Cette étude fait partie d’un projet CYTED - Programme Ibéro-Américain de Science et Technologie pour le Développement, de zonage vitivinicole (Cyted, 2003 ; Sotés & Tonietto, 2004).

L’objectif a été de caractériser l’effet du climat viticole sur la typicité des vins sur la macro région viticole Ibéro-Américaine.

MATERIEL ET MÉTHODELa méthodologie a été appliquée à un ensemble de 45 des principaux régions viticoles

situées sur 6 pays Ibéro-Américains : Argentine (Catania et al., 2007), Bolivie, Brésil, Chili, Espagne et Portugal. Le climat viticole de chaque région viticole a été caractérisé par les 3 indices climatiques viticoles du Système CCM Géoviticole (Tonietto, 1999 ; Tonietto & Carbonneau, 2004) : IH (Indice Héliothermique de Huglin), IF (Indice de Fraîcheur des Nuits) et IS (Indice de Sécheresse). Les indices ont été calculés en utilisant les moyennes climatiques interannuelles d’un poste météorologique représentatif du climat viticole de chaque région.

Les caractéristiques sensorielles moyennes observées de façon fréquente sur les principaux vins rouges secs (jusqu’à l’âge de 12 mois après fermentation alcoolique) élaborés avec le (s) cépage (s) le plus représentatif (s) de chacune des 45 régions viticoles ont été décrites, basée sur les connaissances empiriques, par des œnologues experts en évaluation sensorielle de chaque pays, an utilisant la méthodologie proposée par Zanus & Tonietto (2008). La caractérisation sensorielle réalisée concerne l’intensité de la perception des descripteurs suivants des vins, qui sont très influencés par le climat viticole : Couleur (Cou), Arôme Total (Ar), Arôme – fruit mûr (Ar-Fm), Concentration (Con), Alcool (Al), Tanins (Tan) et Acidité (Ac). La Longueur en Bouche (Lon) a été également évaluée. Les experts ont utilisé un

2

formulaire de caractérisation sensorielle (Tableau 1), avec une échelle de perception sensorielle de l’intensité, qui varie de l’intensité baisse (1) à l’intensité haute (5), classé selon la variabilité d’intensité observée sur les vins à l’échelle mondiale.

Tableau 1. Formulaire de caractérisation sensorielle des vins rouges des régions viticoles.

Les données ont été soumises à l’analyse des corrélations pour l’ensemble des variables et à l’Analyse en Composantes Principales (ACP).

RÉSULTATS ET DISCUSSION

Le Tableau 2 montre les moyennes et l’écart-type des indices climatiques viticoles du Système CCM et des variables sensorielles des 45 régions viticoles. Le IH a présenté une valeur moyenne de 2.398, avec la valeur minimale de 1.700 et la valeur maximale de 3.294 ; le IF a présenté une valeur moyenne de 13,3°C, avec une valeur minimale de 8,1°C et une valeur maximale de 21,0°C ; et le IS a présenté une valeur moyenne de –71 mm, avec une valeur minimale de –276 mm et une valeur maximale de 200 mm, excepte pour les climats très frais. On observe une très bonne représentation de la variabilité observée au niveau de la viticulture mondiale. Les valeurs moyennes sur l’ensemble des variables sensorielles se situent entre 3,0 (Ac) et 3,7 (Al). L’écart-type sur l’ensemble des variables sensorielles se situe entre 0,67 (Al) et 0,81 (Ar-Fm et Ac).

Tableau 2. Moyenne et l’écart-type des indices climatiques du Système CCM et variables sensorielles pour l’ensemble des 45 régions viticoles de l’étude.

Le Tableau 3 présente les coefficients de corrélation des indices climatiques du Système CCM et variables sensorielles pour l’ensemble des 45 régions viticoles de l’étude, avec le niveau de significance statistique.

3

IH IF IS Cou Ar Ar-Fm Conc Al Tan Ac Lon

Moyenne 2398 13,3 -71 3,7 3,6 3,6 3,6 3,7 3,4 3,0 3,6

Ecart-type 363,69 2,87 114,73 0,88 0,72 0,81 0,75 0,67 0,72 0,81 0,75

Baisse Haute

Couleur - intensitéArôme - intensitéArôme - fruit mûr - intensitéConcentration - intensitéAlcool - intensitéTanins - intensitéAcidité - intensitéLongueur en bouche

Descripteur sensorielTendance de l'intensité

3 - 18



viticultural climate index of the Géoviticulture MCC System (Tonietto & Carbonneau, 2004): HI (Heliothermal index), CI (Cool night index) and DI (Dryness index). The main sensory characteristics observed frequently in representative red wines produced with grapes of each of these 45 grape-growing regions were described by enologists in the respective countries, using the methodology of Zanus & Tonietto (2008). The sensory evaluation concerned to the intensity of perception of Color (Cou), Total Aroma (Ar), Aroma - ripe fruit (Ar-Rf), Body – palate concentration (Con), Alcohol (Al), Tannins (Tan) and Acidity (Ac). The Persistence in mouth (Lon) was also evaluated. The data were submitted to a correlation matrix for the variables and to a Principal Component Analysis (PCA). The results showed significant correlation effect for: HI – positive with Al and negative with Ac; DI – positive with Ac and negative with Al and Ar-Rf; CI – negative with Cou, Tan, Lon, Ar and Con. The results confirm the effect of the temperatures on increasing alcohol and reducing acidity perception of red wines. The soil water availability shows that higher values of DI contributes to rise the acidity perception and to diminish alcohol and aroma (ripe fruit) perception. The effect of nycto-temperatures during ripening was confirmed influencing several sensory characteristics of the wines: the cooler the night temperatures during maturation (lower CI values) the higher is the perception of color, aroma, palate concentration, tannins and the persistence in mouth. Part of the wine typicity of the regions was determined by the viticultural climate. Others are related with varieties, viticultural and wine making processes, among others in each region.

KEYWORDS : viticultural climate, climatic index, MCC System, wine, typicity.

INTRODUCTIONIl existe plusieurs études dans le monde qui caractérisent l’effet du climat sur la composition

physique et chimique du raisin-de-cuve et sur la typicité des vins dans des régions et climats viticoles particuliers. Mais il n’existent pas d’études à l’échelle mondiale qui caractérisent cet effet en considérant les différents types de climats mondiaux. Cette étude fait partie d’un projet CYTED - Programme Ibéro-Américain de Science et Technologie pour le Développement, de zonage vitivinicole (Cyted, 2003 ; Sotés & Tonietto, 2004).

L’objectif a été de caractériser l’effet du climat viticole sur la typicité des vins sur la macro région viticole Ibéro-Américaine.

MATERIEL ET MÉTHODELa méthodologie a été appliquée à un ensemble de 45 des principaux régions viticoles

situées sur 6 pays Ibéro-Américains : Argentine (Catania et al., 2007), Bolivie, Brésil, Chili, Espagne et Portugal. Le climat viticole de chaque région viticole a été caractérisé par les 3 indices climatiques viticoles du Système CCM Géoviticole (Tonietto, 1999 ; Tonietto & Carbonneau, 2004) : IH (Indice Héliothermique de Huglin), IF (Indice de Fraîcheur des Nuits) et IS (Indice de Sécheresse). Les indices ont été calculés en utilisant les moyennes climatiques interannuelles d’un poste météorologique représentatif du climat viticole de chaque région.

Les caractéristiques sensorielles moyennes observées de façon fréquente sur les principaux vins rouges secs (jusqu’à l’âge de 12 mois après fermentation alcoolique) élaborés avec le (s) cépage (s) le plus représentatif (s) de chacune des 45 régions viticoles ont été décrites, basée sur les connaissances empiriques, par des œnologues experts en évaluation sensorielle de chaque pays, an utilisant la méthodologie proposée par Zanus & Tonietto (2008). La caractérisation sensorielle réalisée concerne l’intensité de la perception des descripteurs suivants des vins, qui sont très influencés par le climat viticole : Couleur (Cou), Arôme Total (Ar), Arôme – fruit mûr (Ar-Fm), Concentration (Con), Alcool (Al), Tanins (Tan) et Acidité (Ac). La Longueur en Bouche (Lon) a été également évaluée. Les experts ont utilisé un

2

formulaire de caractérisation sensorielle (Tableau 1), avec une échelle de perception sensorielle de l’intensité, qui varie de l’intensité baisse (1) à l’intensité haute (5), classé selon la variabilité d’intensité observée sur les vins à l’échelle mondiale.

Tableau 1. Formulaire de caractérisation sensorielle des vins rouges des régions viticoles.

Les données ont été soumises à l’analyse des corrélations pour l’ensemble des variables et à l’Analyse en Composantes Principales (ACP).

RÉSULTATS ET DISCUSSION

Le Tableau 2 montre les moyennes et l’écart-type des indices climatiques viticoles du Système CCM et des variables sensorielles des 45 régions viticoles. Le IH a présenté une valeur moyenne de 2.398, avec la valeur minimale de 1.700 et la valeur maximale de 3.294 ; le IF a présenté une valeur moyenne de 13,3°C, avec une valeur minimale de 8,1°C et une valeur maximale de 21,0°C ; et le IS a présenté une valeur moyenne de –71 mm, avec une valeur minimale de –276 mm et une valeur maximale de 200 mm, excepte pour les climats très frais. On observe une très bonne représentation de la variabilité observée au niveau de la viticulture mondiale. Les valeurs moyennes sur l’ensemble des variables sensorielles se situent entre 3,0 (Ac) et 3,7 (Al). L’écart-type sur l’ensemble des variables sensorielles se situe entre 0,67 (Al) et 0,81 (Ar-Fm et Ac).

Tableau 2. Moyenne et l’écart-type des indices climatiques du Système CCM et variables sensorielles pour l’ensemble des 45 régions viticoles de l’étude.

Le Tableau 3 présente les coefficients de corrélation des indices climatiques du Système CCM et variables sensorielles pour l’ensemble des 45 régions viticoles de l’étude, avec le niveau de significance statistique.

3

IH IF IS Cou Ar Ar-Fm Conc Al Tan Ac Lon

Moyenne 2398 13,3 -71 3,7 3,6 3,6 3,6 3,7 3,4 3,0 3,6

Ecart-type 363,69 2,87 114,73 0,88 0,72 0,81 0,75 0,67 0,72 0,81 0,75

Baisse Haute

Couleur - intensitéArôme - intensitéArôme - fruit mûr - intensitéConcentration - intensitéAlcool - intensitéTanins - intensitéAcidité - intensitéLongueur en bouche

Descripteur sensorielTendance de l'intensité

3 - 19



Tableau 3. Coefficients de corrélation des indices climatiques du Système CCM et variables sensorielles pour l’ensemble des 45 régions viticoles de l’étude.

* Significatif au niveau de 5% de probabilité. ** Significatif au niveau de 1% de probabilité.

Les résultats montrent une corrélation significative entre indices climatiques viticoles et les variables sensorielles pour : IH – positive avec Al et négative avec Ac ; IS – positive avec Ac et négative avec Al et Ar-Fm ; IF – négative avec Cou, Ar, Con, Tan et Lon.

La Figure 1 présente le cercle des corrélations de l’Analyse en Composantes Principales (ACP) des indices climatiques du Système CCM et variables sensorielles pour l’ensemble des 45 régions viticoles de l’étude. Les composantes principales 1 et 2 expliquent 64,67% de la variabilité. L’ACP renforce les résultats du Tableau 3.

Figure 1. Cercle des corrélations de l’Analyse en Composantes Principales (ACP) des indices climatiques du Système CCM et variables sensorielles pour l’ensemble des 45 régions viticoles de l’étude.

4

IH

IF

IS

Cou

Ar Ar-Fm

Con

Al

Tan

Ac

Lon

CP 1 : 38,46% d’inertie

-1,0

0,0

1,0

-1,0 0,0 1,0

Positionnement des variables sur le cercle des corrélations

CP

2 :

26,

21%

d’i

nert

ie

Variable

IH 1,00 - - - - - - - - - -

IF 0,53 ** 1,00 - - - - - - - - -

IS -0,34 * 0,12 1,00 - - - - - - - -

Cou -0,23 -0,49 ** -0,09 1,00 - - - - - - -

Ar 0,15 -0,33 * -0,21 0,43 ** 1,00 - - - - - -

Ar-Fm 0,17 -0,27 -0,46 ** 0,46 ** 0,66 ** 1,00 - - - - -

Con -0,11 -0,34 * -0,03 0,74 ** 0,52 ** 0,50 ** 1,00 - - - -

Al 0,49 ** 0,11 -0,55 ** 0,13 0,26 0,37 * 0,23 1,00 - - -

Tan -0,18 -0,39 ** -0,01 0,75 ** 0,26 0,26 0,72 ** 0,01 1,00 - -

Ac -0,59 ** -0,26 0,44 ** 0,40 ** -0,21 -0,02 0,36 * -0,47 ** 0,48 ** 1,00 -

Lon -0,09 -0,41 ** -0,23 0,59 ** 0,71 ** 0,63 ** 0,62 ** 0,20 0,37 * 0,06 1,00

IH IF IS Cou Tan Ac LonAr Ar-Fm Con Al

Les résultats confirment l’effet des températures (IH) sur l’augmentation, surtout de la perception l’alcool et sur la réduction de la perception de l’acidité des vins rouges. La réserve en eau du sol montre que les valeurs les plus élevées de IS contribuent, surtout, à augmenter la perception de l’acidité et à réduire la perception de l’alcool et de l’intensité de l’arôme (fruit mûr). L’effet des nycto températures en période de maturation du raisin sur plusieurs caractéristiques sensorielles des vins a été mis en évidence : les nuits fraîches en période de maturation (les valeurs les plus baisses de IF), favorisent la perception de la couleur, des tannins, de l’arôme, de la concentration et de la longueur en bouche.

Evidement que la caractérisation sensorielle de chaque région n’est pas seulement l’expression de l’effet climatique. Bien au contraire, elle intègre également la grande variabilité associé aux différents cépages et ses interactions avec le milieu physique, aux systèmes viticoles et à l’ensemble des pratiques œnologiques adoptées par chaque région.

De toute façon, l’utilisation des résultats obtenues et d’autres dans l’avenir en reliant l’effet du climat sur la typitité des vins peut servir aussi pour avoir une idée de la typicité espéré pour des vins à produire dans des nouvelles régions potentielles pour la viticulture et pour avoir une idée quantifiée du changement de typicité des vins des régions productrices en fonction du changement climatique.

CONCLUSIONS

L’étude indique qu’une partie de la typicité des vins est déterminée par le climat viticole des régions et que les indices du Système CCM Géoviticole sont pertinents pour les relier aux caractéristiques sensorielles des vins. L’effet de l’Indice Héliothermique et de l’Indice de Sécheresse à été confirmé sur les variables sensorielles, surtout sur l’alcool, sur l’acidité et sur l’intensité de l’arôme. Le déterminisme de l’Indice de Fraîcheur des Nuits sur la perception sensorielle des vins – couleur, arôme, tannins, persistance, à été mis en évidence.

REMERCIEMENTS

On voudrait remercier tout d’abord au CYTED pour avoir possibilité le développement du projet qui est à l’origine de ce travail et a toutes les institutions de recherche et développement des pays impliqués. À la FINEP – Financiadora de Estudos e Projetos, pour l’appuie à la consécution du travail au Brésil. Egalement, aux diverses institutions qu’ont fourni les bases des données climatiques des régions viticoles de l’étude et aux œnologues experts de tous les pays pour l’évaluation sensorielle des vins des régions viticoles.

BIBLIOGRAPHY

Catania, C.D.; Avagnina de del Monte, S.; Uliarte, E. M.; F. del Monte, R.; Tonietto, J. 2007. El clima vitícola de las regiones productoras de uvas para vinos de Argentina. In: Tonietto, J.; Sotés, V. (Ed.). Caracterização climática de regiões vitivinícolas ibero-americanas. Bento Gonçalves: Embrapa Uva e Vinho. p.9-55. Disponible à : <http://www.cnpuv.embrapa.br/ccm>.

Cyted. 2003. Metodologías de zonificación y su aplicación a las regiones vitivinícolas Iberoamericanas. Madrid. 20p. (Proyecto de Investigación Cooperativa; Coodinacion de Vicente Sotés Ruiz - UPM, España).

Sotés, V.; Tonietto, J. 2004. Climatic zoning of the Ibero-American viticultural regions. In: Joint International Conference on Viticultural Zoning, 2004, Cape Town. Proceedings. Cape

53 - 20



Tableau 3. Coefficients de corrélation des indices climatiques du Système CCM et variables sensorielles pour l’ensemble des 45 régions viticoles de l’étude.

* Significatif au niveau de 5% de probabilité. ** Significatif au niveau de 1% de probabilité.

Les résultats montrent une corrélation significative entre indices climatiques viticoles et les variables sensorielles pour : IH – positive avec Al et négative avec Ac ; IS – positive avec Ac et négative avec Al et Ar-Fm ; IF – négative avec Cou, Ar, Con, Tan et Lon.

La Figure 1 présente le cercle des corrélations de l’Analyse en Composantes Principales (ACP) des indices climatiques du Système CCM et variables sensorielles pour l’ensemble des 45 régions viticoles de l’étude. Les composantes principales 1 et 2 expliquent 64,67% de la variabilité. L’ACP renforce les résultats du Tableau 3.

Figure 1. Cercle des corrélations de l’Analyse en Composantes Principales (ACP) des indices climatiques du Système CCM et variables sensorielles pour l’ensemble des 45 régions viticoles de l’étude.

4

IH

IF

IS

Cou

Ar Ar-Fm

Con

Al

Tan

Ac

Lon

CP 1 : 38,46% d’inertie

-1,0

0,0

1,0

-1,0 0,0 1,0

Positionnement des variables sur le cercle des corrélations

CP

2 :

26,

21%

d’i

nert

ie

Variable

IH 1,00 - - - - - - - - - -

IF 0,53 ** 1,00 - - - - - - - - -

IS -0,34 * 0,12 1,00 - - - - - - - -

Cou -0,23 -0,49 ** -0,09 1,00 - - - - - - -

Ar 0,15 -0,33 * -0,21 0,43 ** 1,00 - - - - - -

Ar-Fm 0,17 -0,27 -0,46 ** 0,46 ** 0,66 ** 1,00 - - - - -

Con -0,11 -0,34 * -0,03 0,74 ** 0,52 ** 0,50 ** 1,00 - - - -

Al 0,49 ** 0,11 -0,55 ** 0,13 0,26 0,37 * 0,23 1,00 - - -

Tan -0,18 -0,39 ** -0,01 0,75 ** 0,26 0,26 0,72 ** 0,01 1,00 - -

Ac -0,59 ** -0,26 0,44 ** 0,40 ** -0,21 -0,02 0,36 * -0,47 ** 0,48 ** 1,00 -

Lon -0,09 -0,41 ** -0,23 0,59 ** 0,71 ** 0,63 ** 0,62 ** 0,20 0,37 * 0,06 1,00

IH IF IS Cou Tan Ac LonAr Ar-Fm Con Al

Les résultats confirment l’effet des températures (IH) sur l’augmentation, surtout de la perception l’alcool et sur la réduction de la perception de l’acidité des vins rouges. La réserve en eau du sol montre que les valeurs les plus élevées de IS contribuent, surtout, à augmenter la perception de l’acidité et à réduire la perception de l’alcool et de l’intensité de l’arôme (fruit mûr). L’effet des nycto températures en période de maturation du raisin sur plusieurs caractéristiques sensorielles des vins a été mis en évidence : les nuits fraîches en période de maturation (les valeurs les plus baisses de IF), favorisent la perception de la couleur, des tannins, de l’arôme, de la concentration et de la longueur en bouche.

Evidement que la caractérisation sensorielle de chaque région n’est pas seulement l’expression de l’effet climatique. Bien au contraire, elle intègre également la grande variabilité associé aux différents cépages et ses interactions avec le milieu physique, aux systèmes viticoles et à l’ensemble des pratiques œnologiques adoptées par chaque région.

De toute façon, l’utilisation des résultats obtenues et d’autres dans l’avenir en reliant l’effet du climat sur la typitité des vins peut servir aussi pour avoir une idée de la typicité espéré pour des vins à produire dans des nouvelles régions potentielles pour la viticulture et pour avoir une idée quantifiée du changement de typicité des vins des régions productrices en fonction du changement climatique.

CONCLUSIONS

L’étude indique qu’une partie de la typicité des vins est déterminée par le climat viticole des régions et que les indices du Système CCM Géoviticole sont pertinents pour les relier aux caractéristiques sensorielles des vins. L’effet de l’Indice Héliothermique et de l’Indice de Sécheresse à été confirmé sur les variables sensorielles, surtout sur l’alcool, sur l’acidité et sur l’intensité de l’arôme. Le déterminisme de l’Indice de Fraîcheur des Nuits sur la perception sensorielle des vins – couleur, arôme, tannins, persistance, à été mis en évidence.

REMERCIEMENTS

On voudrait remercier tout d’abord au CYTED pour avoir possibilité le développement du projet qui est à l’origine de ce travail et a toutes les institutions de recherche et développement des pays impliqués. À la FINEP – Financiadora de Estudos e Projetos, pour l’appuie à la consécution du travail au Brésil. Egalement, aux diverses institutions qu’ont fourni les bases des données climatiques des régions viticoles de l’étude et aux œnologues experts de tous les pays pour l’évaluation sensorielle des vins des régions viticoles.

BIBLIOGRAPHY

Catania, C.D.; Avagnina de del Monte, S.; Uliarte, E. M.; F. del Monte, R.; Tonietto, J. 2007. El clima vitícola de las regiones productoras de uvas para vinos de Argentina. In: Tonietto, J.; Sotés, V. (Ed.). Caracterização climática de regiões vitivinícolas ibero-americanas. Bento Gonçalves: Embrapa Uva e Vinho. p.9-55. Disponible à : <http://www.cnpuv.embrapa.br/ccm>.

Cyted. 2003. Metodologías de zonificación y su aplicación a las regiones vitivinícolas Iberoamericanas. Madrid. 20p. (Proyecto de Investigación Cooperativa; Coodinacion de Vicente Sotés Ruiz - UPM, España).

Sotés, V.; Tonietto, J. 2004. Climatic zoning of the Ibero-American viticultural regions. In: Joint International Conference on Viticultural Zoning, 2004, Cape Town. Proceedings. Cape

53 - 21



Town, South Africa, South African Society for Enology and Viticulture-OIV-GESCO. p. 202. CD-ROM (Viticultural Terroir Zoning 2004).

Tonietto, J. 1999. Les macroclimats viticoles mondiaux et l'influence du mésoclimat sur la typicité de la Syrah et du Muscat de Hambourg dans le sud de la France : méthodologie de caractérisation. (Thèse Doctorat). École Nationale Supérieure Agronomique de Montpellier - ENSA-M. 233p.

Tonietto, J.; Carbonneau, A. 2004. A multicriteria climatic classification system for grape-growing regions worldwide. Agricultural and Forest Meteorology, 124/1-2, 81-97.

Zanus, M. C.; Tonietto, J. 2007. Elementos metodológicos para a caracterização sensorial de vinhos de regiões climáticas vitivinícolas. In: Tonietto, J.; Sotés, V. (Ed.). Caracterização climática de regiões vitivinícolas ibero-americanas. Bento Gonçalves: Embrapa Uva e Vinho, p.57-64. Disponível em: <http://www.cnpuv.embrapa.br/ccm>.

6

(1)(1)

(1) Department of Environment and Soil Science. University of Lleida. Alcalde Rovira Roure 191, 25198, Lleida, Spain [email protected]; [email protected]

This study present a detailed analysis of the rainfall and temperature changes in the Penedès

region in the period 19959 200809, in comparison with the trends observed during the last 50 years, and its implications on phenology and yield. Temperature increases are higher than in previous time periods, which together with the irregular rainfall distribution throughout the year give rise to significant water deficits for vine development. Water deficits are being exacerbated during the last years by the increase of temperatures which imply an increase of evapotranspiration. The dates at which each phenological stage starts and the length of the different phenological stages are affected by temperature (accumulated degreedays and daily air temperature difference), precipitation and water accumulated into the soil. Winegrape yield was also influenced by soil water availability.

Evapotraspiration Mediterranean climate E Spain phenology trendsyield Climate change and its potential impacts on viticulture and viniculture have become

increasingly important as a consequence of changes in earth surface characteristics associated with the increase in greenhouse gases and changes in global temperatures, radiation budget and hydrological cycles. Vines, one of the most extensive crops in some parts of the Mediterranean Spanish area, which are cultivated under rainfed conditions, may therefore be one of the crops that suffer the consequences of climate change. According to Olesen and Bindi (2002), in the current production areas the yield variability (fruit production and quality) may be higher in the future than it is at present. However, some areas may suffer negative effects such as water stress due to a reduction in water availability and shortening of the ripening period, with harvest occurring during time with high temperatures, which may have negative impacts on wine quality (Duchêne and Schneider, 2005). Examples of impacts of temperature changes, frost occurrence and growing season lengths on grape productivity are found in the literature (Jones et al., 2005).

The Mediterranean climate is characterized by dry and warm summers and two wet seasons (spring and autumn), in which most rainfall is recorded, but there is high variability from year to year. Some authors indicate decreasing precipitation trends for the Mediterranean (Karl, 1998) and significant changes in extreme events concentrated in a small number of events such as more frequent and extreme droughts, increases in cool season precipitation, and warm season drying (Easterling et al. 2000). This study presents an analysis of temperature and precipitation and their

3 - 22



Town, South Africa, South African Society for Enology and Viticulture-OIV-GESCO. p. 202. CD-ROM (Viticultural Terroir Zoning 2004).

Tonietto, J. 1999. Les macroclimats viticoles mondiaux et l'influence du mésoclimat sur la typicité de la Syrah et du Muscat de Hambourg dans le sud de la France : méthodologie de caractérisation. (Thèse Doctorat). École Nationale Supérieure Agronomique de Montpellier - ENSA-M. 233p.

Tonietto, J.; Carbonneau, A. 2004. A multicriteria climatic classification system for grape-growing regions worldwide. Agricultural and Forest Meteorology, 124/1-2, 81-97.

Zanus, M. C.; Tonietto, J. 2007. Elementos metodológicos para a caracterização sensorial de vinhos de regiões climáticas vitivinícolas. In: Tonietto, J.; Sotés, V. (Ed.). Caracterização climática de regiões vitivinícolas ibero-americanas. Bento Gonçalves: Embrapa Uva e Vinho, p.57-64. Disponível em: <http://www.cnpuv.embrapa.br/ccm>.

6

(1)(1)

(1) Department of Environment and Soil Science. University of Lleida. Alcalde Rovira Roure 191, 25198, Lleida, Spain [email protected]; [email protected]

This study present a detailed analysis of the rainfall and temperature changes in the Penedès

region in the period 19959 200809, in comparison with the trends observed during the last 50 years, and its implications on phenology and yield. Temperature increases are higher than in previous time periods, which together with the irregular rainfall distribution throughout the year give rise to significant water deficits for vine development. Water deficits are being exacerbated during the last years by the increase of temperatures which imply an increase of evapotranspiration. The dates at which each phenological stage starts and the length of the different phenological stages are affected by temperature (accumulated degreedays and daily air temperature difference), precipitation and water accumulated into the soil. Winegrape yield was also influenced by soil water availability.

Evapotraspiration Mediterranean climate E Spain phenology trendsyield Climate change and its potential impacts on viticulture and viniculture have become

increasingly important as a consequence of changes in earth surface characteristics associated with the increase in greenhouse gases and changes in global temperatures, radiation budget and hydrological cycles. Vines, one of the most extensive crops in some parts of the Mediterranean Spanish area, which are cultivated under rainfed conditions, may therefore be one of the crops that suffer the consequences of climate change. According to Olesen and Bindi (2002), in the current production areas the yield variability (fruit production and quality) may be higher in the future than it is at present. However, some areas may suffer negative effects such as water stress due to a reduction in water availability and shortening of the ripening period, with harvest occurring during time with high temperatures, which may have negative impacts on wine quality (Duchêne and Schneider, 2005). Examples of impacts of temperature changes, frost occurrence and growing season lengths on grape productivity are found in the literature (Jones et al., 2005).

The Mediterranean climate is characterized by dry and warm summers and two wet seasons (spring and autumn), in which most rainfall is recorded, but there is high variability from year to year. Some authors indicate decreasing precipitation trends for the Mediterranean (Karl, 1998) and significant changes in extreme events concentrated in a small number of events such as more frequent and extreme droughts, increases in cool season precipitation, and warm season drying (Easterling et al. 2000). This study presents an analysis of temperature and precipitation and their

3 - 23



possible impacts on grape development in the Alt Penedès (NE Spain), which is dry farming area with a long tradition on vine cultivation but with significant changes during the last decades.

The Alt Penedès region, located in the northeast Spain, is part of the Penedès Tertiary DepressionThe main soil types in the area are (MartínezCasasnovas and Ramos, 2009). Calcilutites (marls) are the main lithological material, with occasional sandstones and conglomerates and due to labours carried out in the field to facilitate labour mechanization, soil profiles have been disturbed, leaving on top of the surface materials which are poor in organic matter and very week soil structure. This affects water intake and further redistribution, with an important limitation of water availability for the vine.

The climate is Mediterranean, with a mean annual temperature of 15ºC and a mean annual rainfall of 550 mm, mainly concentrated in spring (April to June) and autumn (September to November). Highintensity rainstorms are particularly frequent in autumn.

The Alt Penedès area has a long tradition of vineyard cultivation under the Penedès and Cava Designation of Origins (DO). Vineyards are the main land use, representing 80% of the cultivated area, about 17500 ha (CRDOP, 2008), (about 67% of the DO surface) 82% of them planted with white varieties, being the most representative: Xarelo (31%), Macabeo (25.7%), Parellada (18.6%) and Chardonnay (5%). Among the red varieties Cabernet Sauvignon (4.3%) and Merlot (7.3%) are the most extended (CRDOP, 2008). At present, about 50% of vineyards have been transformed into new vineyards, in which almost all labours are mechanised. For this study four observatories distributed in the area and two vineyards close to them are considered to asses the effects of climate trends on production and phenology. Precipitation and temperature data were recorded in Vilafranca del Penedès (VP) (41.434º; 1.419E; 230 m) from 1952 to 2009 and in Sant Sadurní d’Anoia (SSA) (41,208º; 1.791E; 164 m), La Granada (LG) (41.367º; 1.724E; 238 m) and Els Hostalets de Pierola (EHP) (41.526º; 1.805E; 312 m) from 1996 to 2009. Daily rainfall, maximum and minimum temperatures and evapotranspiration were analysed for the growing season (AprilSeptember) and for each phenological stage (budbreakbloom [], bloomveraison [V] and ripening [R]). Additionally, several bioclimatic indexes such as accumulated effective temperature [Tef(Tm10ºC)] and Huglin [HI] and Winkler [WI] indexes were analyzed.

Dates at which each phenology stage started was examined in two vineyards planted with Xarelo: beginning of budbreak, early bloom (beginning of flowering, grapes colour change and grapes ripe for harvesting. In the text, the different stages are identified as: dormant period (D): time between 1srt Novemberbudbreak start; budbreakbloom (); bloomveraison (V); ripening (veraisonharvest: R), post harvest (PH): time between harvest and 31st October. The length of each phenological period was evaluated for each year.

Crop yield was also evaluated in the same vineyards. The ratio between precipitation and crop evapotranspiration, estimated using the crop coefficients proposed by the FAO (Allen et al., 1998), was evaluated as well as the relationship between temperature and the related indexes and the date at which the different phenological events took place. Results related to Xarelo variety are presented. In addition, the relationship between grape production and temperature and water

availability was analysed. For these analysis, a step multiple regression analysis (forward variable selection) was performed using the Statgrapics 5.1 software.

Table 1 shows the

mean values of climate parameters related to the growing season (GS) recorded in four observatories in the area during the period (19962009). The mean temperature (TGSM) ranged between 18.7 and 22.2ºC, with mean maximum temperature (TGSMax) ranging between 23.8 and 25.9 and mean minimum temperatures (TGSMin) ranging between 12.4 and 13.7ºC. During the last decades, increasing mean, maximum and minimum temperature trends were observed in the growing season in this study area (Ramos et al., 2008). Growing season mean temperatures increased in about 0.04ºC/year in VP, during the last 50 years, which implies and increase of about 2.2 ºC in the period (19522006). If we focus on the last 14 years (19962009), the trend of the mean temperature is similar (0.038 ºC/year), but the minimum temperature seems to increase much more than in previous decades (ranging between 0.105 and 0.149ºC/year depending on the observatory). Other bioclimatic indices which also showed significant increasing trends during the past decades such as the Winkler index (trend=7.81GDD/year) or the Huglin index (7.24GDD/year), show now higher values in VP (20.9 and 22.2 DDD/year), although lower values are found in other observatories (e.g. (8.23 and 11.2 GDD/year, respectively in EHP and 6.5 GDD/year in LG). While the mean WI index value during the past decades (1860 GDD) places this region in Winkler region III (Ramos et al., 2008), it is observed that at present the mean values are clearly in region IV. This means the need of adapting some varieties to this new situation. Particularly, some of the most extended varieties in the region could be affected (Parellada or Chardonnay). One of the direct consequences of the temperature increase is the higher evapotranspiration rates. During the last 14 years evapotraspiration rates showed a significant positive trend in all observatories. However, those ratios were on average 9.3 mm/year in SSA, 11.1 mm/year in and EHP and about 3 mm/year in VP and LG, observatory where the average evapotranspiration was already higher. This increase means an annual evapotraspiration increase ranging between 1 and 2.3%.

The rainfall characteristics of the Mediterranean climate, with high variability from year to year and within the year, makes difficult to confirm precipitation trends. At annual scale no significant trends may be confirmed, but higher variability may be observed: rainfall increases in winter and autumn and decreases in spring, mainly affecting the bloomveraison stage. A significant decrease ranging between 4.6 and 5.9 mm/year was observed in that period, which represent about 10% of the rainfall recorded during that crop stage. The irregular distribution of the rainfall and its decrease during the growing period makes, that very often, the vines suffer deficits which may not be covered by the water reserves accumulated into the soil profile.

Using the data belonging to one of the observatories included in this analysis (EHP) an analysis of the water availability for the crop during the last years was done. In this analysis the relationship between rainfall and evapotranspiration in each crop stage as well as the accumulated water from the previous stage was considered.

3 - 24



possible impacts on grape development in the Alt Penedès (NE Spain), which is dry farming area with a long tradition on vine cultivation but with significant changes during the last decades.

The Alt Penedès region, located in the northeast Spain, is part of the Penedès Tertiary DepressionThe main soil types in the area are (MartínezCasasnovas and Ramos, 2009). Calcilutites (marls) are the main lithological material, with occasional sandstones and conglomerates and due to labours carried out in the field to facilitate labour mechanization, soil profiles have been disturbed, leaving on top of the surface materials which are poor in organic matter and very week soil structure. This affects water intake and further redistribution, with an important limitation of water availability for the vine.

The climate is Mediterranean, with a mean annual temperature of 15ºC and a mean annual rainfall of 550 mm, mainly concentrated in spring (April to June) and autumn (September to November). Highintensity rainstorms are particularly frequent in autumn.

The Alt Penedès area has a long tradition of vineyard cultivation under the Penedès and Cava Designation of Origins (DO). Vineyards are the main land use, representing 80% of the cultivated area, about 17500 ha (CRDOP, 2008), (about 67% of the DO surface) 82% of them planted with white varieties, being the most representative: Xarelo (31%), Macabeo (25.7%), Parellada (18.6%) and Chardonnay (5%). Among the red varieties Cabernet Sauvignon (4.3%) and Merlot (7.3%) are the most extended (CRDOP, 2008). At present, about 50% of vineyards have been transformed into new vineyards, in which almost all labours are mechanised. For this study four observatories distributed in the area and two vineyards close to them are considered to asses the effects of climate trends on production and phenology. Precipitation and temperature data were recorded in Vilafranca del Penedès (VP) (41.434º; 1.419E; 230 m) from 1952 to 2009 and in Sant Sadurní d’Anoia (SSA) (41,208º; 1.791E; 164 m), La Granada (LG) (41.367º; 1.724E; 238 m) and Els Hostalets de Pierola (EHP) (41.526º; 1.805E; 312 m) from 1996 to 2009. Daily rainfall, maximum and minimum temperatures and evapotranspiration were analysed for the growing season (AprilSeptember) and for each phenological stage (budbreakbloom [], bloomveraison [V] and ripening [R]). Additionally, several bioclimatic indexes such as accumulated effective temperature [Tef(Tm10ºC)] and Huglin [HI] and Winkler [WI] indexes were analyzed.

Dates at which each phenology stage started was examined in two vineyards planted with Xarelo: beginning of budbreak, early bloom (beginning of flowering, grapes colour change and grapes ripe for harvesting. In the text, the different stages are identified as: dormant period (D): time between 1srt Novemberbudbreak start; budbreakbloom (); bloomveraison (V); ripening (veraisonharvest: R), post harvest (PH): time between harvest and 31st October. The length of each phenological period was evaluated for each year.

Crop yield was also evaluated in the same vineyards. The ratio between precipitation and crop evapotranspiration, estimated using the crop coefficients proposed by the FAO (Allen et al., 1998), was evaluated as well as the relationship between temperature and the related indexes and the date at which the different phenological events took place. Results related to Xarelo variety are presented. In addition, the relationship between grape production and temperature and water

availability was analysed. For these analysis, a step multiple regression analysis (forward variable selection) was performed using the Statgrapics 5.1 software.

Table 1 shows the

mean values of climate parameters related to the growing season (GS) recorded in four observatories in the area during the period (19962009). The mean temperature (TGSM) ranged between 18.7 and 22.2ºC, with mean maximum temperature (TGSMax) ranging between 23.8 and 25.9 and mean minimum temperatures (TGSMin) ranging between 12.4 and 13.7ºC. During the last decades, increasing mean, maximum and minimum temperature trends were observed in the growing season in this study area (Ramos et al., 2008). Growing season mean temperatures increased in about 0.04ºC/year in VP, during the last 50 years, which implies and increase of about 2.2 ºC in the period (19522006). If we focus on the last 14 years (19962009), the trend of the mean temperature is similar (0.038 ºC/year), but the minimum temperature seems to increase much more than in previous decades (ranging between 0.105 and 0.149ºC/year depending on the observatory). Other bioclimatic indices which also showed significant increasing trends during the past decades such as the Winkler index (trend=7.81GDD/year) or the Huglin index (7.24GDD/year), show now higher values in VP (20.9 and 22.2 DDD/year), although lower values are found in other observatories (e.g. (8.23 and 11.2 GDD/year, respectively in EHP and 6.5 GDD/year in LG). While the mean WI index value during the past decades (1860 GDD) places this region in Winkler region III (Ramos et al., 2008), it is observed that at present the mean values are clearly in region IV. This means the need of adapting some varieties to this new situation. Particularly, some of the most extended varieties in the region could be affected (Parellada or Chardonnay). One of the direct consequences of the temperature increase is the higher evapotranspiration rates. During the last 14 years evapotraspiration rates showed a significant positive trend in all observatories. However, those ratios were on average 9.3 mm/year in SSA, 11.1 mm/year in and EHP and about 3 mm/year in VP and LG, observatory where the average evapotranspiration was already higher. This increase means an annual evapotraspiration increase ranging between 1 and 2.3%.

The rainfall characteristics of the Mediterranean climate, with high variability from year to year and within the year, makes difficult to confirm precipitation trends. At annual scale no significant trends may be confirmed, but higher variability may be observed: rainfall increases in winter and autumn and decreases in spring, mainly affecting the bloomveraison stage. A significant decrease ranging between 4.6 and 5.9 mm/year was observed in that period, which represent about 10% of the rainfall recorded during that crop stage. The irregular distribution of the rainfall and its decrease during the growing period makes, that very often, the vines suffer deficits which may not be covered by the water reserves accumulated into the soil profile.

Using the data belonging to one of the observatories included in this analysis (EHP) an analysis of the water availability for the crop during the last years was done. In this analysis the relationship between rainfall and evapotranspiration in each crop stage as well as the accumulated water from the previous stage was considered.

3 - 25



Figure 1a shows the water available (P ETc) from the dormant period to the end of crop cycle. In only five (1996, 1997, 2002, 2004 and 2007) of the 14 years included in the analysis, precipitation exceed evapotranspiration needs during the whole growing cycle. In four years more (1999, 2000, 2004 and 2008) precipitation recorded during the dormant period helped to cover water needs during the budbreakbloom stage, but during the rest of the year, the plant suffer significant water deficits. However, there were some years, dry years or years with very irregular distribution, in which the accumulated precipitation was not enough to cover evapotraspiration in any stage of the whole growing season (2001 and 2005).

Even more, taking into account the rainfall intensity and the limited soil infiltration capacity in many cases, water restrictions are even higher. A specific study carried out in two vineyards planted with Xarelo, one of the varieties most commonly extended in the area, shows how water deficits recorded, not only during the ripening period but in earlier stages, are even higher. We can see how water deficits during the bloom to veraison and during the ripening period were significantly increased in almost all years (Fig. 1b). These water deficits are affected for both irregular distribution and higher temperatures, which gave rise to higher evapotranspiration rates. Under these results we could understand the difficulty of extracting general conclusions. Even in the case that total precipitation does not change so much, it is clear that a decrease of precipitation during some stages of the crop have additional effects.

Table 1: Mean temperature and precipitation indices and trends covering the growing period in four observatories of the study area (19962009)

Vilafranca del Penedès

Trend/ year

La Granada Trend /year

Sant Sadurní D’Anoia

Trend/ year

Els Hostalets de Pierola

Trend/ year

TGSM 19.8±0.9 0.102* 19.4±0.5 22.2±3.3 18.7±0.9 0.038 TGSMax 24.0±1.3 0.105* 25.9±0.8 25.2±0.9 0.106 23.8±0.8 0.149* TGSMin 13.5±0.9 0.108* 13.7±0.4 0.072* 12.4±0.7 0.084 12.9±0.9 0.068

WI 2104±140 20.9* 2119+180 6.5 * 2120+180 35.2 2066+160 11.9* HI 2813±150 22.2* 2573±170 12.8 2575±170 27.5 2385±130 8.23*

PGS 336±110 322±100 322±120 350±150 ETcGS 414±18 2.7 * 482±20 3.8 * 322±120 9.32 * 366±100 11.1 * PBB 68±50 81±60 87±60 79±60 PBV 47±30 5.34* 34.2±30 4.84 * 36±30 4.6 * 76±70 5.9 * PR 103±38 125±20 96±45 90.6±70

* means significant differences at 95% level

According to the Consell Regulador de la Denominació d’Origen Penedès, the ranks of the vintage from the region are given as good, very good and excellent for both white and red wines. In 12 out of the last 15 years the vintage qualification was very good plus one more excellent (CRDOP, 2010). From this point we could think that there is not a serious impact of climate or water availability on the sector despite different authors having already pointed out the impacts that climate change could have on wine quality (Duchêne and Schneider, 2005; White et al., 2006; Makra et al., 2009).

However, water availability may have a direct impact on yield with a significant economic impact for the sector. In order to evaluate the relationship between grape production and temperature and water stored into the soil, a step multiple regression analysis (forward variable selection) was performed. Although several variables related to temperature and soil water have influence on yield when they are tested isolated (negative effect: effective temperature corresponding to D+BB period [(Tef_D+BB), to BV period (Tef_BV) or to the whole growing

season (Tef_WGS); positive effect: accumulated water during D+BB period (WA_D+BB)], if all these variables are combined in a multiple regression analysis, the only significant value was the accumulated water during the budbreakbloom stage, which represents about 47% of total variance (Tab. 2).

Fig. 1. Water availability during each stage of the growing period (BB: budbreakbloom; BV:

bloomveraison; R: ripening) of a vineyard planted with Xarelo): a) precipitation – evapotranspiration in each stage: b) precipitationrunoff evapotranspiration in each stage

On the other hand, variables related to phenology such as the dates at which different phenological stages start are also influenced by water availability as well as by temperature. In particular, a significant correlation was found for this variety between the data at which veraison starts and the effective temperature during the bloomveraison period (negative correlation, which represents about 29% of total variance), and between that variable and the water available during the same period (positive correlation, representing about 38% of the variance) (Tab. 2)

Tab. 2. Fit parameters for Xarelo grapevine yield and some phenological dates in relation to effective temperature during the bloomveraison period (Tef_BV)) and soil water available

during budbreakbloom (WA_BB) and during bloomveraison (WW_BV) periods Variety variable parameter R2 (%) FRatio PValue Xarelo

Yield (kg/ha) Constant

WA_BB

17.97

43.47

6.92

0.04 Veraison data Tef_BV

WA_ BV 0.106 0.081

28.78 37.95

4.04 6.11

0.07 0.02

The high variability of the Mediterranean climate together with the high intensity rainfall and

the increasing temperature trends give rise to significant water deficits for vine development in the Penedès region, in which vineyards have been cultivated for centuries without irrigation. Significant water deficits have been observed not only in dry years but also in years with total rainfall above the mean value in the area. Water deficits during the growing season may be in part supplemented by water accumulated during the dormant period, but in most years water balance is negative for the whole growing period. Water deficits are being exacerbated during the last years by the increase of temperatures, which results in an increase of evapotranspiration.

Winegrape yield and the dates at which each phenological stage starts are affected by temperature (accumulated degreedays) and by soil water availability.

-250

-200

-150

-100

-50

0

50

100

150

200

250

300

96 97 98 99 00 01 02 03 04 05 06 07 08 09

Wat

erav

aila

ble

( mm

)

BB BV R-250

-200

-150

-100

-50

0

50

100

150

200

250

300

Wat

erav

aila

ble

( mm

)

BB BV R

96 97 98 99 00 01 02 03 04 05 06 07 08 09

-250

-200

-150

-100

-50

0

50

100

150

200

250

300

96 97 98 99 00 01 02 03 04 05 06 07 08 09

Wat

erav

aila

ble

( mm

)

BB BV R-250

-200

-150

-100

-50

0

50

100

150

200

250

300

Wat

erav

aila

ble

( mm

)

BB BV R

96 97 98 99 00 01 02 03 04 05 06 07 08 09

a b -250

-200

-150

-100

-50

0

50

100

150

200

250

300

96 97 98 99 00 01 02 03 04 05 06 07 08 09

Wat

erav

aila

ble

( mm

)

BB BV R-250

-200

-150

-100

-50

0

50

100

150

200

250

300

Wat

erav

aila

ble

( mm

)

BB BV R

96 97 98 99 00 01 02 03 04 05 06 07 08 09

-250

-200

-150

-100

-50

0

50

100

150

200

250

300

96 97 98 99 00 01 02 03 04 05 06 07 08 09

Wat

erav

aila

ble

( mm

)

BB BV R-250

-200

-150

-100

-50

0

50

100

150

200

250

300

Wat

erav

aila

ble

( mm

)

BB BV R

96 97 98 99 00 01 02 03 04 05 06 07 08 09

a b

3 - 26



Figure 1a shows the water available (P ETc) from the dormant period to the end of crop cycle. In only five (1996, 1997, 2002, 2004 and 2007) of the 14 years included in the analysis, precipitation exceed evapotranspiration needs during the whole growing cycle. In four years more (1999, 2000, 2004 and 2008) precipitation recorded during the dormant period helped to cover water needs during the budbreakbloom stage, but during the rest of the year, the plant suffer significant water deficits. However, there were some years, dry years or years with very irregular distribution, in which the accumulated precipitation was not enough to cover evapotraspiration in any stage of the whole growing season (2001 and 2005).

Even more, taking into account the rainfall intensity and the limited soil infiltration capacity in many cases, water restrictions are even higher. A specific study carried out in two vineyards planted with Xarelo, one of the varieties most commonly extended in the area, shows how water deficits recorded, not only during the ripening period but in earlier stages, are even higher. We can see how water deficits during the bloom to veraison and during the ripening period were significantly increased in almost all years (Fig. 1b). These water deficits are affected for both irregular distribution and higher temperatures, which gave rise to higher evapotranspiration rates. Under these results we could understand the difficulty of extracting general conclusions. Even in the case that total precipitation does not change so much, it is clear that a decrease of precipitation during some stages of the crop have additional effects.

Table 1: Mean temperature and precipitation indices and trends covering the growing period in four observatories of the study area (19962009)

Vilafranca del Penedès

Trend/ year

La Granada Trend /year

Sant Sadurní D’Anoia

Trend/ year

Els Hostalets de Pierola

Trend/ year

TGSM 19.8±0.9 0.102* 19.4±0.5 22.2±3.3 18.7±0.9 0.038 TGSMax 24.0±1.3 0.105* 25.9±0.8 25.2±0.9 0.106 23.8±0.8 0.149* TGSMin 13.5±0.9 0.108* 13.7±0.4 0.072* 12.4±0.7 0.084 12.9±0.9 0.068

WI 2104±140 20.9* 2119+180 6.5 * 2120+180 35.2 2066+160 11.9* HI 2813±150 22.2* 2573±170 12.8 2575±170 27.5 2385±130 8.23*

PGS 336±110 322±100 322±120 350±150 ETcGS 414±18 2.7 * 482±20 3.8 * 322±120 9.32 * 366±100 11.1 * PBB 68±50 81±60 87±60 79±60 PBV 47±30 5.34* 34.2±30 4.84 * 36±30 4.6 * 76±70 5.9 * PR 103±38 125±20 96±45 90.6±70

* means significant differences at 95% level

According to the Consell Regulador de la Denominació d’Origen Penedès, the ranks of the vintage from the region are given as good, very good and excellent for both white and red wines. In 12 out of the last 15 years the vintage qualification was very good plus one more excellent (CRDOP, 2010). From this point we could think that there is not a serious impact of climate or water availability on the sector despite different authors having already pointed out the impacts that climate change could have on wine quality (Duchêne and Schneider, 2005; White et al., 2006; Makra et al., 2009).

However, water availability may have a direct impact on yield with a significant economic impact for the sector. In order to evaluate the relationship between grape production and temperature and water stored into the soil, a step multiple regression analysis (forward variable selection) was performed. Although several variables related to temperature and soil water have influence on yield when they are tested isolated (negative effect: effective temperature corresponding to D+BB period [(Tef_D+BB), to BV period (Tef_BV) or to the whole growing

season (Tef_WGS); positive effect: accumulated water during D+BB period (WA_D+BB)], if all these variables are combined in a multiple regression analysis, the only significant value was the accumulated water during the budbreakbloom stage, which represents about 47% of total variance (Tab. 2).

Fig. 1. Water availability during each stage of the growing period (BB: budbreakbloom; BV:

bloomveraison; R: ripening) of a vineyard planted with Xarelo): a) precipitation – evapotranspiration in each stage: b) precipitationrunoff evapotranspiration in each stage

On the other hand, variables related to phenology such as the dates at which different phenological stages start are also influenced by water availability as well as by temperature. In particular, a significant correlation was found for this variety between the data at which veraison starts and the effective temperature during the bloomveraison period (negative correlation, which represents about 29% of total variance), and between that variable and the water available during the same period (positive correlation, representing about 38% of the variance) (Tab. 2)

Tab. 2. Fit parameters for Xarelo grapevine yield and some phenological dates in relation to effective temperature during the bloomveraison period (Tef_BV)) and soil water available

during budbreakbloom (WA_BB) and during bloomveraison (WW_BV) periods Variety variable parameter R2 (%) FRatio PValue Xarelo

Yield (kg/ha) Constant

WA_BB

17.97

43.47

6.92

0.04 Veraison data Tef_BV

WA_ BV 0.106 0.081

28.78 37.95

4.04 6.11

0.07 0.02

The high variability of the Mediterranean climate together with the high intensity rainfall and

the increasing temperature trends give rise to significant water deficits for vine development in the Penedès region, in which vineyards have been cultivated for centuries without irrigation. Significant water deficits have been observed not only in dry years but also in years with total rainfall above the mean value in the area. Water deficits during the growing season may be in part supplemented by water accumulated during the dormant period, but in most years water balance is negative for the whole growing period. Water deficits are being exacerbated during the last years by the increase of temperatures, which results in an increase of evapotranspiration.

Winegrape yield and the dates at which each phenological stage starts are affected by temperature (accumulated degreedays) and by soil water availability.

-250

-200

-150

-100

-50

0

50

100

150

200

250

300

96 97 98 99 00 01 02 03 04 05 06 07 08 09

Wat

erav

aila

ble

( mm

)

BB BV R-250

-200

-150

-100

-50

0

50

100

150

200

250

300

Wat

erav

aila

ble

( mm

)

BB BV R

96 97 98 99 00 01 02 03 04 05 06 07 08 09

-250

-200

-150

-100

-50

0

50

100

150

200

250

300

96 97 98 99 00 01 02 03 04 05 06 07 08 09

Wat

erav

aila

ble

( mm

)

BB BV R-250

-200

-150

-100

-50

0

50

100

150

200

250

300

Wat

erav

aila

ble

( mm

)

BB BV R

96 97 98 99 00 01 02 03 04 05 06 07 08 09

a b -250

-200

-150

-100

-50

0

50

100

150

200

250

300

96 97 98 99 00 01 02 03 04 05 06 07 08 09

Wat

erav

aila

ble

( mm

)

BB BV R-250

-200

-150

-100

-50

0

50

100

150

200

250

300

Wat

erav

aila

ble

( mm

)

BB BV R

96 97 98 99 00 01 02 03 04 05 06 07 08 09

-250

-200

-150

-100

-50

0

50

100

150

200

250

300

96 97 98 99 00 01 02 03 04 05 06 07 08 09

Wat

erav

aila

ble

( mm

)

BB BV R-250

-200

-150

-100

-50

0

50

100

150

200

250

300

Wat

erav

aila

ble

( mm

)

BB BV R

96 97 98 99 00 01 02 03 04 05 06 07 08 09

a b

3 - 27



The lack of water resources in the area makes very difficult or excessively expensive to implement irrigation systems in the fields, and because of that vineyards may be seriously affected if temperature continues increasing. Water deficits, driven by the irregularly rainfall distribution and the increase of evapotranspiration in the stages in which vine water needs should not be restricted, may challenged the sustainability of some vineyards in the area. On the other hand, yield may be reduced due to higher water stress under a scenario of climate change.

This work was developed in the framework of the AMB1995158, AMB980481, REN2002

0042, AGL200500091AGR, AGL2009,085 research projects financed by the Spanish Ministry of Science and Innovation, developed in the same study area. Authors want to thank the farmer of the field for the information given about his vines.

CRDOP (Consejo Regulador Denominación de Origen Penedès), 2008. Anuario estadstico 2002008. Generalitat de Cataluña, Barcelona 24 pp

CRDOP (Consejo Regulador Denominación de Origen Penedès), 2010. Anyades D.O. Penedès (Vintages D.O. Penedès). http://www.dopenedes.es/includes/estadistiques/anyades _do _penedes_cat.pdf

Duchene E., Schneider C., 2005. Grapevine and climatic changes: A glance at the situation in Alsace. 24: 999

Easterling DR., Meehl GA., Parmensan C., Chagnon SA., Karl T., Mearns LO., 2000. Climate extremes: observation, modelling and impacts. 289: 2068204.

Jones GV., 2005. Climate change in the western United States grape growing regions. . 689: 41–60

Jones GV., White MA., Cooper OR., Storchmann K., 2005. Climate change and global wine quality. : 194.

Karl TR., 1998. Regional Trends and Variations of Temperature and Precipitation. In Watson, R.T., Zyinyowera M.C. and Moss R.H. (eds), . IPCC. Cambridge University Press. pp. 4114.

Makra L., Vitányi B., Gál A., Mika J., Matyasovszky I., Hirsch T. 2009. Wine Quantity and Quality Variations in Relation to Climatic Factors in the Tokaj (Hungary) Winegrowing Region. 60: 1221.

MartnezCasasnovas JA., Ramos MC., 2009. Soil alteration due to erosion, ploughing and levelling of vineyards in north east Spain. 25: 18192.

Olesen JE., Bindi M., 2002. Consequences of climate change for European agricultural productivity, land use and policy. European Journal of Agronomy 16: 29262.

Soil Survey Staff, 2006. Soil survey staff, keys to soil taxonomy. Department of Agriculture Soil Conservation Service, Washington D.C. U.S.

White MA., Diffenbaugh NS., Jones GV., Pal JS., Giorgi F., 2006. Extreme heat reduces and shifts United States premium wine production in the 21st century. PNAS 10 (0): 112111222.

EFFETTI DEL CAMBIAMENTO CLIMATICO EUROPEO SULLEEPOCHE DI VENDEMMIA IN ABRUZZO

B. Di Lena(1), (2) , L. Mariani(3), F. Antenucci(2), O. Silvestroni(1)

(1)Dip. Scienze Ambientali e delle Produzioni Vegetali, Università Politecnica delle Marche, Via Breccebianche, 60131 Ancona.(2)Regione Abruzzo – Arssa - Centro Agrometeorologico Regionale, C.da Colle Comune, 66020 Scerni (Chieti).(3)Università di Milano- Dipartimento di Produzione Vegetale, Via Celoria, Milano

RiassuntoI dati termo-pluviometrici del periodo 1971-2009 registrati da alcune stazioni della regioneAbruzzo sono stati analizzati adottando alcuni semplici indici climatici e bioclimatici. E’ statovalutato il verificarsi di cambiamenti climatici così come le loro ripercussioni sulle date diinizio vendemmia. La data di vendemmia è risultata significativamente influenzata dalledisponibilità termiche e in particolare dalle Ore Normali di Caldo (NHH) cumulate nelperiodo marzo-giugno. L’analisi statistica dei trend temporali dell’ accumulo di NHH inmarzo-giugno ha individuato una discontinuità climatica che ricade nel 1984 per la collinalitoranea centrale, nel 1997 per la collina litoranea meridionale e nel 1998 per la collinainterna del pescarese. Questi punti di discontinuità sono risultati in buon accordo con i puntidi discontinuità delle date di inizio raccolta e possono pertanto rappresentare lo spartiacque trala precedente e l’attuale fase climatica. Quest’ultima si caratterizza per un anticipo della datadi raccolta rispettivamente di 10 giorni per la collina litoranea meridionale , 15 per la collinalitoranea centrale e 14 per la collina interna.

Parole chiaveVitis vinifera, fenologia, ore normali di caldo

AbstractThermo-pluviometric data registered in the period 1971-2009 by three hillside stations of theAbruzzi located in maritime areas (central and southern part of the region) and in the internalzone were analyzed adopting some simple climatic and bioclimatic indices. Occurrence ofclimate change was evaluated as well as its influence on harvest dates. Harvest dates weresignificantly influenced by thermal availability, mainly when it was measured by NormalHeat Hours referred to the period March-June (NHH march-june). The statistical analysis ofthe temporal trends of NHH march-june has identified change-points occurred in a lapse oftime from 1984 to 1998. The first abrupt change happened in central maritime area (1984),followed in 1997 and 1998 seasons by change-points respectively registered in southernmaritime area in the internal zone. These NHH march-june break-points were in a goodrelationship with harvest date break-points and seem to well represent the watershed betweenthe previous and the current climatic phase. This latter is characterized by an advance inharvest date around 10 days in southern maritime area and averaging 14-15 days in centralmaritime area and internal zone.

Key-wordsVitis vinifera, climate change, harvest date

3 - 28



The lack of water resources in the area makes very difficult or excessively expensive to implement irrigation systems in the fields, and because of that vineyards may be seriously affected if temperature continues increasing. Water deficits, driven by the irregularly rainfall distribution and the increase of evapotranspiration in the stages in which vine water needs should not be restricted, may challenged the sustainability of some vineyards in the area. On the other hand, yield may be reduced due to higher water stress under a scenario of climate change.

This work was developed in the framework of the AMB1995158, AMB980481, REN2002

0042, AGL200500091AGR, AGL2009,085 research projects financed by the Spanish Ministry of Science and Innovation, developed in the same study area. Authors want to thank the farmer of the field for the information given about his vines.

CRDOP (Consejo Regulador Denominación de Origen Penedès), 2008. Anuario estadstico 2002008. Generalitat de Cataluña, Barcelona 24 pp

CRDOP (Consejo Regulador Denominación de Origen Penedès), 2010. Anyades D.O. Penedès (Vintages D.O. Penedès). http://www.dopenedes.es/includes/estadistiques/anyades _do _penedes_cat.pdf

Duchene E., Schneider C., 2005. Grapevine and climatic changes: A glance at the situation in Alsace. 24: 999

Easterling DR., Meehl GA., Parmensan C., Chagnon SA., Karl T., Mearns LO., 2000. Climate extremes: observation, modelling and impacts. 289: 2068204.

Jones GV., 2005. Climate change in the western United States grape growing regions. . 689: 41–60

Jones GV., White MA., Cooper OR., Storchmann K., 2005. Climate change and global wine quality. : 194.

Karl TR., 1998. Regional Trends and Variations of Temperature and Precipitation. In Watson, R.T., Zyinyowera M.C. and Moss R.H. (eds), . IPCC. Cambridge University Press. pp. 4114.

Makra L., Vitányi B., Gál A., Mika J., Matyasovszky I., Hirsch T. 2009. Wine Quantity and Quality Variations in Relation to Climatic Factors in the Tokaj (Hungary) Winegrowing Region. 60: 1221.

MartnezCasasnovas JA., Ramos MC., 2009. Soil alteration due to erosion, ploughing and levelling of vineyards in north east Spain. 25: 18192.

Olesen JE., Bindi M., 2002. Consequences of climate change for European agricultural productivity, land use and policy. European Journal of Agronomy 16: 29262.

Soil Survey Staff, 2006. Soil survey staff, keys to soil taxonomy. Department of Agriculture Soil Conservation Service, Washington D.C. U.S.

White MA., Diffenbaugh NS., Jones GV., Pal JS., Giorgi F., 2006. Extreme heat reduces and shifts United States premium wine production in the 21st century. PNAS 10 (0): 112111222.

EFFETTI DEL CAMBIAMENTO CLIMATICO EUROPEO SULLEEPOCHE DI VENDEMMIA IN ABRUZZO

B. Di Lena(1), (2) , L. Mariani(3), F. Antenucci(2), O. Silvestroni(1)

(1)Dip. Scienze Ambientali e delle Produzioni Vegetali, Università Politecnica delle Marche, Via Breccebianche, 60131 Ancona.(2)Regione Abruzzo – Arssa - Centro Agrometeorologico Regionale, C.da Colle Comune, 66020 Scerni (Chieti).(3)Università di Milano- Dipartimento di Produzione Vegetale, Via Celoria, Milano

RiassuntoI dati termo-pluviometrici del periodo 1971-2009 registrati da alcune stazioni della regioneAbruzzo sono stati analizzati adottando alcuni semplici indici climatici e bioclimatici. E’ statovalutato il verificarsi di cambiamenti climatici così come le loro ripercussioni sulle date diinizio vendemmia. La data di vendemmia è risultata significativamente influenzata dalledisponibilità termiche e in particolare dalle Ore Normali di Caldo (NHH) cumulate nelperiodo marzo-giugno. L’analisi statistica dei trend temporali dell’ accumulo di NHH inmarzo-giugno ha individuato una discontinuità climatica che ricade nel 1984 per la collinalitoranea centrale, nel 1997 per la collina litoranea meridionale e nel 1998 per la collinainterna del pescarese. Questi punti di discontinuità sono risultati in buon accordo con i puntidi discontinuità delle date di inizio raccolta e possono pertanto rappresentare lo spartiacque trala precedente e l’attuale fase climatica. Quest’ultima si caratterizza per un anticipo della datadi raccolta rispettivamente di 10 giorni per la collina litoranea meridionale , 15 per la collinalitoranea centrale e 14 per la collina interna.

Parole chiaveVitis vinifera, fenologia, ore normali di caldo

AbstractThermo-pluviometric data registered in the period 1971-2009 by three hillside stations of theAbruzzi located in maritime areas (central and southern part of the region) and in the internalzone were analyzed adopting some simple climatic and bioclimatic indices. Occurrence ofclimate change was evaluated as well as its influence on harvest dates. Harvest dates weresignificantly influenced by thermal availability, mainly when it was measured by NormalHeat Hours referred to the period March-June (NHH march-june). The statistical analysis ofthe temporal trends of NHH march-june has identified change-points occurred in a lapse oftime from 1984 to 1998. The first abrupt change happened in central maritime area (1984),followed in 1997 and 1998 seasons by change-points respectively registered in southernmaritime area in the internal zone. These NHH march-june break-points were in a goodrelationship with harvest date break-points and seem to well represent the watershed betweenthe previous and the current climatic phase. This latter is characterized by an advance inharvest date around 10 days in southern maritime area and averaging 14-15 days in centralmaritime area and internal zone.

Key-wordsVitis vinifera, climate change, harvest date

3 - 29



IntroduzioneLe recenti annate caratterizzate da siccità primaverile-estiva hanno alimentato il dibattito sui

probabili impatti dei cambiamenti climatici in specifici areali viticoli e molto interesse è statorivolto alle loro possibili ripercussioni sulla fenologia della vite e sulle date di vendemmia.Riteniamo che un tale dibattito non possa prescindere dagli aspetti di scala in virtù dei quali icambiamenti climatici a livello di macroclima, che nelle medie latitudini del pianeta sono dinorma il frutto di riconfigurazioni nella circolazione generale, si propagano al mesoclima(clima di areali viticoli relativamente ampi) e da questo al microclima (clima del singolovigneto). Da questo angolo di visuale un elemento cruciale per la viticoltura europea èrappresentato dal brusco cambiamento climatico dovuto al mutato regime del grande vorticepolare e di conseguenza delle grandi correnti occidentali (westerlies) che ha interessato lemedie latitudini dell’emisfero boreale negli anni ’80 del 20° secolo (Werner et al., 2000) e cheper l’areale europeo viene efficacemente descritto dal comportamento della NAO (NorthAtlantic Oscillation), indice circolatorio a macroscala che dal 1981 manifesta una sensibileanomalia positiva (Mariani, 2008). Le più immediate conseguenze a livello europeo di talebrusco cambiamento climatico sono state l’affievolirsi dell’apporto invernale di masse d’ariapolare continentale (PC, la gelida aria siberiana) e l’aumentato apporto di masse d’ariasubtropicale marittima (STm).

Da tali fenomeni è disceso un aumento delle temperature medie annue di circa 0,5-1°C nelleparti più settentrionali dell’areale viticolo europeo (tipo Cfb di Koeppen) e di 1-1,5°C nelleparti più meridionali dello stesso (tipo Csa di Koeppen). Meno univoci sono invece daconsiderare gli effetti sul quadro pluviometrico, anche se è noto che le fasi a NAO positivo sicaratterizzano per una maggiore aridità negli areali a clima Csa (Trouet et al., 2009). Fra leconseguenze più evidenti di tali fenomeni rientra l’anticipo di 6-25 giorni delle principali fasifenologiche della vite (fioritura, invaiatura e vendemmia) segnalato da Jones et al. (2005) peralcuni vitigni coltivati in diversi siti. Per cogliere le modalità con cui il cambiamentoclimatico europeo degli anni ’80 si è manifestato nelle diverse aree viticole italiane in ènecessario spingere l’analisi fino alla mesoscala, considerando singolarmente territori nonmolto estesi, come quello dell’Abruzzo, una regione ad orografia complessa dove laviticoltura è diffusa sia nella zona della collina litoranea che in quella della collina interna.Per questa Regione, l’analisi di alcuni indici statistici e bioclimatici ricavati da serie storichedi dati termo-pluviometrici giornalieri del periodo 1965-2007, ha evidenziato, durante il ciclovegetativo, una diminuzione delle precipitazioni nella fascia costiera e un aumento delletemperature nelle aree interne (Silvestroni et al., 2008). La stessa indagine ha evidenziatolungo la fascia collinare litoranea un aumento dell’indice di Huglin nel decennio antecedenteal 2007.

Alla luce di questi presupposti il presente lavoro è stato mirato a cogliere gli effetti delcambiamento climatico sulle date di inizio vendemmia, analizzando in particolare leinformazioni provenienti dai registri di tre importanti cantine abruzzesi e riferite al periodo1974-2009 con lo scopo di porre in luce:

- le relazioni tra date di inizio vendemmia della cv. Montepulciano e alcuni indici climaticie bioclimatici ricavati da serie storiche di dati termo-pluviometrici.

- l’eventuale presenza di discontinuità nelle serie storiche delle date di inizio vendemmia(“change points”) e degli indici bioclimatici applicando opportune metodologie statistiche(Bai e Perron, 1998 e 2003).

Materiali e MetodiLo studio delle serie storiche di date di inizio vendemmia della cv. Montepulciano nella

Regione Abruzzo, è stato effettuato utilizzando le informazioni ricavate dai registri dellecantine di Scerni, Vacri e Loreto Aprutino. Per le prime due, localizzate rispettivamente nellacollina litoranea meridionale e centrale della provincia di Chieti, è stato considerato il periodo1974-2009, mentre per l’ultima localizzata nella collina interna della provincia di Pescara èstato analizzato il periodo 1977-2009 (Fig.1).

La valutazione delle relazioniesistenti fra indici climatici ebioclimatici da un lato, e serie storichedelle date di inizio vendemmiadall’altro, è stata effettuata utilizzandoi dati termo-pluviometrici giornalieridel Servizio Idrografico Regionale.

Il data set era composto dai valoridelle temperature massime e minime edelle precipitazioni giornaliere. Latemperatura media giornaliera è stataottenuta facendo la semisomma deivalori delle temperature massime eminime.Prima di procedere ai normali controllidi consistenza interna e persistenzatemporale delle serie storiche sonostate verificate le informazioni sullestazioni di rilevamento, che non hannosubito, nel periodo considerato,modifiche tecniche e di posizione talida inficiare l’attendibilità dei datirilevati.

In particolare per la collina litoranea meridionale sono stati impiegati i dati della stazione diScerni, per la collina litoranea centrale quelli di Chieti, e per la collina interna quelli di Penne(Fig. 1)

Nel periodo aprile ottobre sono stati determinati: la temperatura media (ottenuta facendo lamedia dei valori medi giornalieri), il numero di giorni con temperature massime superiori a30°C, le precipitazioni totali e la media delle escursioni termiche giornaliere (espresse comedifferenza fra temperature massime e minime).

Nel periodo aprile settembre sono stati determinati per le tre località i tre indici proposti daTonietto e Carbonneau (2004) per la classificazione degli areali viticoli secondo un approcciomulticriteriale: Indice eliotermico di Huglin, Indice di Freschezza della notte (media delletemperature minime del mese di settembre) e Indice di Siccità (stima della disponibilitàpotenziale di acqua nel suolo). Nello stesso periodo sono state determinate anche le orenormali di caldo (NHH) e la sommatoria delle temperature attive considerando quale sogliatermica 10°C. Le NHH hanno consentito di valutare quantitativamente l’efficacia di oretrascorse a temperature diverse, evidenziando l’accumulo complessivo di risorse termicheutili per il processo indagato. Il valore della funzione normalizzata è pari a 0 sia con valoritermici inferiori a 7°C (Cardinale minimo - Cmin) che maggiori di 35°C (Cardinale massimo -

Fig.1 – Localizzazione degli areali oggetto dello studio Stazione meteorologica Cantina

3 - 30



IntroduzioneLe recenti annate caratterizzate da siccità primaverile-estiva hanno alimentato il dibattito sui

probabili impatti dei cambiamenti climatici in specifici areali viticoli e molto interesse è statorivolto alle loro possibili ripercussioni sulla fenologia della vite e sulle date di vendemmia.Riteniamo che un tale dibattito non possa prescindere dagli aspetti di scala in virtù dei quali icambiamenti climatici a livello di macroclima, che nelle medie latitudini del pianeta sono dinorma il frutto di riconfigurazioni nella circolazione generale, si propagano al mesoclima(clima di areali viticoli relativamente ampi) e da questo al microclima (clima del singolovigneto). Da questo angolo di visuale un elemento cruciale per la viticoltura europea èrappresentato dal brusco cambiamento climatico dovuto al mutato regime del grande vorticepolare e di conseguenza delle grandi correnti occidentali (westerlies) che ha interessato lemedie latitudini dell’emisfero boreale negli anni ’80 del 20° secolo (Werner et al., 2000) e cheper l’areale europeo viene efficacemente descritto dal comportamento della NAO (NorthAtlantic Oscillation), indice circolatorio a macroscala che dal 1981 manifesta una sensibileanomalia positiva (Mariani, 2008). Le più immediate conseguenze a livello europeo di talebrusco cambiamento climatico sono state l’affievolirsi dell’apporto invernale di masse d’ariapolare continentale (PC, la gelida aria siberiana) e l’aumentato apporto di masse d’ariasubtropicale marittima (STm).

Da tali fenomeni è disceso un aumento delle temperature medie annue di circa 0,5-1°C nelleparti più settentrionali dell’areale viticolo europeo (tipo Cfb di Koeppen) e di 1-1,5°C nelleparti più meridionali dello stesso (tipo Csa di Koeppen). Meno univoci sono invece daconsiderare gli effetti sul quadro pluviometrico, anche se è noto che le fasi a NAO positivo sicaratterizzano per una maggiore aridità negli areali a clima Csa (Trouet et al., 2009). Fra leconseguenze più evidenti di tali fenomeni rientra l’anticipo di 6-25 giorni delle principali fasifenologiche della vite (fioritura, invaiatura e vendemmia) segnalato da Jones et al. (2005) peralcuni vitigni coltivati in diversi siti. Per cogliere le modalità con cui il cambiamentoclimatico europeo degli anni ’80 si è manifestato nelle diverse aree viticole italiane in ènecessario spingere l’analisi fino alla mesoscala, considerando singolarmente territori nonmolto estesi, come quello dell’Abruzzo, una regione ad orografia complessa dove laviticoltura è diffusa sia nella zona della collina litoranea che in quella della collina interna.Per questa Regione, l’analisi di alcuni indici statistici e bioclimatici ricavati da serie storichedi dati termo-pluviometrici giornalieri del periodo 1965-2007, ha evidenziato, durante il ciclovegetativo, una diminuzione delle precipitazioni nella fascia costiera e un aumento delletemperature nelle aree interne (Silvestroni et al., 2008). La stessa indagine ha evidenziatolungo la fascia collinare litoranea un aumento dell’indice di Huglin nel decennio antecedenteal 2007.

Alla luce di questi presupposti il presente lavoro è stato mirato a cogliere gli effetti delcambiamento climatico sulle date di inizio vendemmia, analizzando in particolare leinformazioni provenienti dai registri di tre importanti cantine abruzzesi e riferite al periodo1974-2009 con lo scopo di porre in luce:

- le relazioni tra date di inizio vendemmia della cv. Montepulciano e alcuni indici climaticie bioclimatici ricavati da serie storiche di dati termo-pluviometrici.

- l’eventuale presenza di discontinuità nelle serie storiche delle date di inizio vendemmia(“change points”) e degli indici bioclimatici applicando opportune metodologie statistiche(Bai e Perron, 1998 e 2003).

Materiali e MetodiLo studio delle serie storiche di date di inizio vendemmia della cv. Montepulciano nella

Regione Abruzzo, è stato effettuato utilizzando le informazioni ricavate dai registri dellecantine di Scerni, Vacri e Loreto Aprutino. Per le prime due, localizzate rispettivamente nellacollina litoranea meridionale e centrale della provincia di Chieti, è stato considerato il periodo1974-2009, mentre per l’ultima localizzata nella collina interna della provincia di Pescara èstato analizzato il periodo 1977-2009 (Fig.1).

La valutazione delle relazioniesistenti fra indici climatici ebioclimatici da un lato, e serie storichedelle date di inizio vendemmiadall’altro, è stata effettuata utilizzandoi dati termo-pluviometrici giornalieridel Servizio Idrografico Regionale.

Il data set era composto dai valoridelle temperature massime e minime edelle precipitazioni giornaliere. Latemperatura media giornaliera è stataottenuta facendo la semisomma deivalori delle temperature massime eminime.Prima di procedere ai normali controllidi consistenza interna e persistenzatemporale delle serie storiche sonostate verificate le informazioni sullestazioni di rilevamento, che non hannosubito, nel periodo considerato,modifiche tecniche e di posizione talida inficiare l’attendibilità dei datirilevati.

In particolare per la collina litoranea meridionale sono stati impiegati i dati della stazione diScerni, per la collina litoranea centrale quelli di Chieti, e per la collina interna quelli di Penne(Fig. 1)

Nel periodo aprile ottobre sono stati determinati: la temperatura media (ottenuta facendo lamedia dei valori medi giornalieri), il numero di giorni con temperature massime superiori a30°C, le precipitazioni totali e la media delle escursioni termiche giornaliere (espresse comedifferenza fra temperature massime e minime).

Nel periodo aprile settembre sono stati determinati per le tre località i tre indici proposti daTonietto e Carbonneau (2004) per la classificazione degli areali viticoli secondo un approcciomulticriteriale: Indice eliotermico di Huglin, Indice di Freschezza della notte (media delletemperature minime del mese di settembre) e Indice di Siccità (stima della disponibilitàpotenziale di acqua nel suolo). Nello stesso periodo sono state determinate anche le orenormali di caldo (NHH) e la sommatoria delle temperature attive considerando quale sogliatermica 10°C. Le NHH hanno consentito di valutare quantitativamente l’efficacia di oretrascorse a temperature diverse, evidenziando l’accumulo complessivo di risorse termicheutili per il processo indagato. Il valore della funzione normalizzata è pari a 0 sia con valoritermici inferiori a 7°C (Cardinale minimo - Cmin) che maggiori di 35°C (Cardinale massimo -

Fig.1 – Localizzazione degli areali oggetto dello studio Stazione meteorologica Cantina

3 - 31



Cmax) mentre è pari a 1 se la temperatura assume il valore ottimale di 26°C (Cardinaleottimale - Copt). L’equazione adottata è quella descritta da Wang e Engel (1998):

Fvn(T) =(2(T-Cmin) *(Copt-Cmin) *(T- Cmin) 2 )/(Copt –Cmin)se Tmin <=T<=Tmax e Fvn(T) =0 se T<Cmin e T>Cmaxove : = ln(2/ln((Cmax-Cmin)/(Copt-Cmin)))Ai fini del calcolo delle NHH, le temperature orarie sono state ottenute dalle temperature

massime e minime applicando l’algoritmo di Parton e Logan (1981). Per il mese di settembre,assunto come periodo di pre-vendemmia, sono state inoltre calcolate le precipitazioni totali ela media delle escursioni termiche giornaliere.

Le relazioni tra le serie storiche delle date di inizio vendemmia (espresse come numero digiorni a decorrere dal 1 aprile) e gli indici climatici e bioclimatici sono state determinate conl’approccio statistico della regressione lineare semplice.

La presenza di discontinuità nell’andamento delle serie storiche (Todaro e Migliardi, 2000,2003 e 2004) è stata indagata utilizzando l’algoritmo di analisi di “change point”, presentenella libreria Strucchange del software R - http://www.r-project.org (Bai e Perron, 1998 e2003).

Risultati e discussioneLa caratterizzazione climatica, basata sui valori medi degli indici climatici e bioclimatici, ha

mostrato una elevata similarità delle località di Scerni e Chieti, situate rispettivamente nellacollina litoranea meridionale e centrale dell’Abruzzo. La località di Penne, situata nellacollina interna, si è distinta dalle altre per la maggiore entità delle precipitazioni, per i valoripiù contenuti delle temperature medie, della sommatoria delle temperature attive, delle orenormali di caldo e dell’Indice di Huglin. In questo areale si sono riscontrate anche minoricondizioni di siccità (Tab.1).

Tab. 1 Statistiche descrittive degli indici climatici e bioclimatici.Collina litoranea

meridionaleScerni

Collina litoraneacentraleChieti

Collina internaPenneINDICI CLIMATICI E BIOCLIMATICI

media dev.st media dev.st media dev.staprile-ottobreTemperatura media (°C) 19,9 0,8 20,1 0,7 19,6 0,8Numero giorni Tmax >30°C 28,8 15,5 34,3 14,8 30,6 13,4Precipitazioni totali totali (mm.) 378 127,7 385 122,0 477 115,4Media escursioni termiche giornaliere (°C) 8,0 1,0 8,3 1,3 8,4 0,8aprile-settembreIndice di Huglin 2364 187,2 2423 179,2 2322 190,3Indice di Siccità (mm.) -120 44,1 -121 49,5 -85 44,7Indice di Freschezza della notte (°C) 16,7 1,4 16,8 1,4 15,9 1,4Ore normali di caldo 3038 159,1 3045 119,1 2941 144,4Sommatoria temperature attive 1924 169,0 1972 140,9 1868 163,8settembreMedia escursioni termiche giornaliere (°C) 7,9 1,2 8,1 1,5 8,6 0,9Percipitazioni totali (mm.) 59 40,9 61 38,9 68 42,8

L’analisi delle serie storiche delle date di inizio vendemmia della cv. Montepulciano,eseguita con l’approccio della regressione lineare semplice (Tab.2), ha portato a stimare unprogressivo sensibile anticipo di questa operazione colturale, che nell’arco di 35 anni, sisarebbe attestato attorno a 16-18 giorni nella collina litoranea, valore del tutto analogo aquello calcolato per Loreto Aprutino, situato nella collina interna.

Gli effetti del cambiamento climatico sulle date di inizio vendemmia sono stati indagatiricorrendo nuovamente allo studio della regressione lineare semplice con gli indici climatici ebioclimatici già riportati in Tabella 1. Le date di inizio vendemmia sono risultatesignificativamente correlate, in tutti i tre areali, con le temperature medie del periodo aprile-ottobre, l’Indice di Huglin, la sommatoria delle temperature attive e le ore normali di caldo. Icoefficienti angolari negativi evidenziano che all’aumento degli indici suddetti ha fattoriscontro un anticipo della data di inizio vendemmia (Tab.3).

Tab. 2 Serie storiche di date di inizio vendemmia della cv. Montepulciano. Coefficientiangolari delle rette di regressione (β) e loro significatività.( ** = P0.01)

Areale Periodo β Anticipo in giorni Collina litoranea meridionale (Scerni) 1974-2009 -0.52 ** 18Collina litoranea centrale ( Vacri) 1974-2009 -0.46 ** 16Collina interna (Loreto Aprutino 1977-2009 -0.56 ** 17

Tab. 3 Relazioni tra gli indici climatici e bioclimatici e le date di inizio vendemmia.Coefficienti angolari delle rette di regressione (β) e coefficienti di determinazione (R2).(ns = non significativo; * = P0.05; ** = P0.01),.

Variabile Y(data inizio vendemmia)

Collina litoraneameridionale

Scerni

Collina litoraneacentrale

Vacri

Collina internaLoreto Aprutino

Variabile X(indici climatici e bioclimatici)

β R2 β R2 β R2

aprile-ottobreTemperatura media (°C) -5,73 ** 0,35 -5,40 ** 0,27 -6,64 ** 0,32Numero giorni Tmax >30°C -0,12 ns 0,06 -0,21 * 0,16 -0,35 ** 0,22Precipitazioni totali totali (mm.) 0,01 ns 0,02 0,02 * 0,12 0,03 * 0,17Media escursioni termiche giornaliere (°C) 2,89 * 0,12 -1,69 ns 0,01 -1,10 ns 0,01aprile-settembreIndice di Huglin -0,02 ** 0,22 -0,02 ** 0,30 -0,03 ** 0,27Indice di Siccità (mm.) 0,04 ns 0,05 0,04 ns 0,07 0,07 ns 0,11Indice di Freschezza della notte -1,47 ns 0,07 0,93 ns 0,03 1,24 ns 0,03Ore normali di caldo (NHH) -0,04 ** 0,56 -0,03 ** 0,29 -0,03 ** 0,23Sommatoria temperature attive -0,03 ** 0,36 -0,03 ** 0,29 -0,03 ** 0,31settembreMedia escursioni termiche giornaliere (°C) 2,97 ** 0,20 -1,15 ns 0.05 -0,45 ns 0,01Percipitazioni totali (mm.) -0,03 ns 0,03 -0,02 ns 0,01 0,03 ns 0,01ore normali di caldo (NHH) parzialiMarzo-Aprile -0.04 ** 0.19 -0.04 ** 0.20 -0.05 ** 0.23Marzo- Giugno -0.04 ** 0.45 -0.03 ** 0.39 -0.03 ** 0.30Maggio-Giugno -0.05 ** 0.44 -0.06 ** 0.39 -0.05 ** 0.22Luglio-Settembre -0.04 * 0.14 0.02 ns 0.03 0.06 ns 0.01

Da un esame approfondito della Tabella 3 emerge che la variabilità a carico delle date di iniziovendemmia sembra strettamente connessa alle variazioni stagionali riscontrate per le ore normalidi caldo (NHH), come testimoniano i valori del coefficiente di determinazione R2, che risultanosuperiori a quelli degli altri indici calcolati per i due siti della collina litoranea. Un ulterioreapprofondimento del potere descrittivo delle NHH rispetto alle date di vendemmia ha riguardatola loro suddivisione in alcuni sottoperiodi dell’anno, perché, in base alla legge del minimo diLiebig, le risorse termiche possono rappresentare un fattore limitante soprattutto nei periodi in cuiil loro livello è particolarmente basso o presenta elevata variabilità interannuale, come inprimavera. Il restringimento del periodo di accumulo di NHH alla sola stagione primaverile

3 - 32



Cmax) mentre è pari a 1 se la temperatura assume il valore ottimale di 26°C (Cardinaleottimale - Copt). L’equazione adottata è quella descritta da Wang e Engel (1998):

Fvn(T) =(2(T-Cmin) *(Copt-Cmin) *(T- Cmin) 2 )/(Copt –Cmin)se Tmin <=T<=Tmax e Fvn(T) =0 se T<Cmin e T>Cmaxove : = ln(2/ln((Cmax-Cmin)/(Copt-Cmin)))Ai fini del calcolo delle NHH, le temperature orarie sono state ottenute dalle temperature

massime e minime applicando l’algoritmo di Parton e Logan (1981). Per il mese di settembre,assunto come periodo di pre-vendemmia, sono state inoltre calcolate le precipitazioni totali ela media delle escursioni termiche giornaliere.

Le relazioni tra le serie storiche delle date di inizio vendemmia (espresse come numero digiorni a decorrere dal 1 aprile) e gli indici climatici e bioclimatici sono state determinate conl’approccio statistico della regressione lineare semplice.

La presenza di discontinuità nell’andamento delle serie storiche (Todaro e Migliardi, 2000,2003 e 2004) è stata indagata utilizzando l’algoritmo di analisi di “change point”, presentenella libreria Strucchange del software R - http://www.r-project.org (Bai e Perron, 1998 e2003).

Risultati e discussioneLa caratterizzazione climatica, basata sui valori medi degli indici climatici e bioclimatici, ha

mostrato una elevata similarità delle località di Scerni e Chieti, situate rispettivamente nellacollina litoranea meridionale e centrale dell’Abruzzo. La località di Penne, situata nellacollina interna, si è distinta dalle altre per la maggiore entità delle precipitazioni, per i valoripiù contenuti delle temperature medie, della sommatoria delle temperature attive, delle orenormali di caldo e dell’Indice di Huglin. In questo areale si sono riscontrate anche minoricondizioni di siccità (Tab.1).

Tab. 1 Statistiche descrittive degli indici climatici e bioclimatici.Collina litoranea

meridionaleScerni

Collina litoraneacentraleChieti

Collina internaPenneINDICI CLIMATICI E BIOCLIMATICI

media dev.st media dev.st media dev.staprile-ottobreTemperatura media (°C) 19,9 0,8 20,1 0,7 19,6 0,8Numero giorni Tmax >30°C 28,8 15,5 34,3 14,8 30,6 13,4Precipitazioni totali totali (mm.) 378 127,7 385 122,0 477 115,4Media escursioni termiche giornaliere (°C) 8,0 1,0 8,3 1,3 8,4 0,8aprile-settembreIndice di Huglin 2364 187,2 2423 179,2 2322 190,3Indice di Siccità (mm.) -120 44,1 -121 49,5 -85 44,7Indice di Freschezza della notte (°C) 16,7 1,4 16,8 1,4 15,9 1,4Ore normali di caldo 3038 159,1 3045 119,1 2941 144,4Sommatoria temperature attive 1924 169,0 1972 140,9 1868 163,8settembreMedia escursioni termiche giornaliere (°C) 7,9 1,2 8,1 1,5 8,6 0,9Percipitazioni totali (mm.) 59 40,9 61 38,9 68 42,8

L’analisi delle serie storiche delle date di inizio vendemmia della cv. Montepulciano,eseguita con l’approccio della regressione lineare semplice (Tab.2), ha portato a stimare unprogressivo sensibile anticipo di questa operazione colturale, che nell’arco di 35 anni, sisarebbe attestato attorno a 16-18 giorni nella collina litoranea, valore del tutto analogo aquello calcolato per Loreto Aprutino, situato nella collina interna.

Gli effetti del cambiamento climatico sulle date di inizio vendemmia sono stati indagatiricorrendo nuovamente allo studio della regressione lineare semplice con gli indici climatici ebioclimatici già riportati in Tabella 1. Le date di inizio vendemmia sono risultatesignificativamente correlate, in tutti i tre areali, con le temperature medie del periodo aprile-ottobre, l’Indice di Huglin, la sommatoria delle temperature attive e le ore normali di caldo. Icoefficienti angolari negativi evidenziano che all’aumento degli indici suddetti ha fattoriscontro un anticipo della data di inizio vendemmia (Tab.3).

Tab. 2 Serie storiche di date di inizio vendemmia della cv. Montepulciano. Coefficientiangolari delle rette di regressione (β) e loro significatività.( ** = P0.01)

Areale Periodo β Anticipo in giorni Collina litoranea meridionale (Scerni) 1974-2009 -0.52 ** 18Collina litoranea centrale ( Vacri) 1974-2009 -0.46 ** 16Collina interna (Loreto Aprutino 1977-2009 -0.56 ** 17

Tab. 3 Relazioni tra gli indici climatici e bioclimatici e le date di inizio vendemmia.Coefficienti angolari delle rette di regressione (β) e coefficienti di determinazione (R2).(ns = non significativo; * = P0.05; ** = P0.01),.

Variabile Y(data inizio vendemmia)

Collina litoraneameridionale

Scerni

Collina litoraneacentrale

Vacri

Collina internaLoreto Aprutino

Variabile X(indici climatici e bioclimatici)

β R2 β R2 β R2

aprile-ottobreTemperatura media (°C) -5,73 ** 0,35 -5,40 ** 0,27 -6,64 ** 0,32Numero giorni Tmax >30°C -0,12 ns 0,06 -0,21 * 0,16 -0,35 ** 0,22Precipitazioni totali totali (mm.) 0,01 ns 0,02 0,02 * 0,12 0,03 * 0,17Media escursioni termiche giornaliere (°C) 2,89 * 0,12 -1,69 ns 0,01 -1,10 ns 0,01aprile-settembreIndice di Huglin -0,02 ** 0,22 -0,02 ** 0,30 -0,03 ** 0,27Indice di Siccità (mm.) 0,04 ns 0,05 0,04 ns 0,07 0,07 ns 0,11Indice di Freschezza della notte -1,47 ns 0,07 0,93 ns 0,03 1,24 ns 0,03Ore normali di caldo (NHH) -0,04 ** 0,56 -0,03 ** 0,29 -0,03 ** 0,23Sommatoria temperature attive -0,03 ** 0,36 -0,03 ** 0,29 -0,03 ** 0,31settembreMedia escursioni termiche giornaliere (°C) 2,97 ** 0,20 -1,15 ns 0.05 -0,45 ns 0,01Percipitazioni totali (mm.) -0,03 ns 0,03 -0,02 ns 0,01 0,03 ns 0,01ore normali di caldo (NHH) parzialiMarzo-Aprile -0.04 ** 0.19 -0.04 ** 0.20 -0.05 ** 0.23Marzo- Giugno -0.04 ** 0.45 -0.03 ** 0.39 -0.03 ** 0.30Maggio-Giugno -0.05 ** 0.44 -0.06 ** 0.39 -0.05 ** 0.22Luglio-Settembre -0.04 * 0.14 0.02 ns 0.03 0.06 ns 0.01

Da un esame approfondito della Tabella 3 emerge che la variabilità a carico delle date di iniziovendemmia sembra strettamente connessa alle variazioni stagionali riscontrate per le ore normalidi caldo (NHH), come testimoniano i valori del coefficiente di determinazione R2, che risultanosuperiori a quelli degli altri indici calcolati per i due siti della collina litoranea. Un ulterioreapprofondimento del potere descrittivo delle NHH rispetto alle date di vendemmia ha riguardatola loro suddivisione in alcuni sottoperiodi dell’anno, perché, in base alla legge del minimo diLiebig, le risorse termiche possono rappresentare un fattore limitante soprattutto nei periodi in cuiil loro livello è particolarmente basso o presenta elevata variabilità interannuale, come inprimavera. Il restringimento del periodo di accumulo di NHH alla sola stagione primaverile

3 - 33



(periodo Marzo-Giugno) ha permesso di mantenere alto il coefficiente di determinazione, che èaddirittura aumentato in due areali su tre.

La variabilità temporale delle NHH per il periodo marzo – giugno è stata pertanto indagata alloscopo di individuare eventuali discontinuità attribuibili a cause climatiche. In tutte le zone si èregistrata una buona corrispondenza tra i change point delle serie storiche delle date di iniziovendemmia e quelle delle ore normali di caldo del suddetto periodo (Fig. 2).

Nella collina litoranea meridionale la data di inizio vendemmia è mediamente passata dal 28settembre dell’arco temporale 1974-1991 al 18 settembre del periodo successivo. L’anticipo dellavendemmia è parso connesso con l’aumento delle ore normali di caldo, salite da 1313 del periodo1974-1997 a 1454 del periodo 1998-2009.

Nella collina litoranea centrale la data di inizio vendemmia è stata progressivamente anticipatadal 9 ottobre dell’arco temporale 1974-1982, al 30 settembre del periodo 1983-1991, e al 24settembre degli ultimi anni. In questo areale si è verificato un significativo incremento delle orenormali di caldo che sono salite da 1290 del periodo 1974-1984 a 1415 in quello successivo.

Nella collina interna l’incremento delle ore normali di caldo e l’anticipo della data di iniziovendemmia si sono manifestati solo negli ultimi anni. Le ore normali di caldo sono passate da1257 del periodo 1977-1998 a 1406 del periodo 1999-2009 mentre l’inizio della vendemmia èstato anticipato dal 14 ottobre dell’arco temporale 1977-2002 al 30 settembre degli ultimi anni.

Fig. 2 Analisi del change point applicata alle ore normali di caldo del periodo marzo giugnoe alle date di inizio vendemmia. Le linee tratteggiate verticali indicano i change pointsmentre le linee orizzontali poste in basso in ogni figura indicano l’intervallo di confidenza al90%. Le linee spesse orizzontali rappresentano la media dei periodi.

ConclusioniFin dalla sua fondazione avvenuta nel 1816 ad opera di Alexander von Humboldt (Mariani, 2002),la climatologia si propone di individuare areali o periodi storici climaticamente omogenei , con loscopo finale di fornire strumenti di supporto per le scelte gestionali. A tale approccio si ispira in

particolare la produzione delle normali climatiche di riferimento (CLINO) rappresentative delclima attuale e che si rivelano necessarie in particolare per:

- esprimere in termini quantitativi i livelli di anomalia di una data situazione meteorologica- ottimizzare le scelte varietali e di portainnesto- ottimizzare le tecniche gestionali a livello di vigneto (gestione della chioma e dei rapporti

source-sink, gestione delle risorse idriche e nutrizionali, gestione fitosanitaria, ecc.)- ottimizzare le tecniche gestionali a livello di cantina (programmazione delle attività di

conferimento e di lavorazione delle uve, tecniche enologiche).In questo lavoro è stato evidenziato che il change point climatico individuato negli anni’80 delventesimo secolo ha inaugurato una nuova fase climatica tuttora in corso ed ai CLINOcaratteristici di tale nuova fase dovrebbero di qui in avanti ispirarsi per il territorio abruzzese lescelte sopraelencate.

BibliografiaBai J., Perron P., 1998. Estimating and Testing Linear Models With Multiple Structural

Changes, Econometrica, 66:47-78.Bai J., Perron P., 2003. Computation and Analysis of Multiple Structural Change Models,

Journal of Applied Econometrics, 18:1-22.Jones G.V., Duchene E., Tomasi D., Yuste J., Braslavka O., Schultzh., Martinez C., Boso S.,

Langellier F., Perruchot C., Guimberteau G., 2005. Change in euopean winegrapephenology and relationship with climate. Atti GESCO. Vol.1. Agosto 2005. Geisenheim,Germania. 55-62.

Mariani L., 2002. Dispensa di Agrometeorologia. Milano, CLESAV, 292 pp.Mariani L., 2008. Note scientifiche per un discorso sul clima, Roma, IF,105 pp.Parton W.J., Logan J.A., 1981. A model for diurnal variation in soil and air temperature.

Agric.Meteorol , 23:205-216.Silvestroni O., Di Lena B., Antenucci A., Palliotti A., 2008. Analysis of climatic change in

different areas of Abruzzo Region (Central Italy) implications for grape growing. Atti VIIth

International Terroir Congress. Nyon (Suisse) 19-23/5/2008. 236-241.Todaro C., Migliardi E., 2000. Opinioni sullo studio delle tendenze climatiche, Bollettino

Geofisico, n: 3-4, luglio dicembre 2000.Todaro C., Migliardi E., 2003. Opinioni sullo studio delle tendenze climatiche (parte

seconda), Bollettino Geofisico, 3-4, luglio dicembre 2003.Todaro C., Migliardi E., 2003. Opinioni sullo studio delle tendenze climatiche (reminiscenze

e suggerimenti), Bollettino Geofisico, 3-4, luglio dicembre 2003.Tonietto J., Carbonneau A., 2004. A multicriteria climatic classification system for grape

growing regions worldwide. Agricultural and Forest Meteorology, 124:81-87.Trouet V., Esper J., Graham N.E., Baker A., Scourse J.D., Frank D.C., 2009. Persistent

positive North Atlantic oscillation mode dominated the medieval climate anomaly.Science, 3 april 2009, Vol 324.

Werner, P. C., Gerstengarbe F.W., Fraedrich K, Oesterle K., 2000. Recent climate change inthe North Atlantic/European sector. International Journal of Climatology, Vol. 20, Issue 5,463-471.

3 - 34



(periodo Marzo-Giugno) ha permesso di mantenere alto il coefficiente di determinazione, che èaddirittura aumentato in due areali su tre.

La variabilità temporale delle NHH per il periodo marzo – giugno è stata pertanto indagata alloscopo di individuare eventuali discontinuità attribuibili a cause climatiche. In tutte le zone si èregistrata una buona corrispondenza tra i change point delle serie storiche delle date di iniziovendemmia e quelle delle ore normali di caldo del suddetto periodo (Fig. 2).

Nella collina litoranea meridionale la data di inizio vendemmia è mediamente passata dal 28settembre dell’arco temporale 1974-1991 al 18 settembre del periodo successivo. L’anticipo dellavendemmia è parso connesso con l’aumento delle ore normali di caldo, salite da 1313 del periodo1974-1997 a 1454 del periodo 1998-2009.

Nella collina litoranea centrale la data di inizio vendemmia è stata progressivamente anticipatadal 9 ottobre dell’arco temporale 1974-1982, al 30 settembre del periodo 1983-1991, e al 24settembre degli ultimi anni. In questo areale si è verificato un significativo incremento delle orenormali di caldo che sono salite da 1290 del periodo 1974-1984 a 1415 in quello successivo.

Nella collina interna l’incremento delle ore normali di caldo e l’anticipo della data di iniziovendemmia si sono manifestati solo negli ultimi anni. Le ore normali di caldo sono passate da1257 del periodo 1977-1998 a 1406 del periodo 1999-2009 mentre l’inizio della vendemmia èstato anticipato dal 14 ottobre dell’arco temporale 1977-2002 al 30 settembre degli ultimi anni.

Fig. 2 Analisi del change point applicata alle ore normali di caldo del periodo marzo giugnoe alle date di inizio vendemmia. Le linee tratteggiate verticali indicano i change pointsmentre le linee orizzontali poste in basso in ogni figura indicano l’intervallo di confidenza al90%. Le linee spesse orizzontali rappresentano la media dei periodi.

ConclusioniFin dalla sua fondazione avvenuta nel 1816 ad opera di Alexander von Humboldt (Mariani, 2002),la climatologia si propone di individuare areali o periodi storici climaticamente omogenei , con loscopo finale di fornire strumenti di supporto per le scelte gestionali. A tale approccio si ispira in

particolare la produzione delle normali climatiche di riferimento (CLINO) rappresentative delclima attuale e che si rivelano necessarie in particolare per:

- esprimere in termini quantitativi i livelli di anomalia di una data situazione meteorologica- ottimizzare le scelte varietali e di portainnesto- ottimizzare le tecniche gestionali a livello di vigneto (gestione della chioma e dei rapporti

source-sink, gestione delle risorse idriche e nutrizionali, gestione fitosanitaria, ecc.)- ottimizzare le tecniche gestionali a livello di cantina (programmazione delle attività di

conferimento e di lavorazione delle uve, tecniche enologiche).In questo lavoro è stato evidenziato che il change point climatico individuato negli anni’80 delventesimo secolo ha inaugurato una nuova fase climatica tuttora in corso ed ai CLINOcaratteristici di tale nuova fase dovrebbero di qui in avanti ispirarsi per il territorio abruzzese lescelte sopraelencate.

BibliografiaBai J., Perron P., 1998. Estimating and Testing Linear Models With Multiple Structural

Changes, Econometrica, 66:47-78.Bai J., Perron P., 2003. Computation and Analysis of Multiple Structural Change Models,

Journal of Applied Econometrics, 18:1-22.Jones G.V., Duchene E., Tomasi D., Yuste J., Braslavka O., Schultzh., Martinez C., Boso S.,

Langellier F., Perruchot C., Guimberteau G., 2005. Change in euopean winegrapephenology and relationship with climate. Atti GESCO. Vol.1. Agosto 2005. Geisenheim,Germania. 55-62.

Mariani L., 2002. Dispensa di Agrometeorologia. Milano, CLESAV, 292 pp.Mariani L., 2008. Note scientifiche per un discorso sul clima, Roma, IF,105 pp.Parton W.J., Logan J.A., 1981. A model for diurnal variation in soil and air temperature.

Agric.Meteorol , 23:205-216.Silvestroni O., Di Lena B., Antenucci A., Palliotti A., 2008. Analysis of climatic change in

different areas of Abruzzo Region (Central Italy) implications for grape growing. Atti VIIth

International Terroir Congress. Nyon (Suisse) 19-23/5/2008. 236-241.Todaro C., Migliardi E., 2000. Opinioni sullo studio delle tendenze climatiche, Bollettino

Geofisico, n: 3-4, luglio dicembre 2000.Todaro C., Migliardi E., 2003. Opinioni sullo studio delle tendenze climatiche (parte

seconda), Bollettino Geofisico, 3-4, luglio dicembre 2003.Todaro C., Migliardi E., 2003. Opinioni sullo studio delle tendenze climatiche (reminiscenze

e suggerimenti), Bollettino Geofisico, 3-4, luglio dicembre 2003.Tonietto J., Carbonneau A., 2004. A multicriteria climatic classification system for grape

growing regions worldwide. Agricultural and Forest Meteorology, 124:81-87.Trouet V., Esper J., Graham N.E., Baker A., Scourse J.D., Frank D.C., 2009. Persistent

positive North Atlantic oscillation mode dominated the medieval climate anomaly.Science, 3 april 2009, Vol 324.

Werner, P. C., Gerstengarbe F.W., Fraedrich K, Oesterle K., 2000. Recent climate change inthe North Atlantic/European sector. International Journal of Climatology, Vol. 20, Issue 5,463-471.

3 - 35



APPLIANCE OF CLIMATE PROJECTIONS FOR CLIMATE CHANGE STUDY IN SERBIAN VINEYARD REGIONS

A. Vuković(1,3), M. Vujadinović(1,3), V. Djurdjević(2,3), Z. Ranković-Vasić(1), N. Marković(1),

Z. Atanacković(1), B. Sivčev(1), N. Petrović(1) (1)Faculty of Agriculture, Belgrade University

Nemanjina 6, 11080 Belgrade, Serbia [email protected]

(2)Institute for Meteorology, Faculty of Physics, Belgrade University Dobracina 16, 11000 Belgrade, Serbia

(3)South East European Virtual Climate Change Center (hosted by Republic Hydrmeteorological Service of Serbia) Bulevar oslobodjenja 8, 11000 Belgrade, Serbia

ABSTRACT Climate projections considered here are for two periods in the future throughout two IPCC

scenarios: 2001 – 2030 (A1B) and 2071 – 2100 (A2) obtained using Coupled Regional Climate Model EBU-POM. Results are used in calculation of Heliothermal, Drought and Cool Night Index for climate classification of vineyard regions in Serbia. Presented results show significant change of climate in the future, indicating that varieties of grapevine must be adaptable or vineyard regions should be shifted in other areas with appropriate climate.

KEYWORD climate projections – grapevine – climate classification INTRODUCTION Grape growing and wine production with present knowledge and experience are largely

weather and climate driven. Many studies have been done about climate – grape growing connection and analysis of climate change impact on vineyard regions (Jones et al., 2005; Metzger et al., 2008). This paper is an example of application of long term climate simulations in viticulture in Serbia. Presented results are obtained from the Coupled Regional Climate Model EBU-POM (Djurdjevic, Rajkovic, 2008) for the periods: 1961–1990 (experiment 20c3m), 2001–2030 (scenario A1B), 2071–2100 (scenario A2). The chosen period for climate simulations is thirty years because 30-years mean values should be able to capture 75% of the variance of the true signal according to Huntingford et al. (2003) as well as statistically significant changes in extreme precipitation. Scenario A1B is characterized as a medium sensitivity and A2 as a high sensitivity scenario (close to 850 ppm at the end of 21st century, equivalent to ~2.2 times higher value compared to today’s value of ~385 ppm), in sense of carbon dioxide concentration. Details on concentrations of other Greenhouse Gases (GHGs) prescribed by these scenarios can be found in IPCC Special Report on Emission Scenarios (SRES; Nakicenovic et al., 2000). The SRES scenarios were also used for the IPCC Fourth Assessment Report (Christensen et al, 2007). Climate models have BIAS especially in the area of Pannonian valley where the northern half of the Serbia region is located. This problem is known as Summer Drying Problem in South-Eastern Europe (Hagemann et al., 2004). For calculation of climate indices it was necessary first to make

corrections for model output data. Method used for this purpose is statistical BIAS correction method (Piani et al. 2009). Heliothermal Index (HI), Drought Index (DI) and Cool Night Index (CI) define a Multicriteria Climatic Classification System (Géoviticulture MCC System) for grape growing regions worldwide (Tonnieto, Carbonneau, 2004). Analysis of indices values up to the end of the century show how climate will change in present grape-growing regions, if some other areas will have appropriate climate for grapevine growing and if climate change could lead to necessary substitution of existing varieties or their adaptation to new climate conditions.

MATERIALS AND METHODS Dynamical downscaling of Atmosphere Ocean Global Circulation Model (AOGCM) SX-G

(Gualdi et al., 2003) is done using Coupled Regional Climate Model (CRCM) EBU-POM (Djurdjevic, Rajkovic, 2008; Gualdi et al., 2008). Climate projections have been made for the periods: 1961–1990 (experiment 20c3m), 2001–2030 (A1B), 2071–2100 (A2). Model domain is Europe region. The model resolution is ~30km. In Fig.1 we can see that in the whole area temperature increases, for the first 30 years 2001-2030 about one degree (upper left) and in the last 30 years 2071-2100 more than three degrees (upper right panel). In the first thirty years (lower left) of the century change in precipitation amount is over 50mm/year near shore and in mountain areas. For the last thirty years (lower right) generally the whole model domain is much drier. Decrease in precipitation is more than 100mm/year. All results are shown as difference from model simulation for base period (mean values for 1961-1990). Results are consentient with those obtained for 21 climate global models (IPCC Fourth Assessment Report, Christensen et al, 2007, chapter 11).

When we deduct model results for two periods BIAS is abrogated and it could be assumed that using this approach model error will have significantly less influence. For the purposes of this paper more complex method for model correction was necessary. Applied method is known as statistical BIAS correction method (Dettinger et al., 2004; Piani et al., 2009). Used data are daily observations (maximum, minimum and mean daily air temperature and daily precipitation amount) and model results at 00, 06, 12, 18h UTC for the present climate period. In Fig.2 is presented placement of 17 measurement stations that are near or within viticultural regions in Serbia. It is assumed that temperature data follow Normal (Gaussian) distribution and precipitation data follow Gamma distribution, with special consideration for dry days. In Fig.3 (left panels) we can see that corrected values are very close to the observations, which means that correction functions are well determined and can be used for correction of future climate projections.

Detailed description and classification of climate indices that define Géoviticulture MCC System can be found at Tonnieto and Carbonneau (2004). Here are presented only basic facts that are necessary for further understanding of obtained results. Heliothernal index (HI) is calculated according to Eq.(1)

d

TTTTHI bxb

.09.30

.04.1 2 (1)

where T is daily average and Tx daily maximum temperature, Tb base temperature (10oC) and d is coefficient of the length of the day. Dryness index (DI) represents the value of soil moisture at the end of the growing season under assumption that the initial soil moisture (W0) was 200mm. It was calculated using following equation

3 - 36



APPLIANCE OF CLIMATE PROJECTIONS FOR CLIMATE CHANGE STUDY IN SERBIAN VINEYARD REGIONS

A. Vuković(1,3), M. Vujadinović(1,3), V. Djurdjević(2,3), Z. Ranković-Vasić(1), N. Marković(1),

Z. Atanacković(1), B. Sivčev(1), N. Petrović(1) (1)Faculty of Agriculture, Belgrade University

Nemanjina 6, 11080 Belgrade, Serbia [email protected]

(2)Institute for Meteorology, Faculty of Physics, Belgrade University Dobracina 16, 11000 Belgrade, Serbia

(3)South East European Virtual Climate Change Center (hosted by Republic Hydrmeteorological Service of Serbia) Bulevar oslobodjenja 8, 11000 Belgrade, Serbia

ABSTRACT Climate projections considered here are for two periods in the future throughout two IPCC

scenarios: 2001 – 2030 (A1B) and 2071 – 2100 (A2) obtained using Coupled Regional Climate Model EBU-POM. Results are used in calculation of Heliothermal, Drought and Cool Night Index for climate classification of vineyard regions in Serbia. Presented results show significant change of climate in the future, indicating that varieties of grapevine must be adaptable or vineyard regions should be shifted in other areas with appropriate climate.

KEYWORD climate projections – grapevine – climate classification INTRODUCTION Grape growing and wine production with present knowledge and experience are largely

weather and climate driven. Many studies have been done about climate – grape growing connection and analysis of climate change impact on vineyard regions (Jones et al., 2005; Metzger et al., 2008). This paper is an example of application of long term climate simulations in viticulture in Serbia. Presented results are obtained from the Coupled Regional Climate Model EBU-POM (Djurdjevic, Rajkovic, 2008) for the periods: 1961–1990 (experiment 20c3m), 2001–2030 (scenario A1B), 2071–2100 (scenario A2). The chosen period for climate simulations is thirty years because 30-years mean values should be able to capture 75% of the variance of the true signal according to Huntingford et al. (2003) as well as statistically significant changes in extreme precipitation. Scenario A1B is characterized as a medium sensitivity and A2 as a high sensitivity scenario (close to 850 ppm at the end of 21st century, equivalent to ~2.2 times higher value compared to today’s value of ~385 ppm), in sense of carbon dioxide concentration. Details on concentrations of other Greenhouse Gases (GHGs) prescribed by these scenarios can be found in IPCC Special Report on Emission Scenarios (SRES; Nakicenovic et al., 2000). The SRES scenarios were also used for the IPCC Fourth Assessment Report (Christensen et al, 2007). Climate models have BIAS especially in the area of Pannonian valley where the northern half of the Serbia region is located. This problem is known as Summer Drying Problem in South-Eastern Europe (Hagemann et al., 2004). For calculation of climate indices it was necessary first to make

corrections for model output data. Method used for this purpose is statistical BIAS correction method (Piani et al. 2009). Heliothermal Index (HI), Drought Index (DI) and Cool Night Index (CI) define a Multicriteria Climatic Classification System (Géoviticulture MCC System) for grape growing regions worldwide (Tonnieto, Carbonneau, 2004). Analysis of indices values up to the end of the century show how climate will change in present grape-growing regions, if some other areas will have appropriate climate for grapevine growing and if climate change could lead to necessary substitution of existing varieties or their adaptation to new climate conditions.

MATERIALS AND METHODS Dynamical downscaling of Atmosphere Ocean Global Circulation Model (AOGCM) SX-G

(Gualdi et al., 2003) is done using Coupled Regional Climate Model (CRCM) EBU-POM (Djurdjevic, Rajkovic, 2008; Gualdi et al., 2008). Climate projections have been made for the periods: 1961–1990 (experiment 20c3m), 2001–2030 (A1B), 2071–2100 (A2). Model domain is Europe region. The model resolution is ~30km. In Fig.1 we can see that in the whole area temperature increases, for the first 30 years 2001-2030 about one degree (upper left) and in the last 30 years 2071-2100 more than three degrees (upper right panel). In the first thirty years (lower left) of the century change in precipitation amount is over 50mm/year near shore and in mountain areas. For the last thirty years (lower right) generally the whole model domain is much drier. Decrease in precipitation is more than 100mm/year. All results are shown as difference from model simulation for base period (mean values for 1961-1990). Results are consentient with those obtained for 21 climate global models (IPCC Fourth Assessment Report, Christensen et al, 2007, chapter 11).

When we deduct model results for two periods BIAS is abrogated and it could be assumed that using this approach model error will have significantly less influence. For the purposes of this paper more complex method for model correction was necessary. Applied method is known as statistical BIAS correction method (Dettinger et al., 2004; Piani et al., 2009). Used data are daily observations (maximum, minimum and mean daily air temperature and daily precipitation amount) and model results at 00, 06, 12, 18h UTC for the present climate period. In Fig.2 is presented placement of 17 measurement stations that are near or within viticultural regions in Serbia. It is assumed that temperature data follow Normal (Gaussian) distribution and precipitation data follow Gamma distribution, with special consideration for dry days. In Fig.3 (left panels) we can see that corrected values are very close to the observations, which means that correction functions are well determined and can be used for correction of future climate projections.

Detailed description and classification of climate indices that define Géoviticulture MCC System can be found at Tonnieto and Carbonneau (2004). Here are presented only basic facts that are necessary for further understanding of obtained results. Heliothernal index (HI) is calculated according to Eq.(1)

d

TTTTHI bxb

.09.30

.04.1 2 (1)

where T is daily average and Tx daily maximum temperature, Tb base temperature (10oC) and d is coefficient of the length of the day. Dryness index (DI) represents the value of soil moisture at the end of the growing season under assumption that the initial soil moisture (W0) was 200mm. It was calculated using following equation

3 - 37



.

.

Sep

Aprsto EEPWDI (2)

where the second term in equation represents sum of monthly difference of precipitation (P) and water that is lost through transpiration (Et) and evaporation from bare soil (Es). In order to calculate Et and Es, we need values of potential evapotranspiration that include radiation data. Because these data are not available we used Thornthwaite method (Willmott et al. 1985). Cool night index (CI) equals mean minimum air temperature (Tmin) in the month of ripening and is calculated using equation

.09.30

.09.1min

1T

NCI (3)

Fig1. Difference between mean annual air temperature for the period 2001-2030 (upper left), 2071-2100 (upper right) from value the period 1961-1990 in degrees and the same difference for precipitation (lower panels) in millimeters.

Fig.2 Position of stations and list of vineyard areas in Serbia

Fig.3 Mean original, corrected model and observed values for growing season period for mean daily temperature (upper left), for cumulative precipitation (lower left) for present climate period (1961-1990) and the same parameters obtained with corrected model values (right) for periods 1961-1990 (20c3m), 2001-2030 (A1B) and 2071-2100 (A2).

3 - 38



.

.

Sep

Aprsto EEPWDI (2)

where the second term in equation represents sum of monthly difference of precipitation (P) and water that is lost through transpiration (Et) and evaporation from bare soil (Es). In order to calculate Et and Es, we need values of potential evapotranspiration that include radiation data. Because these data are not available we used Thornthwaite method (Willmott et al. 1985). Cool night index (CI) equals mean minimum air temperature (Tmin) in the month of ripening and is calculated using equation

.09.30

.09.1min

1T

NCI (3)

Fig1. Difference between mean annual air temperature for the period 2001-2030 (upper left), 2071-2100 (upper right) from value the period 1961-1990 in degrees and the same difference for precipitation (lower panels) in millimeters.

Fig.2 Position of stations and list of vineyard areas in Serbia

Fig.3 Mean original, corrected model and observed values for growing season period for mean daily temperature (upper left), for cumulative precipitation (lower left) for present climate period (1961-1990) and the same parameters obtained with corrected model values (right) for periods 1961-1990 (20c3m), 2001-2030 (A1B) and 2071-2100 (A2).

3 - 39



RESULTS AND DISCUSSION Indices are calculated using corrected model values for three periods: present climate period

(1961-1990, experiment 20c3m), period 2001-2030 (A1B scenario) and period 2071-2100 (A2 scenario). In Fig.4 are presented results obtained for HI (left panel). In the first thirty years of the century index values switch one category up, to temperate warm. Zlatibor, mountain area station where grape is not cultivated nowadays stays in very cool category. In the last thirty years of the century results are in warm or very warm category. Zlatibor area enters in temperate climate. Precipitation has more complex dependence on surrounding geo morphology. This contributes to larger variation in categories of DI (Fig.4, right panel). In the first thirty years of the century DI has small decrease. In the last thirty years all vineyard regions enter the moderately dry category, except West Morava region, and west and east border parts of Nisava-South Morava region. Zlatibor stays in humid category. In Fig.5 are presented values for CI. In present climate all vineyard regions are in very cool nights category. In the first thirty years mean annual temperatures are about 1 degree higher, but most regions stay in the same category. In the last thirty years climate change impact is considerately large. Banat region, Timok region and north part of Nisava-South Morava region enter temperate nights category. Largest change is at mountain station Zlatibor. It is known that areas with higher altitudes will be more affected with climate changes.

Fig. 5 Mean Heliothermal Index (left) and Dryness Index (right) values for present climate (1961-1990) obtained from observations and corrected model values (20c3m), for 2001-2030 obtained from corrected model values (A1B) and for 2071-2100 obtained from corrected model values (A2).

Full grape ripening, for varieties in Serbian vineyard regions that have late time of ripening, is assured when the night temperature conditions belong in very cool nights category. These results show that at the end of the century in the most part of the vineyard regions temperatures in ripening period will be to warm for present varieties that have negative impact on color and aroma.

Fig.6 The same as Fig.5 but for Cool Night Index

CONCLUSIONS Serbian vineyard regions show tendency to become warmer and dryer. In the period 2001-2030

changes in air temperature and precipitation do not have large influence on heat and water requirements of grapevine. In the period 2071-2100, according to A2 SRES scenario, grape could be under influence of higher temperatures and less precipitation during its development. This could lead to early ripening stage, after which high temperatures could disable plants to prepare for the dormant season. Late varieties will be affected with warmer temperatures in ripening period, which will have negative effect on skin color and aroma of grape and wine. Areas on higher altitudes will have more appropriate climate for present varieties and this could lead to relocating vineyards to higher areas.

BIBLIOGRAPHY

Christensen J.H. et al. 2007: Regional Climate Projections. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA

Dettinger M.D., Cayan, D.R., Meyer, M.K., and Jeton, A.E., 2004: Simulated hydrologic responses to climate variations and change in the Merced, Carson, and American River basins, Sierra Nevada, California, 1900-2099: Climatic Change, v. 62, n: 283-317.

Djurdjevic V., Rajkovic B., 2008: Verification of a coupled atmosphere-ocean model using satellite observations over the Adriatic Sea., Ann. Geophys., 26, n: 1935-1954.

Gualdi, S., Navarra A., Guilyardi E., Delecluse P. 2003: Assessment of the tropical Indo-Pacific climate in the SINTEX CGCM, Ann. Geophys., 46, n:1-26.

Gualdi S., Rajkovic B., Djurdjevic V., Castellari S., Scoccimarro E., Navarra A., Dacic M., 2008: SImulations of climate chaNge in the mediTerranean Area (SINTA), Final Scientific report

Hagemann, S., Machenhauer B., Jones R., Christensen O.B., Deque M., Jacob D., Vidale P.L., 2004: Evaluation of water and energy budgets in regional climate models applied over Europe, Climate Dynamics, 23, n: 547-567.

Huntingford, C., Jones R.G., Prudhomme C., Lamb R., Gash J.H.C, 2003: Regional climate model predictions of extreme rainfall for a changing climate, Q. J. R. Meteorol. Soc. 129, n:1607-1621.

Jones G.V., White M.A., Cooper O.R., Storchmann K., 2005: Climate Change and Global Wine Quality, Climatic Change, 73(3), n: 319-343.

Metzger M.J., Bunce R.G.H., Leemans R., Viner D. 2008: Projected environmental shifts under climate change: Europen trends and regional impacts, Environmental Conservation 35 (1), n:64-75.

Nakicenovic N. et al., 2000: Special Report on Emissions Scenarios, Contribution to the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, UK.

Piani C., Haerter J.O., Coppola E., 2010: Statistical bias correction for daily precipitation in regional climate models over Europe, Theor Appl Climatol. 99, n: 187–192.

Tonietto, J., Carbonneau A. 2004: A multicriteria climatic classification system for grape-growing regions worldwide, Agricultural and Forest Meteorology 124, n:81-97.

Willmott C.J., Rowe C.M., Mintz Y. 1985: Climatology of the Terrestrial Seasonal Water Cycle, Journal of Climatology, 5, n: 589-606.

3 - 40



RESULTS AND DISCUSSION Indices are calculated using corrected model values for three periods: present climate period

(1961-1990, experiment 20c3m), period 2001-2030 (A1B scenario) and period 2071-2100 (A2 scenario). In Fig.4 are presented results obtained for HI (left panel). In the first thirty years of the century index values switch one category up, to temperate warm. Zlatibor, mountain area station where grape is not cultivated nowadays stays in very cool category. In the last thirty years of the century results are in warm or very warm category. Zlatibor area enters in temperate climate. Precipitation has more complex dependence on surrounding geo morphology. This contributes to larger variation in categories of DI (Fig.4, right panel). In the first thirty years of the century DI has small decrease. In the last thirty years all vineyard regions enter the moderately dry category, except West Morava region, and west and east border parts of Nisava-South Morava region. Zlatibor stays in humid category. In Fig.5 are presented values for CI. In present climate all vineyard regions are in very cool nights category. In the first thirty years mean annual temperatures are about 1 degree higher, but most regions stay in the same category. In the last thirty years climate change impact is considerately large. Banat region, Timok region and north part of Nisava-South Morava region enter temperate nights category. Largest change is at mountain station Zlatibor. It is known that areas with higher altitudes will be more affected with climate changes.

Fig. 5 Mean Heliothermal Index (left) and Dryness Index (right) values for present climate (1961-1990) obtained from observations and corrected model values (20c3m), for 2001-2030 obtained from corrected model values (A1B) and for 2071-2100 obtained from corrected model values (A2).

Full grape ripening, for varieties in Serbian vineyard regions that have late time of ripening, is assured when the night temperature conditions belong in very cool nights category. These results show that at the end of the century in the most part of the vineyard regions temperatures in ripening period will be to warm for present varieties that have negative impact on color and aroma.

Fig.6 The same as Fig.5 but for Cool Night Index

CONCLUSIONS Serbian vineyard regions show tendency to become warmer and dryer. In the period 2001-2030

changes in air temperature and precipitation do not have large influence on heat and water requirements of grapevine. In the period 2071-2100, according to A2 SRES scenario, grape could be under influence of higher temperatures and less precipitation during its development. This could lead to early ripening stage, after which high temperatures could disable plants to prepare for the dormant season. Late varieties will be affected with warmer temperatures in ripening period, which will have negative effect on skin color and aroma of grape and wine. Areas on higher altitudes will have more appropriate climate for present varieties and this could lead to relocating vineyards to higher areas.

BIBLIOGRAPHY

Christensen J.H. et al. 2007: Regional Climate Projections. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA

Dettinger M.D., Cayan, D.R., Meyer, M.K., and Jeton, A.E., 2004: Simulated hydrologic responses to climate variations and change in the Merced, Carson, and American River basins, Sierra Nevada, California, 1900-2099: Climatic Change, v. 62, n: 283-317.

Djurdjevic V., Rajkovic B., 2008: Verification of a coupled atmosphere-ocean model using satellite observations over the Adriatic Sea., Ann. Geophys., 26, n: 1935-1954.

Gualdi, S., Navarra A., Guilyardi E., Delecluse P. 2003: Assessment of the tropical Indo-Pacific climate in the SINTEX CGCM, Ann. Geophys., 46, n:1-26.

Gualdi S., Rajkovic B., Djurdjevic V., Castellari S., Scoccimarro E., Navarra A., Dacic M., 2008: SImulations of climate chaNge in the mediTerranean Area (SINTA), Final Scientific report

Hagemann, S., Machenhauer B., Jones R., Christensen O.B., Deque M., Jacob D., Vidale P.L., 2004: Evaluation of water and energy budgets in regional climate models applied over Europe, Climate Dynamics, 23, n: 547-567.

Huntingford, C., Jones R.G., Prudhomme C., Lamb R., Gash J.H.C, 2003: Regional climate model predictions of extreme rainfall for a changing climate, Q. J. R. Meteorol. Soc. 129, n:1607-1621.

Jones G.V., White M.A., Cooper O.R., Storchmann K., 2005: Climate Change and Global Wine Quality, Climatic Change, 73(3), n: 319-343.

Metzger M.J., Bunce R.G.H., Leemans R., Viner D. 2008: Projected environmental shifts under climate change: Europen trends and regional impacts, Environmental Conservation 35 (1), n:64-75.

Nakicenovic N. et al., 2000: Special Report on Emissions Scenarios, Contribution to the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, UK.

Piani C., Haerter J.O., Coppola E., 2010: Statistical bias correction for daily precipitation in regional climate models over Europe, Theor Appl Climatol. 99, n: 187–192.

Tonietto, J., Carbonneau A. 2004: A multicriteria climatic classification system for grape-growing regions worldwide, Agricultural and Forest Meteorology 124, n:81-97.

Willmott C.J., Rowe C.M., Mintz Y. 1985: Climatology of the Terrestrial Seasonal Water Cycle, Journal of Climatology, 5, n: 589-606.

3 - 41



1

CLIMATE EFFECT ON RIPENING PROCESS IN Vitis vinifera, L. Cv. CENCIBEL J.A. Amorós Ortiz-Villajos(1), F. Muñoz de Cuerva(2), C. Pérez de los Reyes(1), F.J. García Navarro(1)

and J.A. Campos Gallego(1)

(1) Esc. Univ. Ing. Tecn. Agrícola. UCLM. Ronda de Calatrava, 7. 13071 Ciudad Real. Spain [email protected] (2) Bodegas Naranjo, S.L.,C/ Felipe II. Carrión de Calatrava, Spain.

ABSTRACT A seven years survey (2003 to 2009) has been carried out over old traditional vineyards cv. Cencibel in

La Mancha region (Spain). Seven plots with more than 35 years old were sampled from veraison to harvest, measuring soluble solids (ºBaumé) and acid concentration (g/l in tartaric acid). The ripening process was different each year depending on season climate character (vintage). The monthly mean temperatures (April to September) and the rainfalls accumulated (April to September) have been studied and these factors have been related with the date of vintage and the colour intensity (very important parameter for wine quality). The growing-degree day (GDD) for the variety Cencibel (1551,1ºC) has been calculated. The temperature of May is critical for the development of photosynthetic apparatus of the vineyard and thus, conditions all the ripening process. It has been found two different models of vintage: mild-fresh year (2004, 2007 and 2008) and warm year (2003, 2005, 2006 and 2009). In the warm conditions of La Mancha it is very desirable a delay in the ripening process. As the later will be the process, the cooler will be the nights at the end of ripening. This will improve the quality of the vintage, as it happened in the fresh years.

KEYWORDS

Vintage – ripeness –growing degree day - harvest. INTRODUCTION

The concept of "vintage" is widely spread in the majority of wine-growing regions in the world. The mentioned concept, used since long time ago, reveals that the annual climate has a direct influence on the quality of the wines when the rest of the factors (soil, plant and cultural practices) are equal. These factors constitute the wine agricultural ecosystem ("Terroir") (Van Leewen et al., 2004). The temperature and the rainfall have been the main climatic factors studied traditionally to define vintage (Mesoclimate). Other factors like the sun irradiation, winds, relative humidity and cultural practices have importance as well, but only for microclimatic level and they do not concern to extensive areas (Jackson, Lombard, 1993; Bergquist, et al., 2001; Spayd, et al., 2002). There are many climatic parameters used in viticulture (Winkler, 1965; Hidalgo, 2002) that characterize the suitability of the climate for the quality grapes production and, therefore, for a good ripeness. In some regions, with hot climate, the supposition that the vintage has little importance has been accepted because, except in serious accidents or catastrophes, grapes complete its ripeness without problems year after year. The aim of this work is to verify the effect of climate on the vintage in hot regions (Western of La Mancha, Spain) in an early variety as Cencibel (red variety for quality wines called Tempranillo in La Rioja). In addition, simple parameters, that could predict the future vintage date and quality, will be determined.

2

MATERIALS AND METHODS The present study was carried out on traditional vineyards of Cencibel from 2003 to 2009. They were all dry farming plantations, more than 35 years old, planted 2,5m x 2,5m (between rows and plants) and head-pruned (from 4 to 7 spurs with 2 buds). Their productions were relatively low, from 1.5 to 3 kg/vine. All the vineyards bring the production to Bodegas Naranjo S.A. (Carrión de Calatrava) and constituted the base to elaborate the higher quality red wines, very estimated by the customs of the cellar. The ripening process was followed in 6 to 10 plots (depending on the year) distant less than 10 km from the cellar, over different soils. Since the veraison, chunks of clusters of every control vine, picked from different parts and from different orientations of the vine and the cluster were collected to complete approximately 200 grape grains. Every sample from every plot was kept in an independent bag and labeled with the date and the plot weekly. In every sample the following analytical determinations were carried out: Content in soluble solids (sugars) from the musts obtained from the samples of grapes was measured using a refractometer with compensation of temperature (model Zuzi-50-305000). The expression of the result was in "ºBe", by direct reading in the scale of the device. Total Acidity was measured by acid - base valuation and it was expressed in g/l equivalent in tartaric acid. Colour Intensity: Addition of optical density (O.D.) at 420 nm in tray with 1cm of thickness, O. D. at 520 nm and O.D. at 620 nm in the final wines elaborated every year. Meteorological data were obtained from the SIAR (the Irrigation Advisory Service for Farmers) and later they were confirmed by those of the National Institute of Meteorology in Spain (Meteorological Station of Ciudad Real: Altitude 627m; Latitude 38 º 59 ' 22 "N; Length 03 º 55 ' 11 "W).

RESULTS AND DISCUSSION CLIMATIC CHARACTERIZATION Tab. 1 shows the mean temperatures for every month from April to September (2003-2009) and the rainfall accumulated from April to September. It can be inferred that the studied area presents a climate with hot summer because ripeness temperatures are, in all cases, over 20ºC, zone B (Jackson, Lombard, 1993). The temperature of May is critical for the development of photosynthetic apparatus of the vineyard, conditioning all the ripening process. Watching the information of rainfall exposed in the Tab.1 a low and irregular rainfall can be noted from year to year. The rainfall during the vegetative period is, in average, 150 mm per year and mainly occurs in spring (April to June). The rainfall during the growing season presents important differences between 2007-2008, and 2003, 2005, 2006 and 2009. 2004 presented a low rainfall with low mean temperature. The rainfall in the most humid years (2007, 2008) had influence on the development of the mildiu (Plasmopara vitícola Berl. et De Toni) reducing the harvest. Tab.1. Mean temperatures (ºC) for every month from April to September (2003-2009) and the rainfall (R)

accumulated (mm) from April to September each year. April May June July August September R (mm)

2003 11,79 17,66 24,91 25,53 25,17 20,58 125,60 2004 10,61 13,46 23,13 24,97 23,69 20,69 74,20 2005 12,77 18,03 23,90 25,75 24,71 19,23 93,00 2006 13,56 18,94 22,70 26,78 24,60 20,55 125,80 2007 10,70 15,00 20,02 24,46 23,81 20,06 252,20 2008 11,90 14,38 21,01 24,02 24,40 18,23 240,00 2009 10,72 17,75 22,91 25,21 25,44 15.30 140,80

MEAN 11,72 16,46 22,65 25,25 24,55 23,48 150,23

3 - 42



1

CLIMATE EFFECT ON RIPENING PROCESS IN Vitis vinifera, L. Cv. CENCIBEL J.A. Amorós Ortiz-Villajos(1), F. Muñoz de Cuerva(2), C. Pérez de los Reyes(1), F.J. García Navarro(1)

and J.A. Campos Gallego(1)

(1) Esc. Univ. Ing. Tecn. Agrícola. UCLM. Ronda de Calatrava, 7. 13071 Ciudad Real. Spain [email protected] (2) Bodegas Naranjo, S.L.,C/ Felipe II. Carrión de Calatrava, Spain.

ABSTRACT A seven years survey (2003 to 2009) has been carried out over old traditional vineyards cv. Cencibel in

La Mancha region (Spain). Seven plots with more than 35 years old were sampled from veraison to harvest, measuring soluble solids (ºBaumé) and acid concentration (g/l in tartaric acid). The ripening process was different each year depending on season climate character (vintage). The monthly mean temperatures (April to September) and the rainfalls accumulated (April to September) have been studied and these factors have been related with the date of vintage and the colour intensity (very important parameter for wine quality). The growing-degree day (GDD) for the variety Cencibel (1551,1ºC) has been calculated. The temperature of May is critical for the development of photosynthetic apparatus of the vineyard and thus, conditions all the ripening process. It has been found two different models of vintage: mild-fresh year (2004, 2007 and 2008) and warm year (2003, 2005, 2006 and 2009). In the warm conditions of La Mancha it is very desirable a delay in the ripening process. As the later will be the process, the cooler will be the nights at the end of ripening. This will improve the quality of the vintage, as it happened in the fresh years.

KEYWORDS

Vintage – ripeness –growing degree day - harvest. INTRODUCTION

The concept of "vintage" is widely spread in the majority of wine-growing regions in the world. The mentioned concept, used since long time ago, reveals that the annual climate has a direct influence on the quality of the wines when the rest of the factors (soil, plant and cultural practices) are equal. These factors constitute the wine agricultural ecosystem ("Terroir") (Van Leewen et al., 2004). The temperature and the rainfall have been the main climatic factors studied traditionally to define vintage (Mesoclimate). Other factors like the sun irradiation, winds, relative humidity and cultural practices have importance as well, but only for microclimatic level and they do not concern to extensive areas (Jackson, Lombard, 1993; Bergquist, et al., 2001; Spayd, et al., 2002). There are many climatic parameters used in viticulture (Winkler, 1965; Hidalgo, 2002) that characterize the suitability of the climate for the quality grapes production and, therefore, for a good ripeness. In some regions, with hot climate, the supposition that the vintage has little importance has been accepted because, except in serious accidents or catastrophes, grapes complete its ripeness without problems year after year. The aim of this work is to verify the effect of climate on the vintage in hot regions (Western of La Mancha, Spain) in an early variety as Cencibel (red variety for quality wines called Tempranillo in La Rioja). In addition, simple parameters, that could predict the future vintage date and quality, will be determined.

2

MATERIALS AND METHODS The present study was carried out on traditional vineyards of Cencibel from 2003 to 2009. They were all dry farming plantations, more than 35 years old, planted 2,5m x 2,5m (between rows and plants) and head-pruned (from 4 to 7 spurs with 2 buds). Their productions were relatively low, from 1.5 to 3 kg/vine. All the vineyards bring the production to Bodegas Naranjo S.A. (Carrión de Calatrava) and constituted the base to elaborate the higher quality red wines, very estimated by the customs of the cellar. The ripening process was followed in 6 to 10 plots (depending on the year) distant less than 10 km from the cellar, over different soils. Since the veraison, chunks of clusters of every control vine, picked from different parts and from different orientations of the vine and the cluster were collected to complete approximately 200 grape grains. Every sample from every plot was kept in an independent bag and labeled with the date and the plot weekly. In every sample the following analytical determinations were carried out: Content in soluble solids (sugars) from the musts obtained from the samples of grapes was measured using a refractometer with compensation of temperature (model Zuzi-50-305000). The expression of the result was in "ºBe", by direct reading in the scale of the device. Total Acidity was measured by acid - base valuation and it was expressed in g/l equivalent in tartaric acid. Colour Intensity: Addition of optical density (O.D.) at 420 nm in tray with 1cm of thickness, O. D. at 520 nm and O.D. at 620 nm in the final wines elaborated every year. Meteorological data were obtained from the SIAR (the Irrigation Advisory Service for Farmers) and later they were confirmed by those of the National Institute of Meteorology in Spain (Meteorological Station of Ciudad Real: Altitude 627m; Latitude 38 º 59 ' 22 "N; Length 03 º 55 ' 11 "W).

RESULTS AND DISCUSSION CLIMATIC CHARACTERIZATION Tab. 1 shows the mean temperatures for every month from April to September (2003-2009) and the rainfall accumulated from April to September. It can be inferred that the studied area presents a climate with hot summer because ripeness temperatures are, in all cases, over 20ºC, zone B (Jackson, Lombard, 1993). The temperature of May is critical for the development of photosynthetic apparatus of the vineyard, conditioning all the ripening process. Watching the information of rainfall exposed in the Tab.1 a low and irregular rainfall can be noted from year to year. The rainfall during the vegetative period is, in average, 150 mm per year and mainly occurs in spring (April to June). The rainfall during the growing season presents important differences between 2007-2008, and 2003, 2005, 2006 and 2009. 2004 presented a low rainfall with low mean temperature. The rainfall in the most humid years (2007, 2008) had influence on the development of the mildiu (Plasmopara vitícola Berl. et De Toni) reducing the harvest. Tab.1. Mean temperatures (ºC) for every month from April to September (2003-2009) and the rainfall (R)

accumulated (mm) from April to September each year. April May June July August September R (mm)

2003 11,79 17,66 24,91 25,53 25,17 20,58 125,60 2004 10,61 13,46 23,13 24,97 23,69 20,69 74,20 2005 12,77 18,03 23,90 25,75 24,71 19,23 93,00 2006 13,56 18,94 22,70 26,78 24,60 20,55 125,80 2007 10,70 15,00 20,02 24,46 23,81 20,06 252,20 2008 11,90 14,38 21,01 24,02 24,40 18,23 240,00 2009 10,72 17,75 22,91 25,21 25,44 15.30 140,80

MEAN 11,72 16,46 22,65 25,25 24,55 23,48 150,23

3 - 43



3

Growing Degree Day over 10ºC (GDD, Winkler, 1965) from April to August and from April until September have been calculated and the results appear in the Tab. 2.

Tab.2. Growing Degree Day over 10ºC (ºC) from April to August and from April to September, date of harvest and Colour Intensity (C.I.).

GDD(IV-VIII) GDD(IV-IX) Date of harvest C.I. 2003 1685,2 2000,2 22/08/2003 15,4 2004 1403,0 1724,0 02/09/2004 24,7 2005 1688,4 1964,4 24/08/2005 19,2 2006 1735,3 2050,3 24/08/2006 14,5 2007 1350,2 1650,2 05/09/2007 19,4 2008 1400,3 1646,3 09/09/2008 19,1 2009 1595,3 1754,3 27/08/2009 14,4

The mean GDD (April - September) of seven studied years is 1826.8ºC, zone III or IV Winkler (Hidalgo, 2002). The GDD is higher than the one needed for ripeness in Merlot variety (1693 º), similar in cycle to Cencibel. It can be seen in the Tab.2 that the vintage has been carried out at the end of August. So, the GDD of the Cencibel (April- August) in the studied area is 1551,1ºC. This is consistent with data obtained for the years 1987 to 1991 in nearby locations as Tomelloso (Jiménez, 1993). In the years 2003, 2005, 2006 and 2008 the GDD (April- August) overcome 1551,1ºC (warm years). However, the years 2004, 2007 and 2008 do not reach this amount and the ripeness delays on the first days of September (mild-fresh years). The warm years are correlated with dry springs and the mild-fresh years with the rainy springs (except in 2004). The spring rains, in the studied zone, produce a significant fall of the temperatures associated with overcast skies, less sunlight and more air humidity. Water stress in 2005 was so hard because a very dry spring was preceded for a dry winter (water deficit marked). In 2005 the growth was already being affected from the budbreak, leading to a very early veraison (on the 15th of July) and a long ripeness period with high temperatures. This deficit can explain that 2005 is the warm year with the highest Colour Intensity (Tab. 2). An over-ripening has been produced with a stress water situation, giving high content in polyphenols and in C.I., although the quality of the wine in 2005 is not considered excellent by the cellar. To sum up, the years 2004, 2007 and 2008 have been characterized as mild-fresh years, especially in spring, with moderate temperatures. They can be qualified as not hot and humid years, with late and uniform ripeness, very suitable for red quality wines. On the opposite, the years 2003, 2005, 2006 and 2009 can be noted as hot and dry year model. The whole vegetative cycle of the grapevine was accelerated due to the high temperatures: budbreak, leaf development, veraison and ripeness. Moreover, the ripeness took place during the hottest days of August, so the harvest was delayed respect to the ripeness. It is expected lower contents in coloured compounds (Bergquist, et al., 2001; Spayd, et al., 2002). CONSEQUENCES ON THE RIPENESS Traditionally, the ripening process begins with the veraison and ends with the harvest. Technological Ripeness is defined as the moment in which the quantity of sugars and organic acids accumulated allows to make the optimum wine. It changes according to the climatic characteristics of the year (Jackson, Lombard, 1993) and to the type of wine. In this case, for traditional vineyards of Cencibel, the grapes might be harvested when overcomes 12.5 º Baumé. To determine the technological ripeness two parameters are used: Solid Soluble (º Baume) and Total Acidity (g/l equivalent in tartaric acid). In the Fig. 1 appears the evolution of the above mentioned parameters in the years 2008 and 2009 respectively. The information of all the plots in the years of the

4

study there has been collected and an average of the ripeness of the variety Cencibel in the studied area has been calculated. 2009 represents the model of ripeness in warm year and 2008 is the model of ripeness in mild-fresh year (only these years have been represented in order to clarify the graph). It is interesting for the quality of the vintage the evolution of total polyphenols. Direct information of the above mentioned compounds is not available for every year, but the indirect information is reported by Colour Intensity (C.I.) measured in final wines in Tab.2.

It can be observed in the Fig.1, that in 2009 (model of warm year) the veraison was early (on the 28th of July) and the whole process of ripening was produced in the first fortnight of August (high mean temperatures and, therefore, very warm nights). The harvest (very early) took place later than it was recommended, for what over-ripeness and a fall of the acidity was produced. Nevertheless, in 2008 (model of mild-fresh year) the veraison was late (on the 8th of August) and the process of ripeness took place in the second fortnight of August with lower mean temperatures and night temperatures below 20ºC. The harvest was carried out coinciding with the technological ripeness, obtaining more balanced musts. This allowed better phenolic ripeness, giving wines with more intensity in colour and better quality.

Ripeness Evolution

0,00,51,01,52,02,53,03,54,04,55,05,56,06,57,07,58,08,59,09,5

10,010,511,011,512,012,513,013,514,014,515,015,516,0

28-J

ul-1

0

29-J

ul-1

0

30-J

ul-1

0

31-J

ul-1

0

01-A

ug-1

0

02-A

ug-1

0

03-A

ug-1

0

04-A

ug-1

0

05-A

ug-1

0

06-A

ug-1

0

07-A

ug-1

0

08-A

ug-1

0

09-A

ug-1

0

10-A

ug-1

0

11-A

ug-1

0

12-A

ug-1

0

13-A

ug-1

0

14-A

ug-1

0

15-A

ug-1

0

16-A

ug-1

0

17-A

ug-1

0

18-A

ug-1

0

19-A

ug-1

0

20-A

ug-1

0

21-A

ug-1

0

22-A

ug-1

0

23-A

ug-1

0

24-A

ug-1

0

25-A

ug-1

0

26-A

ug-1

0

27-A

ug-1

0

28-A

ug-1

0

29-A

ug-1

0

30-A

ug-1

0

31-A

ug-1

0

01-S

ep-1

0

02-S

ep-1

0

03-S

ep-1

0

Tot. Ac 09 ºBé 09 Tot. Ac. 08 ºBé 08

Fig. 1.Evolution of the Total Acidity (g/l equivalent in tartaric acid) and Solid Soluble (ºBaume) for the

year 2008 (discontinuous line) and 2009 (continues line) in vineyards of Cencibel in Carrión de Calatrava (Ciudad Real, Spain).

3 - 44



3

Growing Degree Day over 10ºC (GDD, Winkler, 1965) from April to August and from April until September have been calculated and the results appear in the Tab. 2.

Tab.2. Growing Degree Day over 10ºC (ºC) from April to August and from April to September, date of harvest and Colour Intensity (C.I.).

GDD(IV-VIII) GDD(IV-IX) Date of harvest C.I. 2003 1685,2 2000,2 22/08/2003 15,4 2004 1403,0 1724,0 02/09/2004 24,7 2005 1688,4 1964,4 24/08/2005 19,2 2006 1735,3 2050,3 24/08/2006 14,5 2007 1350,2 1650,2 05/09/2007 19,4 2008 1400,3 1646,3 09/09/2008 19,1 2009 1595,3 1754,3 27/08/2009 14,4

The mean GDD (April - September) of seven studied years is 1826.8ºC, zone III or IV Winkler (Hidalgo, 2002). The GDD is higher than the one needed for ripeness in Merlot variety (1693 º), similar in cycle to Cencibel. It can be seen in the Tab.2 that the vintage has been carried out at the end of August. So, the GDD of the Cencibel (April- August) in the studied area is 1551,1ºC. This is consistent with data obtained for the years 1987 to 1991 in nearby locations as Tomelloso (Jiménez, 1993). In the years 2003, 2005, 2006 and 2008 the GDD (April- August) overcome 1551,1ºC (warm years). However, the years 2004, 2007 and 2008 do not reach this amount and the ripeness delays on the first days of September (mild-fresh years). The warm years are correlated with dry springs and the mild-fresh years with the rainy springs (except in 2004). The spring rains, in the studied zone, produce a significant fall of the temperatures associated with overcast skies, less sunlight and more air humidity. Water stress in 2005 was so hard because a very dry spring was preceded for a dry winter (water deficit marked). In 2005 the growth was already being affected from the budbreak, leading to a very early veraison (on the 15th of July) and a long ripeness period with high temperatures. This deficit can explain that 2005 is the warm year with the highest Colour Intensity (Tab. 2). An over-ripening has been produced with a stress water situation, giving high content in polyphenols and in C.I., although the quality of the wine in 2005 is not considered excellent by the cellar. To sum up, the years 2004, 2007 and 2008 have been characterized as mild-fresh years, especially in spring, with moderate temperatures. They can be qualified as not hot and humid years, with late and uniform ripeness, very suitable for red quality wines. On the opposite, the years 2003, 2005, 2006 and 2009 can be noted as hot and dry year model. The whole vegetative cycle of the grapevine was accelerated due to the high temperatures: budbreak, leaf development, veraison and ripeness. Moreover, the ripeness took place during the hottest days of August, so the harvest was delayed respect to the ripeness. It is expected lower contents in coloured compounds (Bergquist, et al., 2001; Spayd, et al., 2002). CONSEQUENCES ON THE RIPENESS Traditionally, the ripening process begins with the veraison and ends with the harvest. Technological Ripeness is defined as the moment in which the quantity of sugars and organic acids accumulated allows to make the optimum wine. It changes according to the climatic characteristics of the year (Jackson, Lombard, 1993) and to the type of wine. In this case, for traditional vineyards of Cencibel, the grapes might be harvested when overcomes 12.5 º Baumé. To determine the technological ripeness two parameters are used: Solid Soluble (º Baume) and Total Acidity (g/l equivalent in tartaric acid). In the Fig. 1 appears the evolution of the above mentioned parameters in the years 2008 and 2009 respectively. The information of all the plots in the years of the

4

study there has been collected and an average of the ripeness of the variety Cencibel in the studied area has been calculated. 2009 represents the model of ripeness in warm year and 2008 is the model of ripeness in mild-fresh year (only these years have been represented in order to clarify the graph). It is interesting for the quality of the vintage the evolution of total polyphenols. Direct information of the above mentioned compounds is not available for every year, but the indirect information is reported by Colour Intensity (C.I.) measured in final wines in Tab.2.

It can be observed in the Fig.1, that in 2009 (model of warm year) the veraison was early (on the 28th of July) and the whole process of ripening was produced in the first fortnight of August (high mean temperatures and, therefore, very warm nights). The harvest (very early) took place later than it was recommended, for what over-ripeness and a fall of the acidity was produced. Nevertheless, in 2008 (model of mild-fresh year) the veraison was late (on the 8th of August) and the process of ripeness took place in the second fortnight of August with lower mean temperatures and night temperatures below 20ºC. The harvest was carried out coinciding with the technological ripeness, obtaining more balanced musts. This allowed better phenolic ripeness, giving wines with more intensity in colour and better quality.

Ripeness Evolution

0,00,51,01,52,02,53,03,54,04,55,05,56,06,57,07,58,08,59,09,5

10,010,511,011,512,012,513,013,514,014,515,015,516,0

28-J

ul-1

0

29-J

ul-1

0

30-J

ul-1

0

31-J

ul-1

0

01-A

ug-1

0

02-A

ug-1

0

03-A

ug-1

0

04-A

ug-1

0

05-A

ug-1

0

06-A

ug-1

0

07-A

ug-1

0

08-A

ug-1

0

09-A

ug-1

0

10-A

ug-1

0

11-A

ug-1

0

12-A

ug-1

0

13-A

ug-1

0

14-A

ug-1

0

15-A

ug-1

0

16-A

ug-1

0

17-A

ug-1

0

18-A

ug-1

0

19-A

ug-1

0

20-A

ug-1

0

21-A

ug-1

0

22-A

ug-1

0

23-A

ug-1

0

24-A

ug-1

0

25-A

ug-1

0

26-A

ug-1

0

27-A

ug-1

0

28-A

ug-1

0

29-A

ug-1

0

30-A

ug-1

0

31-A

ug-1

0

01-S

ep-1

0

02-S

ep-1

0

03-S

ep-1

0

Tot. Ac 09 ºBé 09 Tot. Ac. 08 ºBé 08

Fig. 1.Evolution of the Total Acidity (g/l equivalent in tartaric acid) and Solid Soluble (ºBaume) for the

year 2008 (discontinuous line) and 2009 (continues line) in vineyards of Cencibel in Carrión de Calatrava (Ciudad Real, Spain).

3 - 45



5

CONCLUSIONS In the studied years, two models of ripeness have been observed:

- Early technological ripeness. In these years (2003, 2005, 2006 and 2009), the veraison has been early and the process of ripeness has developed in the first fortnight of August. The harvest has been early too, the colour intensity of the wines has been low (except in 2005) and the quality of the wines has been medium. They correspond with a warm year model (high mean temperatures, high Growing Degree Day and low spring rainfalls).

- Late technological ripeness. In these years (2004, 2007 and 2008), the process of ripeness has been carried out in the second fortnight of August and the first days of September. The harvest has been coinciding with the technological ripeness, obtaining more balanced musts. The colour intensity of the wines has been high and the quality of the wines has been very good. They correspond with a model mild-fresh year (moderate mean temperatures, low Growing Degree Day and high spring rainfalls).

The study of the environmental parameters (mean temperature and rainfalls from April to September) can be useful in the prediction of the vintage model. May monthly mean temperature seems to be the key for the development of the photosynthetic apparatus of the grapevine and, therefore, it determines the whole vegetative cycle in the studied years: A fresh May (Tm <15ºC, indicated in grey in the Tab. 1) delays the growth of the buds and the whole cycle. On the other hand, a hot May (Tm> 15ºC) supposes a rapid growth during this month, developing the photosynthetic apparatus that is not stopped in the following months that are always hot and leads to an early ripeness with high temperatures.

ACKNOWLEDGMENTS The present work has been elaborated in collaboration with Bodegas Naranjo S.L. that has provided the

facilities and the staff of the cellar. Special gratefulness for Mrs Pilar Almansa and Mr Ramon Muñoz de Cuerva.

BIBLIOGRAPHY

Bergquist, J., Dookozlian, N., Ebisuda, N., 2001. Sunlight Exposure and Temperature Effects on Berry Growth and Composition of Cabernet Sauvignon and Grenache in the Central San Joaquin Valley of California. Am.J. Enol. Vitic. 52(1): 1-7. Hidalgo, L., 2002.Tratado de Viticultura General. 3rd edition, Madrid: Mundiprensa. Jackson, D., Lombard, P., 1993. Enviromental and Management Practices Affecting Grape Composition and Wine Quality- A Review. Am. J. Enol. Vitic., 44( 4):409-430. Jiménez, J., 1993. Adaptación de 10 cultivares tintos de vid (Vitis vinífera L.) a la región de La Mancha. Doctoral Thesis. Univ. Politéc. Madrid. Dept. Fitotécnia y Producción Vegetal. 205 págs. Spayd, S., Tarara, J., Mee, D., Ferguson, J., 2002. Separation of Sunlight and temperature Effects on the Composition of Vitis vinifera cv. Merlot Berries. Am.J. Enol. Vitic, 53(3): 171-182. Van Leeuwen, C., Friant, P., Choné, X., Tregoat, O., Koundouras, S., Dubordieu, D., 2004. Influence of Climate, Soil and Cultivar on Terroir. Am J. Enol. Vitic., 55(3): 207-217. Winkler, A.J., 1965. Viticultura. 1st edition, Mexico:Compañia Editorial Continental.

THE INFLUENCE OF CLIMATE ON THE GRAPEVINEPHENOLOGY AND CONTENT OF SUGAR AND TOTAL ACIDS IN

THE MUST

Violeta Dimovska1, Klime Beleski 2, Krum Boskov 3

1 University Goce Delcev, Faculty of Agriculture, Goce Delcev 89 2000 Stip, Republic of Macedoniaemail: [email protected]

2 University St. Cyril and Methodius , Institute of Agriculture, Aleksandar Makedonski bb, 1000 Skopje,Republic of Macedonia

3 University St Cyril and Methodius, Faculty for agricultural sciences and food, Aleksandar Makedonskibb,1000 Skopje, Republic of Macedonia

SUMMARYFor the period of 10 years in the condition of Skopje vineyard area, at two regional (Vranecand Smederevka) and two international (Cabernet sauvignon and Chardonnay) grapevinecultivars, the researches are done.The influences of temperatures sum on the duration of following phenological stages (numberof days) are analyzed: from budburst to full maturity; from budburst to flowering and fromveraison to full maturity. The temperature sum has a high impact on the duration of eachphenological stage, especially from budburst to full maturity and from budburst to flowering.The climate has the influence on the content of sugar and total acids in the must. Theseparameters show grater variation at the cultivars Cabernet sauvignon and Chardonnay thancvs. Smedervka and Vranec.Key words: climate, fenology, grape variety, sugar, acids

INTRODUCTION

Temperature sum during the period vegetation are the basic indicator in the choice of variety ina region. Varieties with a different epoch in maturity to grapes in need of different woods in atemperature of individual phenophases vegetation period. In phenophase from verasion to fullmaturity, varieties have the need to sum the highest temperature and lowest in the phenophase-budburs. The need of the vine varieties of a particular temperature depends on the amount ofpoints (biological properties and varietal characteristics as it is length of vegetation period,epoch of maturity on grapes, etc. Skopje vineyards are characterized by the annual averagetemperature air of 12.4 0C (1984/2006), vegetation of 18.8 0C (1984/2006), the annualtemperature sum (from 4.789 0C vegetation temperature and amount of 3673.9 0C (1984/2006).

MATERIALS AND METHODS

Analasys are made for the period 1984-2006 among the varieties vranec, smederevka,merlot and chardonnay which are grown in the national collection at the Agricultural Institut,the Department of viticulture and wine – Skopje.Climate data are used by RHMZ-RM, and the climatic conditions for the Skopje vineyards areanalysed, especially the temperature sum and their influence on particular phenophases of the

3 - 46



5

CONCLUSIONS In the studied years, two models of ripeness have been observed:

- Early technological ripeness. In these years (2003, 2005, 2006 and 2009), the veraison has been early and the process of ripeness has developed in the first fortnight of August. The harvest has been early too, the colour intensity of the wines has been low (except in 2005) and the quality of the wines has been medium. They correspond with a warm year model (high mean temperatures, high Growing Degree Day and low spring rainfalls).

- Late technological ripeness. In these years (2004, 2007 and 2008), the process of ripeness has been carried out in the second fortnight of August and the first days of September. The harvest has been coinciding with the technological ripeness, obtaining more balanced musts. The colour intensity of the wines has been high and the quality of the wines has been very good. They correspond with a model mild-fresh year (moderate mean temperatures, low Growing Degree Day and high spring rainfalls).

The study of the environmental parameters (mean temperature and rainfalls from April to September) can be useful in the prediction of the vintage model. May monthly mean temperature seems to be the key for the development of the photosynthetic apparatus of the grapevine and, therefore, it determines the whole vegetative cycle in the studied years: A fresh May (Tm <15ºC, indicated in grey in the Tab. 1) delays the growth of the buds and the whole cycle. On the other hand, a hot May (Tm> 15ºC) supposes a rapid growth during this month, developing the photosynthetic apparatus that is not stopped in the following months that are always hot and leads to an early ripeness with high temperatures.

ACKNOWLEDGMENTS The present work has been elaborated in collaboration with Bodegas Naranjo S.L. that has provided the

facilities and the staff of the cellar. Special gratefulness for Mrs Pilar Almansa and Mr Ramon Muñoz de Cuerva.

BIBLIOGRAPHY

Bergquist, J., Dookozlian, N., Ebisuda, N., 2001. Sunlight Exposure and Temperature Effects on Berry Growth and Composition of Cabernet Sauvignon and Grenache in the Central San Joaquin Valley of California. Am.J. Enol. Vitic. 52(1): 1-7. Hidalgo, L., 2002.Tratado de Viticultura General. 3rd edition, Madrid: Mundiprensa. Jackson, D., Lombard, P., 1993. Enviromental and Management Practices Affecting Grape Composition and Wine Quality- A Review. Am. J. Enol. Vitic., 44( 4):409-430. Jiménez, J., 1993. Adaptación de 10 cultivares tintos de vid (Vitis vinífera L.) a la región de La Mancha. Doctoral Thesis. Univ. Politéc. Madrid. Dept. Fitotécnia y Producción Vegetal. 205 págs. Spayd, S., Tarara, J., Mee, D., Ferguson, J., 2002. Separation of Sunlight and temperature Effects on the Composition of Vitis vinifera cv. Merlot Berries. Am.J. Enol. Vitic, 53(3): 171-182. Van Leeuwen, C., Friant, P., Choné, X., Tregoat, O., Koundouras, S., Dubordieu, D., 2004. Influence of Climate, Soil and Cultivar on Terroir. Am J. Enol. Vitic., 55(3): 207-217. Winkler, A.J., 1965. Viticultura. 1st edition, Mexico:Compañia Editorial Continental.

THE INFLUENCE OF CLIMATE ON THE GRAPEVINEPHENOLOGY AND CONTENT OF SUGAR AND TOTAL ACIDS IN

THE MUST

Violeta Dimovska1, Klime Beleski 2, Krum Boskov 3

1 University Goce Delcev, Faculty of Agriculture, Goce Delcev 89 2000 Stip, Republic of Macedoniaemail: [email protected]

2 University St. Cyril and Methodius , Institute of Agriculture, Aleksandar Makedonski bb, 1000 Skopje,Republic of Macedonia

3 University St Cyril and Methodius, Faculty for agricultural sciences and food, Aleksandar Makedonskibb,1000 Skopje, Republic of Macedonia

SUMMARYFor the period of 10 years in the condition of Skopje vineyard area, at two regional (Vranecand Smederevka) and two international (Cabernet sauvignon and Chardonnay) grapevinecultivars, the researches are done.The influences of temperatures sum on the duration of following phenological stages (numberof days) are analyzed: from budburst to full maturity; from budburst to flowering and fromveraison to full maturity. The temperature sum has a high impact on the duration of eachphenological stage, especially from budburst to full maturity and from budburst to flowering.The climate has the influence on the content of sugar and total acids in the must. Theseparameters show grater variation at the cultivars Cabernet sauvignon and Chardonnay thancvs. Smedervka and Vranec.Key words: climate, fenology, grape variety, sugar, acids

INTRODUCTION

Temperature sum during the period vegetation are the basic indicator in the choice of variety ina region. Varieties with a different epoch in maturity to grapes in need of different woods in atemperature of individual phenophases vegetation period. In phenophase from verasion to fullmaturity, varieties have the need to sum the highest temperature and lowest in the phenophase-budburs. The need of the vine varieties of a particular temperature depends on the amount ofpoints (biological properties and varietal characteristics as it is length of vegetation period,epoch of maturity on grapes, etc. Skopje vineyards are characterized by the annual averagetemperature air of 12.4 0C (1984/2006), vegetation of 18.8 0C (1984/2006), the annualtemperature sum (from 4.789 0C vegetation temperature and amount of 3673.9 0C (1984/2006).


Analasys are made for the period 1984-2006 among the varieties vranec, smederevka,merlot and chardonnay which are grown in the national collection at the Agricultural Institut,the Department of viticulture and wine – Skopje.Climate data are used by RHMZ-RM, and the climatic conditions for the Skopje vineyards areanalysed, especially the temperature sum and their influence on particular phenophases of the

3 - 47



development of the vine . The monitoring of the phenophases development is carried out bythe method of phonological monitoring among each variety separately.The chemical content is determined by standard methods for he content of sugar and totalacids in the must .

RESULTS AND DISCUSSION

In table 1 are presented the vegetational temperature sum data and the number of days fromshoot elongation to full (technological) maturity. In the years of examination the vegetacionaltemperature sum varies and ranges from 3838.4°C in 1984 to 4260,7°C in the year 2000 .The period(number of days) from shoot elongation to full maturity of the grape varies as in the years ofexamination also among species examined . The shortest period is concluded in 2000 among allvarieties, when we have the highest temperature sum (4260.7°C). Among examined species it rangesfrom 126 days among Chardonnay, 129 days among Vranec, 139 among Cabernet sauvignon to 152days among Smederevka. Among the specie Vranec the period of shoot elogation to full maturity lasts140 days in average for the examination period, among Cabernet sauvignon 155 days , amongSmederevka 163 days and among Chardonnay 137 days .

Table 1 Temperature sum during vegetation and numbers of days from shoot elongation tofull maturity

From shoot elogation to full maturityYear Temperature sumduring

vegetation0CVranec

Cabernetsauvignon Smederevka Chardonnay

1984 3838.4 151 152 160 1271998 4207.1 148 171 177 1471999 4128.4 142 161 166 1402000 4260.7 129 139 152 1262001 4225.1 138 140 167 1322002 3893.7 126 151 161 1382003 4173.2 131 142 158 1382004 4010.5 143 155 173 1392005 3961.9 148 148 155 1482006 4038.4 153 148 166 14784/06 3673.9 140 155 163 137

Table 2 Temperature sum during vegetation and numbers of days from budburst to flowerinFrom budburst to floweringYear Temperature sum

duringvegetation 0C

Vranec Cabernetsauvignon

Smederevka Chardonnay

1984 854.6 42 42 44 461998 959.1 51 42 54 521999 944.8 40 42 46 432000 1035.9 32 34 37 312001 924.4 42 40 48 352002 890.3 43 44 40 462003 944.2 35 44 38 462004 879.3 45 50 49 452005 936 33 42 41 422006 956.8 39 36 40 3884/06 41 42 43 42

Figure 1 Numbers of days from budburst to full maturity

In table 2 vegetational temperature sum data and number of days from shoot elongation to fullmaturity are presented.

Vegetational temperature sum in the phenophase from shoot elongation to full maturity rangesfrom 856.6°C in 1984 to 1036°C in 2000.

The period (number of days) from shoot elogation to flowering varies as in years ofexamination and also among examined species.The shortest period is concluded in 2000 among allspecies, when we have the highest vegetational temperature sum (1036.9°C) , and it amounts 31 daysamong Chardonnay , 32 among Vranec, 34 among Cabernet sauvignon to 37 days among Smederevka.

In Table 3 vegetational temperature sum data and number of days from verasion to fullmaturity are presented .

According to the data analysed, it has been concluded that vegetational temperature sum in thisphenophase during the years of investigation slightly varies,and the grape of the sorts examined hasbeen matured in time.

Vegetational temperature sum in the phenophase of verasion to full maturity is rangingfrom 1891.4 in 1984 to 2175.5 degrees Celsius in 2001.The Phenophase (number of days) from grapes turning red to full maturity varies through theyears of investigation,and among the sorts as well. Regarding the Vranec variety, rangingfrom 32 days in 1984 up to 53 days in 2005 or 42 in average;Cabernet sauvignon from 37days (1984,2001)to 53 days in (1988/1989)-avarage 45 days; at Smederevka it ranges from 43in 2000 to 58 days in 2004-average 52 days; Chardonnay ranges from 28 days in 1999 up to38 days(2002,2003,2005,2006) or 34 days average phenophase in the period of investigation.

165

160

155

150

145

140

135

130

125

120

Vranec C.sauvignon Smedervka Shardonnay

3 - 48



development of the vine . The monitoring of the phenophases development is carried out bythe method of phonological monitoring among each variety separately.The chemical content is determined by standard methods for he content of sugar and totalacids in the must .


In table 1 are presented the vegetational temperature sum data and the number of days fromshoot elongation to full (technological) maturity. In the years of examination the vegetacionaltemperature sum varies and ranges from 3838.4°C in 1984 to 4260,7°C in the year 2000 .The period(number of days) from shoot elongation to full maturity of the grape varies as in the years ofexamination also among species examined . The shortest period is concluded in 2000 among allvarieties, when we have the highest temperature sum (4260.7°C). Among examined species it rangesfrom 126 days among Chardonnay, 129 days among Vranec, 139 among Cabernet sauvignon to 152days among Smederevka. Among the specie Vranec the period of shoot elogation to full maturity lasts140 days in average for the examination period, among Cabernet sauvignon 155 days , amongSmederevka 163 days and among Chardonnay 137 days .

Table 1 Temperature sum during vegetation and numbers of days from shoot elongation tofull maturity

From shoot elogation to full maturityYear Temperature sumduring

vegetation0CVranec

Cabernetsauvignon Smederevka Chardonnay

1984 3838.4 151 152 160 1271998 4207.1 148 171 177 1471999 4128.4 142 161 166 1402000 4260.7 129 139 152 1262001 4225.1 138 140 167 1322002 3893.7 126 151 161 1382003 4173.2 131 142 158 1382004 4010.5 143 155 173 1392005 3961.9 148 148 155 1482006 4038.4 153 148 166 14784/06 3673.9 140 155 163 137

Table 2 Temperature sum during vegetation and numbers of days from budburst to flowerinFrom budburst to floweringYear Temperature sum

duringvegetation 0C

Vranec Cabernetsauvignon

Smederevka Chardonnay

1984 854.6 42 42 44 461998 959.1 51 42 54 521999 944.8 40 42 46 432000 1035.9 32 34 37 312001 924.4 42 40 48 352002 890.3 43 44 40 462003 944.2 35 44 38 462004 879.3 45 50 49 452005 936 33 42 41 422006 956.8 39 36 40 3884/06 41 42 43 42

Figure 1 Numbers of days from budburst to full maturity

In table 2 vegetational temperature sum data and number of days from shoot elongation to fullmaturity are presented.

Vegetational temperature sum in the phenophase from shoot elongation to full maturity rangesfrom 856.6°C in 1984 to 1036°C in 2000.

The period (number of days) from shoot elogation to flowering varies as in years ofexamination and also among examined species.The shortest period is concluded in 2000 among allspecies, when we have the highest vegetational temperature sum (1036.9°C) , and it amounts 31 daysamong Chardonnay , 32 among Vranec, 34 among Cabernet sauvignon to 37 days among Smederevka.

In Table 3 vegetational temperature sum data and number of days from verasion to fullmaturity are presented .

According to the data analysed, it has been concluded that vegetational temperature sum in thisphenophase during the years of investigation slightly varies,and the grape of the sorts examined hasbeen matured in time.

Vegetational temperature sum in the phenophase of verasion to full maturity is rangingfrom 1891.4 in 1984 to 2175.5 degrees Celsius in 2001.The Phenophase (number of days) from grapes turning red to full maturity varies through theyears of investigation,and among the sorts as well. Regarding the Vranec variety, rangingfrom 32 days in 1984 up to 53 days in 2005 or 42 in average;Cabernet sauvignon from 37days (1984,2001)to 53 days in (1988/1989)-avarage 45 days; at Smederevka it ranges from 43in 2000 to 58 days in 2004-average 52 days; Chardonnay ranges from 28 days in 1999 up to38 days(2002,2003,2005,2006) or 34 days average phenophase in the period of investigation.

165

160

155

150

145

140

135

130

125

120

Vranec C.sauvignon Smedervka Shardonnay

3 - 49



Table 3 Temperature sum during vegetation and numbers of days from verasion to full maturitFrom verasion to full maturityYear Temperature sum

duringvegetation 0C

VranecVranec

Cabernetsauvignon

SmederevkaSmederevka

[ardoneChardonnay

1984 1891.4 41 37 45 291998 2148.7 32 53 48 301999 2106.6 42 53 50 282000 2144.8 37 44 43 352001 2175.7 42 37 51 322002 1970.2 40 49 51 382003 2124.4 43 40 57 382004 2024.1 49 43 58 302005 2005.2 53 38 49 382006 2029.6 52 45 53 38

84/06 42 45 52 34

Тhe results of the sugar contents and total acids in the must of the sorts of grapes thathave been investigated are shown in table 4.The must of Vranec grapevine has 222 grams in cubic decimeter(g/dm3) average sugarcontent, therefore in the years of investigation ranges from 214 g/dm3 in the year 2000, to 237g/dm3 in 1998.According to Bozinovic (1996) Vranec grapevine concentrates between 210-230 g/dm3. According to Pejovic and his associates (1996) it is 213 g/dm3.The contents of total acids averages 6.2 g/dm3 for the same period of investigation.The contents of sugar in the must of Cabernet sauvignon is cca. 227 g/dm3 for the years of

investigation, and the total acid contents is 7.3 g/dm3. However, Milosavljevic (1998) claimsthat this sort could concentrate more than 8.0 g/dm3 of total acids.The must of the Smederevka grape variety, average concentration of the sugar content is 183

g/dm3,but in the years of investigation varies from 173 g/dm3 in 2004 to 199 g/dm3 in theyear 2000. As for the total acid contents,average value is 6.0 g/dm3,but it also varies from 5.5in 1998 to 6.3 g/dm3 in 1984/2002.

The must in the Chardonnay grapevine contains average 194 g/dm3 and 6.9 g/dm3 totalacid content for the investigation period. According to Srebra Ilic-Popova and ascts.(1999)Chardonnay grapes grown in the area of Skopje concentrates 219 g/dm3 of sugar and 7.2g/dm3 total acids.

The common conclusion is that the content of sugar and total acids in the sorts abovementioned are within the limits of the biological features,so it meets the requirements forquality wine producing.

Table 4 Content of sugar and total acids in the must ( g/dm3)Vranec Cabernet

sauvignonSmederevka ShardonnayYear

sugar t.acids sugar t.acids sugar t.acids sugar t.acids

1984 235 5.0 245 8.6 178 6.3 219 7.51998 237 6.0 235 7.2 178 5.5 197 5.91999 223 6.9 233 7.8 185 5.7 222 6.92000 214 6.8 205 7.1 199 5.3 228 6.42001 221 5.9 212 6.5 188 5.9 230 6.62002 220 6.9 223 7.1 177 6.3 208 8.22003 215 6.7 220 7.5 182 6.2 208 6.92004 218 6.3 248 5.5 173 6.1 223 6.72005 218 6.0 226 7.9 182 5.9 232 6.72006 223 5.5 226 7.9 189 5.4 202 6.8

84/06 222 6.2 227 7.3 183 6.0 194 6.9

CONCLUSIONSThe Vineyard area of Skopje is characterized by substantial vegetation temperatute

summ,enabling succesfully growing of the mentioned sorts of grapewine,and the maturity ofthe grapes in time.

At the average vegetation temperature of 3.673.9 degrees Celsius, the period frombudburst to full maturity is as follows: Vranec-140 days; Cabenet sauvignon -155 days;Smederevka-163 days and Chardonnay-137 days.

Average for the period of investigation,there is 1-2 days difference among grapevinesinvestigated- from budburst to blooming phenophase.

The sugar content and total acids in the must among these sorts of grapes meets the criteriasfor production od quality vines.

Average for the period of investigation, there is 1-2 days difference among grapevinesinvestigated- from budburst to blooming phenophase.


BIBLIOGRAPHY

GREGORY V. JONES and ROBERT E. DAVIS (2000): Climate Influenses on GrapevinePhenology,Grape Composition and Wine Production and quality for Bordeaux,France. Am.J.Enol.vitic. Vol.51, No 3.DAVIS R.E. and G.V.JONES (1998): Climatic factors influensing grapevine phenology in Bordeaux.p.p. 62-65. 23rd Conferance on Agrucultural and Forest Meteorology of the American MeteorologicalSociety. Alboquerque.New Mexico.COOMBE B.G. (1988): Grapevine Phenology in Viticulture,Volume 1. p.p. 139-153. AustralianIndustrial Publishers, Adelaide.SREBRA-ILIC-POPOVA, B.VOJNOSKI, VIOLETA JEVTIMOVA, K.BELESKI and BILJANAMISIC (1999): The influence of climate condition on content of sugar and total acids in the must ofsome wine cultivars. Jurnal of Agricultural Scienses, vol.44, No 2, 167-172.Belgrade.Klime BELESKI, Zvonimir BOZINOVIC, Violeta DIMOVSKA, Srebra ILIC-POPOVA,Donka DONEVA-SAPCESKA (2008): Climate influence on the grapevine phenology andanthocyanins content in wines from the Skopje vineyard area, Republic of Macedonia. VIIe

Congres International des terroirs viticoles/VIIth International terroir Congress. May, Nyon,Switzerland.

3 - 50



Table 3 Temperature sum during vegetation and numbers of days from verasion to full maturitFrom verasion to full maturityYear Temperature sum

duringvegetation 0C

VranecVranec

Cabernetsauvignon

SmederevkaSmederevka

[ardoneChardonnay

1984 1891.4 41 37 45 291998 2148.7 32 53 48 301999 2106.6 42 53 50 282000 2144.8 37 44 43 352001 2175.7 42 37 51 322002 1970.2 40 49 51 382003 2124.4 43 40 57 382004 2024.1 49 43 58 302005 2005.2 53 38 49 382006 2029.6 52 45 53 38

84/06 42 45 52 34

Тhe results of the sugar contents and total acids in the must of the sorts of grapes thathave been investigated are shown in table 4.The must of Vranec grapevine has 222 grams in cubic decimeter(g/dm3) average sugarcontent, therefore in the years of investigation ranges from 214 g/dm3 in the year 2000, to 237g/dm3 in 1998.According to Bozinovic (1996) Vranec grapevine concentrates between 210-230 g/dm3. According to Pejovic and his associates (1996) it is 213 g/dm3.The contents of total acids averages 6.2 g/dm3 for the same period of investigation.The contents of sugar in the must of Cabernet sauvignon is cca. 227 g/dm3 for the years of

investigation, and the total acid contents is 7.3 g/dm3. However, Milosavljevic (1998) claimsthat this sort could concentrate more than 8.0 g/dm3 of total acids.The must of the Smederevka grape variety, average concentration of the sugar content is 183

g/dm3,but in the years of investigation varies from 173 g/dm3 in 2004 to 199 g/dm3 in theyear 2000. As for the total acid contents,average value is 6.0 g/dm3,but it also varies from 5.5in 1998 to 6.3 g/dm3 in 1984/2002.

The must in the Chardonnay grapevine contains average 194 g/dm3 and 6.9 g/dm3 totalacid content for the investigation period. According to Srebra Ilic-Popova and ascts.(1999)Chardonnay grapes grown in the area of Skopje concentrates 219 g/dm3 of sugar and 7.2g/dm3 total acids.

The common conclusion is that the content of sugar and total acids in the sorts abovementioned are within the limits of the biological features,so it meets the requirements forquality wine producing.

Table 4 Content of sugar and total acids in the must ( g/dm3)Vranec Cabernet

sauvignonSmederevka ShardonnayYear

sugar t.acids sugar t.acids sugar t.acids sugar t.acids

1984 235 5.0 245 8.6 178 6.3 219 7.51998 237 6.0 235 7.2 178 5.5 197 5.91999 223 6.9 233 7.8 185 5.7 222 6.92000 214 6.8 205 7.1 199 5.3 228 6.42001 221 5.9 212 6.5 188 5.9 230 6.62002 220 6.9 223 7.1 177 6.3 208 8.22003 215 6.7 220 7.5 182 6.2 208 6.92004 218 6.3 248 5.5 173 6.1 223 6.72005 218 6.0 226 7.9 182 5.9 232 6.72006 223 5.5 226 7.9 189 5.4 202 6.8

84/06 222 6.2 227 7.3 183 6.0 194 6.9

CONCLUSIONSThe Vineyard area of Skopje is characterized by substantial vegetation temperatute

summ,enabling succesfully growing of the mentioned sorts of grapewine,and the maturity ofthe grapes in time.

At the average vegetation temperature of 3.673.9 degrees Celsius, the period frombudburst to full maturity is as follows: Vranec-140 days; Cabenet sauvignon -155 days;Smederevka-163 days and Chardonnay-137 days.

Average for the period of investigation,there is 1-2 days difference among grapevinesinvestigated- from budburst to blooming phenophase.


Average for the period of investigation, there is 1-2 days difference among grapevinesinvestigated- from budburst to blooming phenophase.


BIBLIOGRAPHY

GREGORY V. JONES and ROBERT E. DAVIS (2000): Climate Influenses on GrapevinePhenology,Grape Composition and Wine Production and quality for Bordeaux,France. Am.J.Enol.vitic. Vol.51, No 3.DAVIS R.E. and G.V.JONES (1998): Climatic factors influensing grapevine phenology in Bordeaux.p.p. 62-65. 23rd Conferance on Agrucultural and Forest Meteorology of the American MeteorologicalSociety. Alboquerque.New Mexico.COOMBE B.G. (1988): Grapevine Phenology in Viticulture,Volume 1. p.p. 139-153. AustralianIndustrial Publishers, Adelaide.SREBRA-ILIC-POPOVA, B.VOJNOSKI, VIOLETA JEVTIMOVA, K.BELESKI and BILJANAMISIC (1999): The influence of climate condition on content of sugar and total acids in the must ofsome wine cultivars. Jurnal of Agricultural Scienses, vol.44, No 2, 167-172.Belgrade.Klime BELESKI, Zvonimir BOZINOVIC, Violeta DIMOVSKA, Srebra ILIC-POPOVA,Donka DONEVA-SAPCESKA (2008): Climate influence on the grapevine phenology andanthocyanins content in wines from the Skopje vineyard area, Republic of Macedonia. VIIe

Congres International des terroirs viticoles/VIIth International terroir Congress. May, Nyon,Switzerland.

3 - 51



EFFECTS OF MESOCLIMATE ON THE YIELD, QUALITY AND PHENOLIC MATURITY OF GRENACHE

M. Nadal1, F. de Herralde2, M. Edo1, M. Lampreave1, R.Savé1

(1) Grup de Recerca Viti-vinicultura. Facultat d’Enologia, Dept. Bioquímica i Biotecnologia, URV

Marcel·lí Domingo s/n. Campus Sant Pere Sescelades. 43007 Tarragona. Spain

[email protected](2) IRTA Torre Marimon, Ecofisiologia.

Torre Marimon. 08140. Caldes de Montbui. Spain

ABSTRACT

The potential climate change, due to global change, will increase temperature general and could increase

at local level. These changes are not going to be the same in different parts of the world, being especially

important in the Mediterranean Basin. Thus, according to the most pessimistic predictions temperature can

rise until 4ºC and precipitation can be reduced close to 20% but this would be different according local

conditions, being also changes in the distribution. In order to study the differences promoted by these

climate differences we compared the phenology, yield and quality parameters of Grenache, grafted onto

110-R in two mesoclimatic areas in Catalonia (Spain), Batea (TA: Terra Alta Appellation) and Caldes de

Montbui (CAT: Catalunya Appellation) during two consecutive years 2007 and 2008.

In TA rainfall and potential evapotranspiration (ET0) were higher than in CAT, but accumulated

growing degree days (∑GDD) were lower, due to lower maximum temperatures and higher minimum

temperatures in winter in CAT. The year 2007 was drier and warmer in both locations. Yield was

significantly lower only in CAT2007, being no differences in leaf area, nor pruning weight. Veraison and

harvest were advanced in 2007 in both locations. Phenological stages were longer in CAT both years. The

length of the period between flowering to veraison, and from veraison to harvest is longer when

accumulated rainfall during each period is higher. On the other hand, the higher the average of GDD

during the period, the shorter the period was. Probable alcohol degree (PAD), Total Phenol Index (TPI),

Color Index (CI), Anthocyanin Content (ANTT and ANTE), were higher and Flavan-3-ols content

(DMACH) and Seed Maturity (SM) were lower in 2008, in both locations than in 2007, which could

indicate that these parameters are very affected by drought, that in 2007 was one of the most dry ripening

periods of last century in Catalonia.

KEYWORDClimate – Vitis – grapevine – drought – phenols

INTRODUCTION

The environmental factors and the cultural practices influence the phenological stages of vines, as

well as in the accumulation of primary and secondary metabolites in the berry (Cortell et al.

2007). Growth measurements give essential information to compare plots under different

conditions (Deloire et al. 2005). High vegetative density in a vine can modify its microclimate by

increasing the shadow of the canes, reducing air circulation, increasing humidity and reducing

temperature. On the contrary, vines with scarce vegetation can have an excessive solar exposition

(Jackson and Lombard, 1993). Different authors (Bergqvist et al. 2001; Coombe, 1987; Smart,

1987) studied the effects of radiation and temperature on the composition and concentration of

primary metabolites. During ripening, the evolution of sugar concentration is positively

correlated with the anthocyanins concentration and the total polyphenol index (Hardie and

Considine, 1976; Hrazdina et al. 1984; Río Segade et al. 2008), and negatively correlated with

the total acidity and malic acid (Barbeau et al. 2004). Nevertheless, the phenolic composition in

relation with the sugars, evolutions differently along the ripening depending on the

edaphoclimatic factors (radiation, temperature, humidity, soil composition, water availability,

etc.) genetic factors (variety, clone, rootstock) and cultural factors (training and pruning system,

nutrition, wanted degree of berry maturity, etc.) (Jackson and Lombard, 1993; Van Leeuwen et

al. 2004). The influence of climate on berry composition has been widely studied. Meriaux

(1982), Coombe (1987), Bergqvist et al. (2001), Spayd et al. (2002) founded that higher the

temperatures, lower the total acidity and higher the sugar levels. Moreover, usually color

(anthocyanins) increases with temperature, proportionally to sugar concentration. Cacho et al.

(1992), Hermosín and García-Romero (2004), found that the anthocyanin content of berries of

the same variety can remarkably vary from vintage to vintage or, within the same vintage,

between plots due to the strong influence of climate. Furthermore sugars, organic acids, and

phenolic compounds concentrations in must can vary depending on the shape of the vine, since

the radiation received by the bunches changes (Smart 1987; Bergqvist et al. 2001; Spayd et al.

2002; Downey et al. 2006; Tarara et al. 2008).


The study was carried out in two different denominations of origin in the North East part of

Spain. The first one was located in Batea (41º11'N, 0º21'E; altitude 236 m), in the Terra Alta

Apellation (DO) (from now TA). The soil is calcareous of clay-loam texture, with 11% limestone

content, 7.7 pH and good content of potassium The average annual temperature is 14.5ºC and the

average rainfall of 450 mm·yr-1. Grenache (Vitis vinifera) vines grafted onto 110-Richter (V.

rupestris Martin x V. berlandieri Resseguier nº2) were 10 years old. Bush vines were pruned in

spurs, with a vine spacing of 1.4 x 2.8 m, and grow without any irrigation. The second location

was in Caldes de Montbui (41º38'N 2º9'E; altitude 176 m) in the Catalunya Appellation (DO)

(from now CAT), where the soil is loam. The average annual temperature is 14.4ºC and the

average rainfall of 641 mm·yr-1. 3-year-old vines of Grenache grafted onto 110-R, were trained to

a bilateral cordon with vertical shoot positioning (VSP) with a vine spacing of 1.5 x 3.0 m, and

grow without any irrigation.

From veraison to harvest each year, 500 berries per plot were sampled several times. The berry

sample was divided in 3 subsamples in order to do: a) classical maturity controls (berry weight,

total acidity and sugar content (OIV, 1990), b) phenolic ripeness was analyzed by the Glories

method (Ribéreau-Gayon et al. 2000). The easily extractable anthocyanins were extracted at pH

3.6 instead of pH 3.2. We used the pH 1 and pH 3.6 buffer extractions to analyze the following:

b1) the total (pH1) and extractable (pH3.6) anthocyanins (analyzed using the bisulphite

discoloration method) at 520nm; b2) the total phenols index (A280) and b3) the phenolic potential

was calculated as Extractability index [(%AE) =(ApH1 – ApH3.6) x 100/ ApH1] and Seed maturity

index [(%SM)= (A280(pH3.6) – (ApH3.6 x 40)/1000) x 100 / A280(pH3.6)]; and c) berry skin anthocyanins

extracted by acid solution (200 ml HCl: ethanol 800 ml and read at 535 nm). All the analyses

were done by triplicate. ANOVA procedure was applied when appropriate. Leaf area was done

by leaf count and measuring the length of central nerve (Carbonneau, 1976; Baeza et al 1997;

Cuevas, 2001)

3 - 52



EFFECTS OF MESOCLIMATE ON THE YIELD, QUALITY AND PHENOLIC MATURITY OF GRENACHE

M. Nadal1, F. de Herralde2, M. Edo1, M. Lampreave1, R.Savé1

(1) Grup de Recerca Viti-vinicultura. Facultat d’Enologia, Dept. Bioquímica i Biotecnologia, URV

Marcel·lí Domingo s/n. Campus Sant Pere Sescelades. 43007 Tarragona. Spain

[email protected](2) IRTA Torre Marimon, Ecofisiologia.

Torre Marimon. 08140. Caldes de Montbui. Spain

ABSTRACT

The potential climate change, due to global change, will increase temperature general and could increase

at local level. These changes are not going to be the same in different parts of the world, being especially

important in the Mediterranean Basin. Thus, according to the most pessimistic predictions temperature can

rise until 4ºC and precipitation can be reduced close to 20% but this would be different according local

conditions, being also changes in the distribution. In order to study the differences promoted by these

climate differences we compared the phenology, yield and quality parameters of Grenache, grafted onto

110-R in two mesoclimatic areas in Catalonia (Spain), Batea (TA: Terra Alta Appellation) and Caldes de

Montbui (CAT: Catalunya Appellation) during two consecutive years 2007 and 2008.

In TA rainfall and potential evapotranspiration (ET0) were higher than in CAT, but accumulated

growing degree days (∑GDD) were lower, due to lower maximum temperatures and higher minimum

temperatures in winter in CAT. The year 2007 was drier and warmer in both locations. Yield was

significantly lower only in CAT2007, being no differences in leaf area, nor pruning weight. Veraison and

harvest were advanced in 2007 in both locations. Phenological stages were longer in CAT both years. The

length of the period between flowering to veraison, and from veraison to harvest is longer when

accumulated rainfall during each period is higher. On the other hand, the higher the average of GDD

during the period, the shorter the period was. Probable alcohol degree (PAD), Total Phenol Index (TPI),

Color Index (CI), Anthocyanin Content (ANTT and ANTE), were higher and Flavan-3-ols content

(DMACH) and Seed Maturity (SM) were lower in 2008, in both locations than in 2007, which could

indicate that these parameters are very affected by drought, that in 2007 was one of the most dry ripening

periods of last century in Catalonia.

KEYWORDClimate – Vitis – grapevine – drought – phenols

INTRODUCTION

The environmental factors and the cultural practices influence the phenological stages of vines, as

well as in the accumulation of primary and secondary metabolites in the berry (Cortell et al.

2007). Growth measurements give essential information to compare plots under different

conditions (Deloire et al. 2005). High vegetative density in a vine can modify its microclimate by

increasing the shadow of the canes, reducing air circulation, increasing humidity and reducing

temperature. On the contrary, vines with scarce vegetation can have an excessive solar exposition

(Jackson and Lombard, 1993). Different authors (Bergqvist et al. 2001; Coombe, 1987; Smart,

1987) studied the effects of radiation and temperature on the composition and concentration of

primary metabolites. During ripening, the evolution of sugar concentration is positively

correlated with the anthocyanins concentration and the total polyphenol index (Hardie and

Considine, 1976; Hrazdina et al. 1984; Río Segade et al. 2008), and negatively correlated with

the total acidity and malic acid (Barbeau et al. 2004). Nevertheless, the phenolic composition in

relation with the sugars, evolutions differently along the ripening depending on the

edaphoclimatic factors (radiation, temperature, humidity, soil composition, water availability,

etc.) genetic factors (variety, clone, rootstock) and cultural factors (training and pruning system,

nutrition, wanted degree of berry maturity, etc.) (Jackson and Lombard, 1993; Van Leeuwen et

al. 2004). The influence of climate on berry composition has been widely studied. Meriaux

(1982), Coombe (1987), Bergqvist et al. (2001), Spayd et al. (2002) founded that higher the

temperatures, lower the total acidity and higher the sugar levels. Moreover, usually color

(anthocyanins) increases with temperature, proportionally to sugar concentration. Cacho et al.

(1992), Hermosín and García-Romero (2004), found that the anthocyanin content of berries of

the same variety can remarkably vary from vintage to vintage or, within the same vintage,

between plots due to the strong influence of climate. Furthermore sugars, organic acids, and

phenolic compounds concentrations in must can vary depending on the shape of the vine, since

the radiation received by the bunches changes (Smart 1987; Bergqvist et al. 2001; Spayd et al.

2002; Downey et al. 2006; Tarara et al. 2008).


The study was carried out in two different denominations of origin in the North East part of

Spain. The first one was located in Batea (41º11'N, 0º21'E; altitude 236 m), in the Terra Alta

Apellation (DO) (from now TA). The soil is calcareous of clay-loam texture, with 11% limestone

content, 7.7 pH and good content of potassium The average annual temperature is 14.5ºC and the

average rainfall of 450 mm·yr-1. Grenache (Vitis vinifera) vines grafted onto 110-Richter (V.

rupestris Martin x V. berlandieri Resseguier nº2) were 10 years old. Bush vines were pruned in

spurs, with a vine spacing of 1.4 x 2.8 m, and grow without any irrigation. The second location

was in Caldes de Montbui (41º38'N 2º9'E; altitude 176 m) in the Catalunya Appellation (DO)

(from now CAT), where the soil is loam. The average annual temperature is 14.4ºC and the

average rainfall of 641 mm·yr-1. 3-year-old vines of Grenache grafted onto 110-R, were trained to

a bilateral cordon with vertical shoot positioning (VSP) with a vine spacing of 1.5 x 3.0 m, and

grow without any irrigation.

From veraison to harvest each year, 500 berries per plot were sampled several times. The berry

sample was divided in 3 subsamples in order to do: a) classical maturity controls (berry weight,

total acidity and sugar content (OIV, 1990), b) phenolic ripeness was analyzed by the Glories

method (Ribéreau-Gayon et al. 2000). The easily extractable anthocyanins were extracted at pH

3.6 instead of pH 3.2. We used the pH 1 and pH 3.6 buffer extractions to analyze the following:

b1) the total (pH1) and extractable (pH3.6) anthocyanins (analyzed using the bisulphite

discoloration method) at 520nm; b2) the total phenols index (A280) and b3) the phenolic potential

was calculated as Extractability index [(%AE) =(ApH1 – ApH3.6) x 100/ ApH1] and Seed maturity

index [(%SM)= (A280(pH3.6) – (ApH3.6 x 40)/1000) x 100 / A280(pH3.6)]; and c) berry skin anthocyanins

extracted by acid solution (200 ml HCl: ethanol 800 ml and read at 535 nm). All the analyses

were done by triplicate. ANOVA procedure was applied when appropriate. Leaf area was done

by leaf count and measuring the length of central nerve (Carbonneau, 1976; Baeza et al 1997;

Cuevas, 2001)

3 - 53



At harvest we measured yield (kg·vine-1) and pruning weight in winter dormancy (kg·vine-1);

and we calculated the Ravaz index.

Weather conditions were monitored during the extent of the experiment through weather

stations belonging to the official Agro-meteorological network in Catalunya (XAC), located in

the same municipalities than the vineyards.


The two considered locations are representative of different viticultural areas and have different

mesoclimates. The two studied years also differed in both places, especially in rainfall amount

and distribution (Fig. 1A). In TA, 2007 was 15% drier (384 mm), than the average (450 mm),

whereas 2008 was 31% rainier (588mm) than average. In CAT, 2007 was 31% drier (440mm)

than the average (641mm) and 2008 16% rainier (742mm). These changes in rainfall amount and

distribution were less negative in TA than in CAT. In both years ET0 was higher in TA,

particularly due to higher vapor pressure deficit and wind. The ∑GDD10 did not differ very

much between locations but years. 2007 accumulated more GDD from April to the end of the

year in CAT than in TA, whereas in 2008 the two locations had an equal GDD value (lower than

in 2007) up to harvest time.

0

20

40

60

80

100

120

140

160

180

200

0

50

100

150

200

250

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

Potential evapotranspiration (mm)

Rainfall (mm)

TA Rainfall 07 TA Rainfall 08 CAT Rainfall 07 CAT Rainfall 08

TA ET0 07 TA ET0 08 CAT ET0 07 CAT ET0 08A

0

250

500

750

1000

1250

1500

1750

2000

2250

2500

J an Feb Mar Apr May J un J ul Aug Sep Oct Nov Dec

ΣGDD10(ºC)

B TA2007 TA2008 CAT2007 CAT2008

Figure 1. A) Monthly rainfall (mm) and potential evapotranspiration (ET0 (mm) and B) Accumulated growing degree days

(∑GDD (ºC)) in DO Terra Alta (TA) and DO Catalunya (CAT) during 2007 and 2008.

Yield and pruning weight were lower in CAT 2007 than in the others (Tab. 1), due to the youth

of the vines. Yield, Leaf area and Ravaz index (Tab. 1) did not show any significant differences

between neither places nor years.

The length of the ripening process was different between locations and years. In 2007 it was

shortest in TA (33 days), whereas the longest in CAT (46 days), but veraison was set on the same

date in both locations. In 2008, veraison and the length of the period in TA resulted similar to

2007 (34 days). In CAT, the veraison was delayed 8 days and moreover the harvest was advanced

resulting in a shorter ripening period in CAT. The length of the ripening period had a significant

positive relationship with the accumulation of growing degree days (∑GDD10 (ºC)) during this

period (r2=0.88), but not with other parameters such as accumulated rainfall or ET0 or mean

thermal amplitude.

The evolution of classical maturity parameters from veraison to harvest for both locations and

years is shown in fig. 2.

In CAT 2007 the berry weight was too high due to the youth of the vines and the poor and not

steady fertility. In 2008, berries were significantly higher in CAT than in TA both years. Target

PAD was around 14.5º, so harvest date was adjusted to it. In CAT in 2007 was impossible to

achieve the level due to intense rainfall around harvest time. In TA 2008 harvest was a bit later

than wanted. The total acidity values were similar for all the years and locations.

Table 1. Yield components for Terra Alta (TA) and Catalunya (CAT) plots in 2007 and 2008. Data are mean of n=3 ± S.E.

Different letters in the same columns indicate significant differences (P≤0.05/ Tukey test)

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

0 10 20 30 40 50

Berry weight (g)

DAV

Berry weight CAT 2007 TA 2007

CAT 2008 TA 2008

8.0

10.0

12.0

14.0

16.0

0 10 20 30 40 50

PAD (º)

DAV

Probable alcohol degree CAT 2007 TA 2007

CAT 2008 TA 2008

3.0

5.0

7.0

9.0

11.0

13.0

15.0

0 10 20 30 40 50

TTA

DAV

Total Tritable Acidity CAT 2007 TA 2007

CAT 2008 TA 2008

Figure 2. Berry weight (g), Probable alcohol degree (º) and total tritable acidity in DO Terra Alta (TA) and DO Catalunya

(CAT) in 2007 and 2008. Horizontal axis is represented in number of days after veraison (DAV) for each location and year.

At harvest TPI, ANTT and ANTE were higher in 2008 than in 2007, whereas DMACH was

lower. This indicates that 2008 was a better year in terms of polyphenol and anthocyanin

accumulation in the berries. Moreover, ANTE was higher in TA than in CAT, showing a higher

value of extractable anthocyanins. The SM at harvest was better in TA than in CAT and in 2008

than in 2007, because of the low values, that indicated less astringency in seeds. The color (CI)

showed the highest values in TA2008 and the lowest in CAT2007. Finally AE was much lower in

TA 2008 than in the other years and locations. The values of AE match well with extractability of

the color (Ribereau-Gayon et al. 2000).

The accumulated rainfall during the veraison to harvest period was much less in TA (2,6 and

21,5 mm in 2007 and 2008 respectively) than in CAT (71,8 and 68,3 mm in 2007 and 2008

respectively). It was positively correlated with BW, PAD, CI, ANTT and ANTE, and negatively

with TTA, DMACH, SM and AE. In TA, where the rainfall was lower these relationships were

significant and last all the period. On the other hand, the CAT samples showed the same trend at

the beginning of the period, but when it rained intensively at the end of the ripening, the trend

was reversed, except in the case of BW and PAD. The rainfall in CAT at the end of the ripening

period is characteristic of this mesoclimate due to the sea influence.

Location Year Pruning weight

(kg·vine-1)

Leaf area at veraison

(m2·vine-1)

Leaf area at harvest

(m2·vine-1)

Yield

(kg·vine-1)

Ravaz Index

TA 2007 0,270 ± 0,047b 2,931 ± 0,863 2,586 ± 0,518 2,581 ± 0,199b 9,557 ± 0,199

2008 0,320 ± 0,010b 2,585 ± 0,438 3,392 ± 0,753 2,872 ± 0,588b 8,978 ± 0,588

CAT 2007 0,079 ± 0,016a 2,684 ± 0,335 3,252 ± 0,196 0,465 ± 0,17a 5,877 ± 0,170

2008 0,385 ± 0,069b 3,004 ± 0,513 3,131 ± 0,355 3,74 ± 0,469b 9,707 ± 0,469

TA 0,295 ± 0,024 2,758 ± 0,44 2,989 ± 0,447 2,726 ± 0,285 9,243 ± 0,285

CAT 0,232 ± 0,061 2,867 ± 0,309 3,183 ± 0,205 2,512 ± 0,664 10,818 ± 0,664

2007 0,151 ± 0,039 2,808 ± 0,418 2,919 ± 0,289 1,523 ± 0,487 10,105 ± 0,487

2008 0,361 ± 0,043 2,824 ± 0,332 3,243 ± 0,346 3,415 ± 0,375 9,465 ± 0,375

Global 0,256 ± 0,157 2,817 ± 0,906 3,093 ± 0,81 2,604 ± 1,449 10,182 ± 1,449

3 - 54



At harvest we measured yield (kg·vine-1) and pruning weight in winter dormancy (kg·vine-1);

and we calculated the Ravaz index.

Weather conditions were monitored during the extent of the experiment through weather

stations belonging to the official Agro-meteorological network in Catalunya (XAC), located in

the same municipalities than the vineyards.


The two considered locations are representative of different viticultural areas and have different

mesoclimates. The two studied years also differed in both places, especially in rainfall amount

and distribution (Fig. 1A). In TA, 2007 was 15% drier (384 mm), than the average (450 mm),

whereas 2008 was 31% rainier (588mm) than average. In CAT, 2007 was 31% drier (440mm)

than the average (641mm) and 2008 16% rainier (742mm). These changes in rainfall amount and

distribution were less negative in TA than in CAT. In both years ET0 was higher in TA,

particularly due to higher vapor pressure deficit and wind. The ∑GDD10 did not differ very

much between locations but years. 2007 accumulated more GDD from April to the end of the

year in CAT than in TA, whereas in 2008 the two locations had an equal GDD value (lower than

in 2007) up to harvest time.

0

20

40

60

80

100

120

140

160

180

200

0

50

100

150

200

250

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

Potential evapotranspiration (mm)

Rainfall (mm)

TA Rainfall 07 TA Rainfall 08 CAT Rainfall 07 CAT Rainfall 08

TA ET0 07 TA ET0 08 CAT ET0 07 CAT ET0 08A

0

250

500

750

1000

1250

1500

1750

2000

2250

2500

J an Feb Mar Apr May J un J ul Aug Sep Oct Nov Dec

ΣGDD10(ºC)

B TA2007 TA2008 CAT2007 CAT2008

Figure 1. A) Monthly rainfall (mm) and potential evapotranspiration (ET0 (mm) and B) Accumulated growing degree days

(∑GDD (ºC)) in DO Terra Alta (TA) and DO Catalunya (CAT) during 2007 and 2008.

Yield and pruning weight were lower in CAT 2007 than in the others (Tab. 1), due to the youth

of the vines. Yield, Leaf area and Ravaz index (Tab. 1) did not show any significant differences

between neither places nor years.

The length of the ripening process was different between locations and years. In 2007 it was

shortest in TA (33 days), whereas the longest in CAT (46 days), but veraison was set on the same

date in both locations. In 2008, veraison and the length of the period in TA resulted similar to

2007 (34 days). In CAT, the veraison was delayed 8 days and moreover the harvest was advanced

resulting in a shorter ripening period in CAT. The length of the ripening period had a significant

positive relationship with the accumulation of growing degree days (∑GDD10 (ºC)) during this

period (r2=0.88), but not with other parameters such as accumulated rainfall or ET0 or mean

thermal amplitude.

The evolution of classical maturity parameters from veraison to harvest for both locations and

years is shown in fig. 2.

In CAT 2007 the berry weight was too high due to the youth of the vines and the poor and not

steady fertility. In 2008, berries were significantly higher in CAT than in TA both years. Target

PAD was around 14.5º, so harvest date was adjusted to it. In CAT in 2007 was impossible to

achieve the level due to intense rainfall around harvest time. In TA 2008 harvest was a bit later

than wanted. The total acidity values were similar for all the years and locations.

Table 1. Yield components for Terra Alta (TA) and Catalunya (CAT) plots in 2007 and 2008. Data are mean of n=3 ± S.E.

Different letters in the same columns indicate significant differences (P≤0.05/ Tukey test)

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

0 10 20 30 40 50

Berry weight (g)

DAV

Berry weight CAT 2007 TA 2007

CAT 2008 TA 2008

8.0

10.0

12.0

14.0

16.0

0 10 20 30 40 50

PAD (º)

DAV

Probable alcohol degree CAT 2007 TA 2007

CAT 2008 TA 2008

3.0

5.0

7.0

9.0

11.0

13.0

15.0

0 10 20 30 40 50

TTA

DAV

Total Tritable Acidity CAT 2007 TA 2007

CAT 2008 TA 2008

Figure 2. Berry weight (g), Probable alcohol degree (º) and total tritable acidity in DO Terra Alta (TA) and DO Catalunya

(CAT) in 2007 and 2008. Horizontal axis is represented in number of days after veraison (DAV) for each location and year.

At harvest TPI, ANTT and ANTE were higher in 2008 than in 2007, whereas DMACH was

lower. This indicates that 2008 was a better year in terms of polyphenol and anthocyanin

accumulation in the berries. Moreover, ANTE was higher in TA than in CAT, showing a higher

value of extractable anthocyanins. The SM at harvest was better in TA than in CAT and in 2008

than in 2007, because of the low values, that indicated less astringency in seeds. The color (CI)

showed the highest values in TA2008 and the lowest in CAT2007. Finally AE was much lower in

TA 2008 than in the other years and locations. The values of AE match well with extractability of

the color (Ribereau-Gayon et al. 2000).

The accumulated rainfall during the veraison to harvest period was much less in TA (2,6 and

21,5 mm in 2007 and 2008 respectively) than in CAT (71,8 and 68,3 mm in 2007 and 2008

respectively). It was positively correlated with BW, PAD, CI, ANTT and ANTE, and negatively

with TTA, DMACH, SM and AE. In TA, where the rainfall was lower these relationships were

significant and last all the period. On the other hand, the CAT samples showed the same trend at

the beginning of the period, but when it rained intensively at the end of the ripening, the trend

was reversed, except in the case of BW and PAD. The rainfall in CAT at the end of the ripening

period is characteristic of this mesoclimate due to the sea influence.

Location Year Pruning weight

(kg·vine-1)

Leaf area at veraison

(m2·vine-1)

Leaf area at harvest

(m2·vine-1)

Yield

(kg·vine-1)

Ravaz Index

TA 2007 0,270 ± 0,047b 2,931 ± 0,863 2,586 ± 0,518 2,581 ± 0,199b 9,557 ± 0,199

2008 0,320 ± 0,010b 2,585 ± 0,438 3,392 ± 0,753 2,872 ± 0,588b 8,978 ± 0,588

CAT 2007 0,079 ± 0,016a 2,684 ± 0,335 3,252 ± 0,196 0,465 ± 0,17a 5,877 ± 0,170

2008 0,385 ± 0,069b 3,004 ± 0,513 3,131 ± 0,355 3,74 ± 0,469b 9,707 ± 0,469

TA 0,295 ± 0,024 2,758 ± 0,44 2,989 ± 0,447 2,726 ± 0,285 9,243 ± 0,285

CAT 0,232 ± 0,061 2,867 ± 0,309 3,183 ± 0,205 2,512 ± 0,664 10,818 ± 0,664

2007 0,151 ± 0,039 2,808 ± 0,418 2,919 ± 0,289 1,523 ± 0,487 10,105 ± 0,487

2008 0,361 ± 0,043 2,824 ± 0,332 3,243 ± 0,346 3,415 ± 0,375 9,465 ± 0,375

Global 0,256 ± 0,157 2,817 ± 0,906 3,093 ± 0,81 2,604 ± 1,449 10,182 ± 1,449

3 - 55



Table 2. Phenolic maturity parameters at harvest for Terra Alta (TA) and Catalunya (CAT) plots in 2007 and 2008. From left to

right: Total Phenol Index (TPI), Color Index (IC), Total Anthocyanin Content (ANTT, mg·l-1), Extractable Anthocyanin Content

(ANTE, mg·l-1), Flavan-3-ols content (mg·l-1) (DMACH), Seed maturity index (SM, %), Extractability index (AE, %). Data are

mean of n=3 ± S.E. Different letters in the same columns indicate significant differences (P≤0.05/ Tukey test).

The response of the quality parameters to the GDD10 and ET0 accumulation during the ripening

period was quite similar in sign and significance. BW, PAD and ANTT showed a positive

correlation with those parameters. The response of TTA and SM was negative and, when

separating by locations and years, the TA2008 data showed a steeper slope. AE only showed a

negative correlation with ET0, but only being significant in TA2008. TPI and CI seemed not to

respond to the tested climatic parameters. Nadal et al. (2008) observed less anthocyanin content

in berries of vines grown under the sea influence (lower VPD, milder temperatures and higher

rainfall).

CONCLUSIONS

In Grenache, a variety with low phenolic potential, the driest years reduce phenol synthesis,

especially anthocyanins synthesis. In both years, the evaporative demand in TA was higher than

in CAT, increasing the effects of the drought.

In CAT appellation, for both years the quality was reduced due to the rainfall just before

harvest, typical in the region. Probable alcohol degree and berry weight increased but

anthocyanins were diluted. Rainfall also increases the length of ripening period in CAT compared

with TA, due to the sea influence.

Extrapolating at a more global scale, those areas in the future exposed to higher ET0 and drier

summers (in length and amount) seem to be more unfavorable to the color synthesis and

extractability and could be a constraint to winegrowing in the future.

ACKNOWLEDGMENTS

The results that appear in this work were partially funded by the Spanish Ministry of Education

AGL2005-06927-C02-02 and by National project CDTI with “Unió de Cooperatives” Company.

Location Year TPI CI ANTT

(mg·L-1)

ANTE

(mg·L-1)

DMACH

(mg·L-1)

SM

(%)

AE

(%)

TA 2007 53,5 ± 0,5 14,44 ± 0,28b 351,1 ± 33,7 282,6 ± 16,9 243,2 ±14,0 78,89 ± 1,24 19,0 ± 2,8a

2008 60,2 ± 1,0 20,58 ± 0,44a 453,7 ± 7,9 441,5 ± 5,4 218,6 ± 3,2 70,68 ± 0,21 2,7 ± 0,9b

CAT 2007 52,5 ± 3,5 6,19 ± 0,94c 290,6 ± 47,2 226,9 ± 49,9 280,7 ± 6,0 82,91 ± 2,85 23,1 ± 4,8a

2008 62,0 ± 2,5 17,09 ± 1,02b 420,1 ± 12,4 309,3 ± 8,8 199,1 ± 18,7 79,93 ± 0,97 26,2 ± 2,6a

TA 56,9 ± 1,6 17,51 ± 1,39 402,4 ± 27,7 362,0 ± 36,4a 230,9 ± 8,4 74,78 ± 1,92b 10,8 ± 3,9

CAT 57,8 ± 2,7 13,14 ± 2,29 362,2 ± 32,3 265,4 ± 25,6b 228,0 ± 19,4 81,79 ± 1,34a 26,7 ± 2,4

2007 53,0 ± 1,6b 10,32 ± 1,90 320,9 ± 29,3b 254,7 ± 26,6b 262,0 ± 10,8a 80,90 ± 1,66a 21,0 ± 2,7

2008 61,2 ± 1,4a 18,58 ± 0,91 434,5 ± 10,0a 366 ,0± 27,2a 207,5 ± 10,8b 75,97 ± 1,94b 16,1 ± 5,0

Global 57,4 ± 1,6 14,77 ± 1,53 382,0 ± 21,4 314,6 ± 24,3 232,6 ± 10,7 78,24 ± 1,43 18,4 ± 2,9

BIBLIOGRAPHY

Baeza P, Ruiz C, Bartolomé MC, and Lissarrague JR 1997. Differences in gas exchange in cv. Tempranillo (Vitis

vinifera L) as affected by training system. Acta Horticulturae (ISHS). 526, 145-156.

Barbeau, G.; Bournand, S.; Champenois, R. ; Bouvet, M. ; Blin, A. ; and Cosneau, M. (2004) Comportement de

quatre cépages rouges du Val de Loire en fonction des variables climatiques. J. Int. Vigne Vin, 38 (1), 35-40.

Bergqvist, J.; Dokoozlian, N. & Ebisuda, N. 2001. Sunlighht exposure and temperature effects on berry growth and

composition of Cabernet Sauvignon and Grenache in the Central San Joaquin Valley of California. Am. J. Enol.

Vitic, 52:1, 1-7.

Cacho, J.; Fernández, P.; Ferreira, V.; and Castells, J.E. (1992). Evolution of five Anthocyanidin-3-glucosides in the

skin of the Tempranillo, Moristel, and Garnacha grape varieties and influence of climatological variables. Am. J.

Enol. Vitic., 43 (3), 244-248.

Carbonneau A. 1976. Principes et méthodes de mesure de la surface foliaire. Essai de caractérisation des types de

feuilles dans le genre Vitis. Ann. Amélio. Plantes, 26:327-343

Coombe,B.G.1987.Influence of temperature on composition and quality of grapes.Acta Hortic.206:23-35.

Cortell, J.M.; Halbleib, M.; Gallagher, A.V.; Righetti, T.L. & Kennedy, J. A. 2007. Influence of vine vigor on grape

(Vitis vinifera L. cv. Pinot Noir) anthocyanins. 1. Anthocyanin concentration and composition in fruit. J. Agric.

Food Chem., 55, 6575-6584.

Cuevas E. 2001. Estudio de mecanismos de adaptación ecofisiológica de la vid (Vitis vinifera L.cv.Tempranillo) al

déficit hídrico. Evaluación del consumo de agua y de las respuestas agronómicas en diferentes regímenes

hídricos. Tesis doctoral. Universidad Politécnica de Madrid.

Deloire, A.; Vaudour, E.; Carey, V. ; Bonnardot, V. & van Leeuwen, C. 2005. Grapevine responses to terroir: a

global approach. J. Int. Sci. Vigne Vin, 39 (4), 149-162.

Downey, M.O.; Dokoozlian, N.K.; and Krstic, M.P. (2006) Cultural Practice and Environmental Impacts on the

Flavonoid Composition of Grapes and Wine: A Review of Recent Research. Am. J. Enol. Vitic., 57:3, 257-268.

Hardie, W. J.; and Considine, J. A. (1976) Response of grapes to water deficit stress in particular stages of

development. Am. J. Enol. Vitic., 27, 55-61.

Hermosín Gutiérrez, I.; and García-Romero, E. (2004) Antocianos de variedades tintas cultivadas en la Mancha:

perfiles varietales característicos de la uva y de los vinos monovarietales, y evolución durante la maduración de la

baya. Alimentaria, Abril 04/127-139.

Hrazdina, G.; Parsons, G.F.; and Mattick, L.R. (1984) Physiological and biochemical events during deveopment and

maturation of grape berries. Am. J. Enol. Vitic. 35 (4), 220-227.

Jackson, J.I. & Lombard, P.B. 1993. Environmental and management practices affecting grape composition and wine

quality – A review. Am. J. Enol. Vitic. 44 (4), 409-429.

Nadal M, Mateos S, and Lampreave, M. (2008) Influence de la topographie etdu mésoclimat sur la composition des

raisins et rendement dans le terroir del’AOC Priorat, VII CongresInt. terroirs viticoles 19 - 23 mai 2008

Nyon,Suïssa,(2):590-595

Meriaux, S. (1982). La vigne et l´eau dans le Midi méditerranéen. Vignes et Vins, nº especial Septiembre: 23–26.

OIV. (1990). “Récueil des Méthodes Internacionales d’Analyse des vins et des Mouts “; Office Internacional de la

Vigne et du Vin: Paris, France.

Ribéreau-Gayon Y, Glories Y, Maujean A. & Dubourdieu D, 2000. Handbook of Enology, Volume 2: The chemistry

of wine and stabilization and treatments. John Wiley & Sons Ltd.

Río Segade, S.; Soto Vázquez, E.; and Díaz Losada, E. (2008) Influence of ripeness grade on accumulation and

extractability of grape skin anthocyanins in different cultivars. J. Food Comp. Anal., 21, 599-607.

Smart, R. 1987. Influence of light on composition and quality of grapes. Acta Hortic. 206:37-43.

Spayd, S.E.; Tarara, J.M.; Mee, D.L.; and Ferguson, J.C. (2002) Separation of sunlight and temperature effects on

the composition of Vitis vinifera cv. Merlot berries. Am. J. Enol. Vitic., 53:3, 171-182.

Tarara, J.M.; Lee, J.; Spayd, S.E.; and Scagel, C.F. (2008) Berry temperature and solar radiation alter acylation,

proportion, and concentration of anthocyanin in Merlot grapes. Am. J. Enol. Vitic., 59:3, 235-246.

Van Leeuwen, C.; Friant, P.; Choné, X.; Tregoat, O.; Koundouras, S.; and Dubourdieu, D. (2004) Influence of

climate, soil, and culrivar on terroir. Am. J. Enol. Vitic. 55:3, 207-217

3 - 56



Table 2. Phenolic maturity parameters at harvest for Terra Alta (TA) and Catalunya (CAT) plots in 2007 and 2008. From left to

right: Total Phenol Index (TPI), Color Index (IC), Total Anthocyanin Content (ANTT, mg·l-1), Extractable Anthocyanin Content

(ANTE, mg·l-1), Flavan-3-ols content (mg·l-1) (DMACH), Seed maturity index (SM, %), Extractability index (AE, %). Data are

mean of n=3 ± S.E. Different letters in the same columns indicate significant differences (P≤0.05/ Tukey test).

The response of the quality parameters to the GDD10 and ET0 accumulation during the ripening

period was quite similar in sign and significance. BW, PAD and ANTT showed a positive

correlation with those parameters. The response of TTA and SM was negative and, when

separating by locations and years, the TA2008 data showed a steeper slope. AE only showed a

negative correlation with ET0, but only being significant in TA2008. TPI and CI seemed not to

respond to the tested climatic parameters. Nadal et al. (2008) observed less anthocyanin content

in berries of vines grown under the sea influence (lower VPD, milder temperatures and higher

rainfall).

CONCLUSIONS

In Grenache, a variety with low phenolic potential, the driest years reduce phenol synthesis,

especially anthocyanins synthesis. In both years, the evaporative demand in TA was higher than

in CAT, increasing the effects of the drought.

In CAT appellation, for both years the quality was reduced due to the rainfall just before

harvest, typical in the region. Probable alcohol degree and berry weight increased but

anthocyanins were diluted. Rainfall also increases the length of ripening period in CAT compared

with TA, due to the sea influence.

Extrapolating at a more global scale, those areas in the future exposed to higher ET0 and drier

summers (in length and amount) seem to be more unfavorable to the color synthesis and

extractability and could be a constraint to winegrowing in the future.

ACKNOWLEDGMENTS

The results that appear in this work were partially funded by the Spanish Ministry of Education

AGL2005-06927-C02-02 and by National project CDTI with “Unió de Cooperatives” Company.

Location Year TPI CI ANTT

(mg·L-1)

ANTE

(mg·L-1)

DMACH

(mg·L-1)

SM

(%)

AE

(%)

TA 2007 53,5 ± 0,5 14,44 ± 0,28b 351,1 ± 33,7 282,6 ± 16,9 243,2 ±14,0 78,89 ± 1,24 19,0 ± 2,8a

2008 60,2 ± 1,0 20,58 ± 0,44a 453,7 ± 7,9 441,5 ± 5,4 218,6 ± 3,2 70,68 ± 0,21 2,7 ± 0,9b

CAT 2007 52,5 ± 3,5 6,19 ± 0,94c 290,6 ± 47,2 226,9 ± 49,9 280,7 ± 6,0 82,91 ± 2,85 23,1 ± 4,8a

2008 62,0 ± 2,5 17,09 ± 1,02b 420,1 ± 12,4 309,3 ± 8,8 199,1 ± 18,7 79,93 ± 0,97 26,2 ± 2,6a

TA 56,9 ± 1,6 17,51 ± 1,39 402,4 ± 27,7 362,0 ± 36,4a 230,9 ± 8,4 74,78 ± 1,92b 10,8 ± 3,9

CAT 57,8 ± 2,7 13,14 ± 2,29 362,2 ± 32,3 265,4 ± 25,6b 228,0 ± 19,4 81,79 ± 1,34a 26,7 ± 2,4

2007 53,0 ± 1,6b 10,32 ± 1,90 320,9 ± 29,3b 254,7 ± 26,6b 262,0 ± 10,8a 80,90 ± 1,66a 21,0 ± 2,7

2008 61,2 ± 1,4a 18,58 ± 0,91 434,5 ± 10,0a 366 ,0± 27,2a 207,5 ± 10,8b 75,97 ± 1,94b 16,1 ± 5,0

Global 57,4 ± 1,6 14,77 ± 1,53 382,0 ± 21,4 314,6 ± 24,3 232,6 ± 10,7 78,24 ± 1,43 18,4 ± 2,9

BIBLIOGRAPHY

Baeza P, Ruiz C, Bartolomé MC, and Lissarrague JR 1997. Differences in gas exchange in cv. Tempranillo (Vitis

vinifera L) as affected by training system. Acta Horticulturae (ISHS). 526, 145-156.

Barbeau, G.; Bournand, S.; Champenois, R. ; Bouvet, M. ; Blin, A. ; and Cosneau, M. (2004) Comportement de

quatre cépages rouges du Val de Loire en fonction des variables climatiques. J. Int. Vigne Vin, 38 (1), 35-40.

Bergqvist, J.; Dokoozlian, N. & Ebisuda, N. 2001. Sunlighht exposure and temperature effects on berry growth and

composition of Cabernet Sauvignon and Grenache in the Central San Joaquin Valley of California. Am. J. Enol.

Vitic, 52:1, 1-7.

Cacho, J.; Fernández, P.; Ferreira, V.; and Castells, J.E. (1992). Evolution of five Anthocyanidin-3-glucosides in the

skin of the Tempranillo, Moristel, and Garnacha grape varieties and influence of climatological variables. Am. J.

Enol. Vitic., 43 (3), 244-248.

Carbonneau A. 1976. Principes et méthodes de mesure de la surface foliaire. Essai de caractérisation des types de

feuilles dans le genre Vitis. Ann. Amélio. Plantes, 26:327-343

Coombe,B.G.1987.Influence of temperature on composition and quality of grapes.Acta Hortic.206:23-35.

Cortell, J.M.; Halbleib, M.; Gallagher, A.V.; Righetti, T.L. & Kennedy, J. A. 2007. Influence of vine vigor on grape

(Vitis vinifera L. cv. Pinot Noir) anthocyanins. 1. Anthocyanin concentration and composition in fruit. J. Agric.

Food Chem., 55, 6575-6584.

Cuevas E. 2001. Estudio de mecanismos de adaptación ecofisiológica de la vid (Vitis vinifera L.cv.Tempranillo) al

déficit hídrico. Evaluación del consumo de agua y de las respuestas agronómicas en diferentes regímenes

hídricos. Tesis doctoral. Universidad Politécnica de Madrid.

Deloire, A.; Vaudour, E.; Carey, V. ; Bonnardot, V. & van Leeuwen, C. 2005. Grapevine responses to terroir: a

global approach. J. Int. Sci. Vigne Vin, 39 (4), 149-162.

Downey, M.O.; Dokoozlian, N.K.; and Krstic, M.P. (2006) Cultural Practice and Environmental Impacts on the

Flavonoid Composition of Grapes and Wine: A Review of Recent Research. Am. J. Enol. Vitic., 57:3, 257-268.

Hardie, W. J.; and Considine, J. A. (1976) Response of grapes to water deficit stress in particular stages of

development. Am. J. Enol. Vitic., 27, 55-61.

Hermosín Gutiérrez, I.; and García-Romero, E. (2004) Antocianos de variedades tintas cultivadas en la Mancha:

perfiles varietales característicos de la uva y de los vinos monovarietales, y evolución durante la maduración de la

baya. Alimentaria, Abril 04/127-139.

Hrazdina, G.; Parsons, G.F.; and Mattick, L.R. (1984) Physiological and biochemical events during deveopment and

maturation of grape berries. Am. J. Enol. Vitic. 35 (4), 220-227.

Jackson, J.I. & Lombard, P.B. 1993. Environmental and management practices affecting grape composition and wine

quality – A review. Am. J. Enol. Vitic. 44 (4), 409-429.

Nadal M, Mateos S, and Lampreave, M. (2008) Influence de la topographie etdu mésoclimat sur la composition des

raisins et rendement dans le terroir del’AOC Priorat, VII CongresInt. terroirs viticoles 19 - 23 mai 2008

Nyon,Suïssa,(2):590-595

Meriaux, S. (1982). La vigne et l´eau dans le Midi méditerranéen. Vignes et Vins, nº especial Septiembre: 23–26.

OIV. (1990). “Récueil des Méthodes Internacionales d’Analyse des vins et des Mouts “; Office Internacional de la

Vigne et du Vin: Paris, France.

Ribéreau-Gayon Y, Glories Y, Maujean A. & Dubourdieu D, 2000. Handbook of Enology, Volume 2: The chemistry

of wine and stabilization and treatments. John Wiley & Sons Ltd.

Río Segade, S.; Soto Vázquez, E.; and Díaz Losada, E. (2008) Influence of ripeness grade on accumulation and

extractability of grape skin anthocyanins in different cultivars. J. Food Comp. Anal., 21, 599-607.

Smart, R. 1987. Influence of light on composition and quality of grapes. Acta Hortic. 206:37-43.

Spayd, S.E.; Tarara, J.M.; Mee, D.L.; and Ferguson, J.C. (2002) Separation of sunlight and temperature effects on

the composition of Vitis vinifera cv. Merlot berries. Am. J. Enol. Vitic., 53:3, 171-182.

Tarara, J.M.; Lee, J.; Spayd, S.E.; and Scagel, C.F. (2008) Berry temperature and solar radiation alter acylation,

proportion, and concentration of anthocyanin in Merlot grapes. Am. J. Enol. Vitic., 59:3, 235-246.

Van Leeuwen, C.; Friant, P.; Choné, X.; Tregoat, O.; Koundouras, S.; and Dubourdieu, D. (2004) Influence of

climate, soil, and culrivar on terroir. Am. J. Enol. Vitic. 55:3, 207-217

3 - 57



“TERROIR” AND CLIMATE CHANGE IN FRANCONIA / GERMANY

Ulrike Maaß and Arnold Schwab

Bavarian State Institute for Viticulture and Horticulture

An der Steige 15, D-97209 Veitshöchheim

[email protected]

ABSTRACT

Franconia which is a „cool climate” winegrowing region is well known for its fruity white

wines. The most common grape cultivars are Silvaner and Mueller-Thurgau.

Franconia is a landscape of contrasts with various climatic conditions. The vineyard sites are

located at a height between 120 m and 420 m above sea level on slopes and steep slopes as

well as on terraces.

In favourable south orientated sites the maximum temperatures reach about 40° C (peak value

year 2003), while winter frosts cause deep temperatures down to about -27°C (year 2002) in

valleys or exposed sites.

At present, the Franconian winegrowing region is being affected by the global climate

change. Several forecasts predict an average annual temperature increase of approximately

2°C for Southern Germany until the year 2050. During the same period an increased

occurrence of temperature-related extreme events is expected.

In case of permanent increase of the average air temperatures and temperature-related extreme

events, the cultivation of grapes on E, W and NW slopes could be considered appropriate to

preserve the fruity character of traditional white wines.

KEYWORD

Vineyard Climate, Climate change, Terroir, Topoclimate, Microclimate

INTRODUCTION The project “Franconian vineyard climate” aims at recording local climatic peculiarities of

Franconian vineyard sites. The research is based on data of different meteorological networks

(about 30 stations, see Fig.1).

This empirical basis will serve to reveal changes of regional winegrowing conditions and to

develop adaptional measures to manage the expected climate change.

The past development of the climatic conditions of different sites will be analyzed and

projected into the future. On the basis of these results the future prospects of individual sites

and especially of former favourable sites will be considered.

Object of this research is a modified ranking of vineyard sites with respect to prospective

climatic conditions.

Furthermore the selected parameters air temperature, soil temperature, global radiation,

precipitation and others enable comparisons of the Franconian vineyard climate with that of

other German and European vineyards (Fig.2).

Fig. 1: The Franconian winegrowing area comprises an area of approximately 6,000 hectares and extends to a

length of about 130 km (beeline) from east to west between Bamberg an

The coloured spots mark different meteorological stations

Fig. 2: Meteorological station of Nordheim. The station collects

like air and soil temperature, global radiation, precipitation and wind velocity

Alzenau

Miltenberg

Aschaffenburg

Marktheidenfeld

The Franconian winegrowing area comprises an area of approximately 6,000 hectares and extends to a

length of about 130 km (beeline) from east to west between Bamberg and Alzenau along the river Main.

The coloured spots mark different meteorological stations.

: Meteorological station of Nordheim. The station collects parameters


Castell

Volkach

Kizingen

Würzburg

Röttingen

Ippesheim

Karlstadt

Zeil a. M.

Hammelburg

Gerolzhofen

Schweinfurt

Bad Windsheim

Marktheidenfeld

Rothenburg o.d. Tauber

Bamberg


d Alzenau along the river Main.

3 - 58



“TERROIR” AND CLIMATE CHANGE IN FRANCONIA / GERMANY

Ulrike Maaß and Arnold Schwab

Bavarian State Institute for Viticulture and Horticulture

An der Steige 15, D-97209 Veitshöchheim

[email protected]

ABSTRACT

Franconia which is a „cool climate” winegrowing region is well known for its fruity white

wines. The most common grape cultivars are Silvaner and Mueller-Thurgau.

Franconia is a landscape of contrasts with various climatic conditions. The vineyard sites are

located at a height between 120 m and 420 m above sea level on slopes and steep slopes as

well as on terraces.

In favourable south orientated sites the maximum temperatures reach about 40° C (peak value

year 2003), while winter frosts cause deep temperatures down to about -27°C (year 2002) in

valleys or exposed sites.

At present, the Franconian winegrowing region is being affected by the global climate

change. Several forecasts predict an average annual temperature increase of approximately

2°C for Southern Germany until the year 2050. During the same period an increased

occurrence of temperature-related extreme events is expected.

In case of permanent increase of the average air temperatures and temperature-related extreme

events, the cultivation of grapes on E, W and NW slopes could be considered appropriate to

preserve the fruity character of traditional white wines.

KEYWORD

Vineyard Climate, Climate change, Terroir, Topoclimate, Microclimate

INTRODUCTION The project “Franconian vineyard climate” aims at recording local climatic peculiarities of

Franconian vineyard sites. The research is based on data of different meteorological networks

(about 30 stations, see Fig.1).

This empirical basis will serve to reveal changes of regional winegrowing conditions and to

develop adaptional measures to manage the expected climate change.

The past development of the climatic conditions of different sites will be analyzed and

projected into the future. On the basis of these results the future prospects of individual sites

and especially of former favourable sites will be considered.

Object of this research is a modified ranking of vineyard sites with respect to prospective

climatic conditions.

Furthermore the selected parameters air temperature, soil temperature, global radiation,

precipitation and others enable comparisons of the Franconian vineyard climate with that of

other German and European vineyards (Fig.2).

Fig. 1: The Franconian winegrowing area comprises an area of approximately 6,000 hectares and extends to a

length of about 130 km (beeline) from east to west between Bamberg an

The coloured spots mark different meteorological stations

Fig. 2: Meteorological station of Nordheim. The station collects


Alzenau

Miltenberg

Aschaffenburg

Marktheidenfeld


length of about 130 km (beeline) from east to west between Bamberg and Alzenau along the river Main.

The coloured spots mark different meteorological stations.

: Meteorological station of Nordheim. The station collects parameters


Castell

Volkach

Kizingen

Würzburg

Röttingen

Ippesheim

Karlstadt

Zeil a. M.

Hammelburg

Gerolzhofen

Schweinfurt

Bad Windsheim

Marktheidenfeld

Rothenburg o.d. Tauber

Bamberg


d Alzenau along the river Main.

3 - 59



FIRST RESULTS

Fig.3 displays the permanent increase of the average air temperature of the growing period

(including months April to October) in Würzburg (DWD) since about 1980.

According to studies in Württemberg/Germany (Rupp & Kast 2010) the maximum increase in

temperature appears during the months April to June. Thus the energy gain of grapes during

growing season is much higher than 30 years ago, and ripening is shifted from

September/October to August.

Fig. 3: Average air temperatures (including months April to October)

of the meteorological station Würzburg, 1947-2009 with 11-year moving averages

Regarding temperature-related extreme events, the year 2003 with its “record summer” in

Germany has to be mentioned. Station Würzburg (DWD) recorded 15 hot days in August

2003 (30-year climate reference: 3 days!) and 6 tropical nights with daily minimum

temperature >20°C during the first half of August (30-year climate reference: 0!), on the other

hand 6 frost days in April.

Fig. 4: Number of hot days (daily maximum temperature >30°C) in August 2009,

recorded by stations of LFL (Bavarian State Institute for Agriculture) and LWG, compared with Würzburg

10

11

12

13

14

15

16

17

18

1947 1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 2002 2007

Tem

pera

ture

C

Air Temperature 11-year Moving Averages

Data source: German Weather Service (DWD)

Average air temperatures of the growing season,

Meteorological Station Würzburg

Würzburg

Würzburg, 30-year

climate ref.

Escherndorf

Nordheim

Randersacker

LWGRandersacker LFL

Iphofen LWG

Iphofen LFL

0

1

2

3

4

5

6

7

8

9

Num

ber

of h

ot d

ays

Hot days in August 2009: Comparison

SE

230 m

268 m

NE

215 m

SW

275 m

WSW

185 m

N

280 m

S

215 m

Data source: LFL and LWG stations; German Weather Service; slope orientation and

height above sea level indicated on the bars

In Fig.4 the number of hot days in August 2009 as measured by LWG stations are compared

to that of station Würzburg (DWD).

As is well known, a series of hot days during maturation influences the typical character of

local wines negatively. The stations of Randersacker (SW, WSW) measured eight hot days,

i.e. three days more than Würzburg (DWD) and nearly three times more than the 30-year

climate reference. The NE slope of Nordheim recorded only 1 hot day.

CONCLUSIONS

The Franconian winegrowing region experienced increased temperatures mainly during the

growing season within the last decades. This leads to advanced shooting and flowering of

traditional grape cultivars and thus a higher risk of damages by late frost events. On the other

hand an increased occurrence of temperature-related extreme events enhances the danger of

draught damage.

According to the German Weather Service (DWD), the latest decade 2000-2009 was the

warmest in Germany since the beginning of weather recording.

Assuming a continuous increase of the air temperatures and temperature-related extreme

events in Franconia, the cultivation of grapes on E, W and NW slopes could be considered

appropriate to preserve the fruity character of traditional white wines.

BIBLIOGRAPHY

German Weather Service, 2009, Clima Report 2008, Offenbach, 24 pages

Rupp, Kast, 2010, Cultivation Methods in a Changing Climate; oral communication

3 - 60



FIRST RESULTS

Fig.3 displays the permanent increase of the average air temperature of the growing period

(including months April to October) in Würzburg (DWD) since about 1980.

According to studies in Württemberg/Germany (Rupp & Kast 2010) the maximum increase in

temperature appears during the months April to June. Thus the energy gain of grapes during

growing season is much higher than 30 years ago, and ripening is shifted from

September/October to August.

Fig. 3: Average air temperatures (including months April to October)

of the meteorological station Würzburg, 1947-2009 with 11-year moving averages

Regarding temperature-related extreme events, the year 2003 with its “record summer” in

Germany has to be mentioned. Station Würzburg (DWD) recorded 15 hot days in August

2003 (30-year climate reference: 3 days!) and 6 tropical nights with daily minimum

temperature >20°C during the first half of August (30-year climate reference: 0!), on the other

hand 6 frost days in April.

Fig. 4: Number of hot days (daily maximum temperature >30°C) in August 2009,

recorded by stations of LFL (Bavarian State Institute for Agriculture) and LWG, compared with Würzburg

10

11

12

13

14

15

16

17

18

1947 1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 2002 2007

Tem

pera

ture

C

Air Temperature 11-year Moving Averages


Average air temperatures of the growing season,

Meteorological Station Würzburg

Würzburg

Würzburg, 30-year

climate ref.

Escherndorf

Nordheim

Randersacker

LWGRandersacker LFL

Iphofen LWG

Iphofen LFL

0

1

2

3

4

5

6

7

8

9

Num

ber

of h

ot d

ays

Hot days in August 2009: Comparison

SE

230 m

268 m

NE

215 m

SW

275 m

WSW

185 m

N

280 m

S

215 m

Data source: LFL and LWG stations; German Weather Service; slope orientation and

height above sea level indicated on the bars

In Fig.4 the number of hot days in August 2009 as measured by LWG stations are compared

to that of station Würzburg (DWD).

As is well known, a series of hot days during maturation influences the typical character of

local wines negatively. The stations of Randersacker (SW, WSW) measured eight hot days,

i.e. three days more than Würzburg (DWD) and nearly three times more than the 30-year

climate reference. The NE slope of Nordheim recorded only 1 hot day.

CONCLUSIONS

The Franconian winegrowing region experienced increased temperatures mainly during the

growing season within the last decades. This leads to advanced shooting and flowering of

traditional grape cultivars and thus a higher risk of damages by late frost events. On the other

hand an increased occurrence of temperature-related extreme events enhances the danger of

draught damage.

According to the German Weather Service (DWD), the latest decade 2000-2009 was the

warmest in Germany since the beginning of weather recording.

Assuming a continuous increase of the air temperatures and temperature-related extreme

events in Franconia, the cultivation of grapes on E, W and NW slopes could be considered

appropriate to preserve the fruity character of traditional white wines.

BIBLIOGRAPHY

German Weather Service, 2009, Clima Report 2008, Offenbach, 24 pages

Rupp, Kast, 2010, Cultivation Methods in a Changing Climate; oral communication

3 - 61



CLIMATE CHANGE – VARIETY CHANGE?

Arnold Schwab and Ulrike Maaß

Bavarian State Institute for Viticulture and Horticulture, An der Steige 15, D-97332 Veitshöchheim [email protected]; [email protected]

ABSTRACT

In Franconia, the northern part of Bavaria in Germany, climate change, visible in earlier bud break, advanced flowering and earlier grape maturity, leads to a decrease of traditionally cultivated early ripening aromatic white wine varieties as Mueller-Thurgau (30 % of the wine growing area) and Bacchus (12 %). With the predicted rise of temperature in all European wine regions the conditions for white wine grape varieties will decline and the grapes themselves will lose a part of their aromatic and fruity expression. Variety change towards the cultivation of later ripening white wine varieties is a very expensive and long-term process, and must be accompanied by special marketing efforts. In the “cool climate” region Franconia, adapted methods are required for the longer use of traditionally grown aromatic early ripening varieties. Studies about maturity management of the early ripening variety Mueller-Thurgau show first results. Cordon pruning compared with traditional spur pruned training system, leads in dependence of botrytis infection to a maturity delay of 4 up to 6 days. The new natural growth training system, also called “minimal pruning”, results in a maturity delay of 8 up to 12 days in the same varieties. Later grape harvest times economize energy for must cooling and fermentation control. Lower night temperatures can better conserve the fresh and fruity flavours of these aromatic grapes. The consequences of maturity retardation effects on must and wine quality will be studied.

KEYWORD: Climate change, Franconia, earlier harvest time, variety change, canopy management

Introduction

Franconia´s wine region along the river Main, has a mainly continental climate with a yearly mean temperature of about 10,0 ° C and a yearly precipitation of about 550 mm. The vineyard area of about 6000 ha covers the slopes and steep slopes along the river mainly in S-SW exposition. White wine varieties are grown on 80 % of the area consisting of Mueller -Thurgau (28 %) and Bacchus ((Riesling x Silvaner) x Mueller-Thurgau) (12%). The later ripening Silvaner is Franconia´s most famos variety and covers 21 %, the late ripening Riesling only 5 %. The residual area is planted with minor white wine varieties. 20 % of the area is covered with red wine varieties as for example Pinot Noir (4,3 %) and Domina (Portugieser x Pinot Noir) 5,7 %. Franconia is well known for its fresh and fruity white wines as Mueller-Thurgau and Bacchus, and for the soft and creamy Silvaner. With the climate change the early ripening varieties will diminish more and more.

Signs of climate change

As shown in Figure 1, the average temperature has been increasing in our region (Würzburg) since 1950. Compared with the Rhine valley (Rhingau region/Geisenheim) both datalines show the same rising development. Because of the increasing temperatures, earlier bud break (8 days), earlier flowering (8 days) and earlier veraison (11 days) is documented for Mueller-Thurgau in Franconia (Hönig & Schwappach, 2003). Earlier harvest time is associated with a higher risk of fungus

infection as Botrytis cinerea, an augmention of acetic acid, a faster reduction of total acid because of high night temperatures and a rising demand for cooling energy.

Figure 1: Development of the Huglin-Index for the region Franconia and Rhingau since 1947 and results of the mean yearly regional Franconian must weight (°Oechsle) since 1970

An increased cultivation of late ripening varieties for a longer utilization of the whole possible ripening period in September and October is in the phase of investigation. The introduction of new later ripening varieties will influence or even change the special regional wine character. This change needs a comprehensive discussion and a careful and integrated transfer.

Variety needs in temperature sums - the Huglin-Index (HI)

Table 1 shows a comparison of the different grape variety needs in temperature sums during the vegetation period by means of the Huglin-Index. The Huglin-Index (HI) adds up all daily mean temperatures above 10° C in the vegetation period by means of a special formula. The temperature requirements to reach a good maturity level for the main German and European grape varieties are summarized in Table 1.

Huglin-Index German grape varieties selected European grape varieties

1300 Siegerrebe, Ortega, Reberger 1400 Mueller-Thurgau, Bacchus, 1500 Kerner, Portugieser Gamay

1600 Silvaner, Grauburgunder, Schwarzriesling Chasselas, Pinot Meunier

1700 Weissburgunder, Sauvignon Blanc, Spätburgunder

Sauvignon Blanc, Pinot Noir Grüner Veltliner,

1800 Riesling, Scheurebe, Gewürztraminer, Chardonnay, 1900 Muskateller, Trollinger, Blaufränkisch Merlot, Syrah, Viognier 2000 Cabernet Cubin, Cabernet Sauvignon,Tempranillo 2100 Grenache, Cinsault, Sangiovese 2200 Carignan, Trebbiano, Airen 2300 Nebbiolo,

Table 1: Huglin-Index – for German and European grape varieties (preliminary classification)

As presented in Figure 2, the calculated HI has sightly increased in our vineyard region of Franconia/Würzburg in the last decades. A prognostic estimation of this development indicates a rapid change in the ripening period and requires a suitable reaction in direction to variety change.

y = 2.6971x + 1448R² = 0.0602

0

10

20

30

40

50

60

70

80

90

100

0

500

1.000

1.500

2.000

2.500

1947 1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 2002 2007

[°Oe]HuglinIndex

Huglin Index Würzburg Huglin Index GeisenheimMust weight linear trend (Würzburg)

~1450~1600

Data Source: German Weather Service (DWD); Bavarian State Office for Statistics and Data processing

3 - 62



CLIMATE CHANGE – VARIETY CHANGE?

Arnold Schwab and Ulrike Maaß

Bavarian State Institute for Viticulture and Horticulture, An der Steige 15, D-97332 Veitshöchheim [email protected]; [email protected]

ABSTRACT

In Franconia, the northern part of Bavaria in Germany, climate change, visible in earlier bud break, advanced flowering and earlier grape maturity, leads to a decrease of traditionally cultivated early ripening aromatic white wine varieties as Mueller-Thurgau (30 % of the wine growing area) and Bacchus (12 %). With the predicted rise of temperature in all European wine regions the conditions for white wine grape varieties will decline and the grapes themselves will lose a part of their aromatic and fruity expression. Variety change towards the cultivation of later ripening white wine varieties is a very expensive and long-term process, and must be accompanied by special marketing efforts. In the “cool climate” region Franconia, adapted methods are required for the longer use of traditionally grown aromatic early ripening varieties. Studies about maturity management of the early ripening variety Mueller-Thurgau show first results. Cordon pruning compared with traditional spur pruned training system, leads in dependence of botrytis infection to a maturity delay of 4 up to 6 days. The new natural growth training system, also called “minimal pruning”, results in a maturity delay of 8 up to 12 days in the same varieties. Later grape harvest times economize energy for must cooling and fermentation control. Lower night temperatures can better conserve the fresh and fruity flavours of these aromatic grapes. The consequences of maturity retardation effects on must and wine quality will be studied.

KEYWORD: Climate change, Franconia, earlier harvest time, variety change, canopy management

Introduction

Franconia´s wine region along the river Main, has a mainly continental climate with a yearly mean temperature of about 10,0 ° C and a yearly precipitation of about 550 mm. The vineyard area of about 6000 ha covers the slopes and steep slopes along the river mainly in S-SW exposition. White wine varieties are grown on 80 % of the area consisting of Mueller -Thurgau (28 %) and Bacchus ((Riesling x Silvaner) x Mueller-Thurgau) (12%). The later ripening Silvaner is Franconia´s most famos variety and covers 21 %, the late ripening Riesling only 5 %. The residual area is planted with minor white wine varieties. 20 % of the area is covered with red wine varieties as for example Pinot Noir (4,3 %) and Domina (Portugieser x Pinot Noir) 5,7 %. Franconia is well known for its fresh and fruity white wines as Mueller-Thurgau and Bacchus, and for the soft and creamy Silvaner. With the climate change the early ripening varieties will diminish more and more.

Signs of climate change

As shown in Figure 1, the average temperature has been increasing in our region (Würzburg) since 1950. Compared with the Rhine valley (Rhingau region/Geisenheim) both datalines show the same rising development. Because of the increasing temperatures, earlier bud break (8 days), earlier flowering (8 days) and earlier veraison (11 days) is documented for Mueller-Thurgau in Franconia (Hönig & Schwappach, 2003). Earlier harvest time is associated with a higher risk of fungus

infection as Botrytis cinerea, an augmention of acetic acid, a faster reduction of total acid because of high night temperatures and a rising demand for cooling energy.

Figure 1: Development of the Huglin-Index for the region Franconia and Rhingau since 1947 and results of the mean yearly regional Franconian must weight (°Oechsle) since 1970

An increased cultivation of late ripening varieties for a longer utilization of the whole possible ripening period in September and October is in the phase of investigation. The introduction of new later ripening varieties will influence or even change the special regional wine character. This change needs a comprehensive discussion and a careful and integrated transfer.

Variety needs in temperature sums - the Huglin-Index (HI)

Table 1 shows a comparison of the different grape variety needs in temperature sums during the vegetation period by means of the Huglin-Index. The Huglin-Index (HI) adds up all daily mean temperatures above 10° C in the vegetation period by means of a special formula. The temperature requirements to reach a good maturity level for the main German and European grape varieties are summarized in Table 1.

Huglin-Index German grape varieties selected European grape varieties

1300 Siegerrebe, Ortega, Reberger 1400 Mueller-Thurgau, Bacchus, 1500 Kerner, Portugieser Gamay

1600 Silvaner, Grauburgunder, Schwarzriesling Chasselas, Pinot Meunier

1700 Weissburgunder, Sauvignon Blanc, Spätburgunder

Sauvignon Blanc, Pinot Noir Grüner Veltliner,

1800 Riesling, Scheurebe, Gewürztraminer, Chardonnay, 1900 Muskateller, Trollinger, Blaufränkisch Merlot, Syrah, Viognier 2000 Cabernet Cubin, Cabernet Sauvignon,Tempranillo 2100 Grenache, Cinsault, Sangiovese 2200 Carignan, Trebbiano, Airen 2300 Nebbiolo,

Table 1: Huglin-Index – for German and European grape varieties (preliminary classification)

As presented in Figure 2, the calculated HI has sightly increased in our vineyard region of Franconia/Würzburg in the last decades. A prognostic estimation of this development indicates a rapid change in the ripening period and requires a suitable reaction in direction to variety change.

y = 2.6971x + 1448R² = 0.0602

0

10

20

30

40

50

60

70

80

90

100

0

500

1.000

1.500

2.000

2.500

1947 1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 2002 2007

[°Oe]HuglinIndex

Huglin Index Würzburg Huglin Index GeisenheimMust weight linear trend (Würzburg)

~1450~1600

Data Source: German Weather Service (DWD); Bavarian State Office for Statistics and Data processing

3 - 63



Figure 2: Huglin-Index – Development of HI in decades in the Franconian region since 1950

Besides Riesling, Chardonnay and Sauvignon Blanc other late ripening European white wine varieties as Viognier, Muscat, and new German crossings are tested to improve the existing variety panel. Adapted measures are necessary to manage the present vineyards with early ripening varieties during the next decades.

Experimental results

In addition to variety change, clonal selection and canopy management can result in a better adaptation to higher temperatures during the ripening period. First experiments with canopy reduction under our climate conditions showed negative results in the wine sensoric. By means of natural growing/minimal pruning acceptable ripening delay were obtained in the varieties Mueller-Thurgau and Bacchus. Aromatic and fruity, alcohol balanced wines were the result. Cordon pruning with one bud cuttings delayed the ripening period only for 4 up to 6 days, is however accompanied by a strong and not always desired yield reduction.

Bibliography

Hönig, Petra and Peter Schwappach, 2003: Klimaänderung: Wie reagiert die Rebe?; Rebe und Wein, Nr. 11, p. 23-25

13

13,5

14

14,5

15

15,5

1300

1400

1500

1600

1700

1800

1950-1959 1960-1969 1970-1979 1980-1989 1990-1999 2000-2009

Huglin IndexAir temperature (April-October)


Tem

pera

ture

°C

Hug

lin I

ndex

CLIMATE CHANGE IMPACT STUDY BASED ON GRAPEVINEPHENOLOGY MODELLING

M. Ladányi(1), E. Hlaszny(2), Gy. Pernesz(3), Gy. Bisztray(2)

(1)Corvinus Univ. of Budapest, Dpt. of Mathematics and Informatics, Villányi út 29-43. H-1118, Budapest, [email protected]

(2)Corvinus Univ. of Budapest, Dpt. Of Viticulture, Villányi út 29-43. H-1118, Budapest, [email protected]

Central Agricultural Office; Budapest, [email protected]

[email protected]

ABSTRACTIn this work we present a joint model of calculation the budbreak and full bloom starting dates

which considers the heat sums and allows reliable estimations for five white wine grape varieties(Chardonnay, Szürkebarát (Pinot gris), Pinot blanc, Riesling, Hárslevelű) and their clone varieties in Hungary (Chardonnay 75 and 96, Riesling 239, 378, 391 and 49, Hárslevelű P.41 and K.9., Pinot blanc 54, 55 and D55, Szürkebarát 34 and 52). The base lower and upper temperatures havebeen determined by optimization, above which (threshold temperature) the accumulation of dailymeans is most active, or alternatively, below which the daily means are most sensitivelyexpressed in the phenology. The model has been extended to the calculation of the end of the restperiod (endodormancy), by optimization as well. We determined the lower and upper basetemperatures separately for the budbreak and full bloom starting dates such that the lowest(normalized) sum of squares error, the lowest average absolute and the lowest maximum error ofpredictions can be achieved. We determined the optimal (lower) base temperature as 6 °C and theoptimal starting date as the 41st Julian day of the year for the budbreak. Moreover, we set 10,45°C and 26 °C as lower and upper optimal base temperatures for full bloom. The joint model wasthen applied to study the impact of climate change on budbreak and full bloom starting datesbased on RegCM3.1 (regional) climate model. We calculated the expected shifts of budbreak andfull bloom and proved that the changes are significant.

KEYWORDSbudbreak – vegetation period – phenology model – biologically effective day degrees – full

bloom – starting dates of phenological stages – Vitis vinifera L.

INTRODUCTIONThe success of viticultural production depends highly on weather parameters. The effects of

climate change are already visible in the phenology of several varieties. Modelling the startingdates of budbreak and full bloom is very important because the success of plant protection andtechnology techniques scheduling is depending mainly on phenological information. Moreover,several risk factors can be traced back to the connection of weather and phenological timing. Thiskind of research is of even greater importance nowadays when usual phenological timing ischanging.

3 - 64



Figure 2: Huglin-Index – Development of HI in decades in the Franconian region since 1950

Besides Riesling, Chardonnay and Sauvignon Blanc other late ripening European white wine varieties as Viognier, Muscat, and new German crossings are tested to improve the existing variety panel. Adapted measures are necessary to manage the present vineyards with early ripening varieties during the next decades.

Experimental results

In addition to variety change, clonal selection and canopy management can result in a better adaptation to higher temperatures during the ripening period. First experiments with canopy reduction under our climate conditions showed negative results in the wine sensoric. By means of natural growing/minimal pruning acceptable ripening delay were obtained in the varieties Mueller-Thurgau and Bacchus. Aromatic and fruity, alcohol balanced wines were the result. Cordon pruning with one bud cuttings delayed the ripening period only for 4 up to 6 days, is however accompanied by a strong and not always desired yield reduction.

Bibliography

Hönig, Petra and Peter Schwappach, 2003: Klimaänderung: Wie reagiert die Rebe?; Rebe und Wein, Nr. 11, p. 23-25

13

13,5

14

14,5

15

15,5

1300

1400

1500

1600

1700

1800

1950-1959 1960-1969 1970-1979 1980-1989 1990-1999 2000-2009

Huglin IndexAir temperature (April-October)


Tem

pera

ture

°C

Hug

lin I

ndex

CLIMATE CHANGE IMPACT STUDY BASED ON GRAPEVINEPHENOLOGY MODELLING

M. Ladányi(1), E. Hlaszny(2), Gy. Pernesz(3), Gy. Bisztray(2)

(1)Corvinus Univ. of Budapest, Dpt. of Mathematics and Informatics, Villányi út 29-43. H-1118, Budapest, [email protected]

(2)Corvinus Univ. of Budapest, Dpt. Of Viticulture, Villányi út 29-43. H-1118, Budapest, [email protected]

Central Agricultural Office; Budapest, [email protected]

[email protected]

ABSTRACTIn this work we present a joint model of calculation the budbreak and full bloom starting dates

which considers the heat sums and allows reliable estimations for five white wine grape varieties(Chardonnay, Szürkebarát (Pinot gris), Pinot blanc, Riesling, Hárslevelű) and their clone varieties in Hungary (Chardonnay 75 and 96, Riesling 239, 378, 391 and 49, Hárslevelű P.41 and K.9., Pinot blanc 54, 55 and D55, Szürkebarát 34 and 52). The base lower and upper temperatures havebeen determined by optimization, above which (threshold temperature) the accumulation of dailymeans is most active, or alternatively, below which the daily means are most sensitivelyexpressed in the phenology. The model has been extended to the calculation of the end of the restperiod (endodormancy), by optimization as well. We determined the lower and upper basetemperatures separately for the budbreak and full bloom starting dates such that the lowest(normalized) sum of squares error, the lowest average absolute and the lowest maximum error ofpredictions can be achieved. We determined the optimal (lower) base temperature as 6 °C and theoptimal starting date as the 41st Julian day of the year for the budbreak. Moreover, we set 10,45°C and 26 °C as lower and upper optimal base temperatures for full bloom. The joint model wasthen applied to study the impact of climate change on budbreak and full bloom starting datesbased on RegCM3.1 (regional) climate model. We calculated the expected shifts of budbreak andfull bloom and proved that the changes are significant.

KEYWORDSbudbreak – vegetation period – phenology model – biologically effective day degrees – full

bloom – starting dates of phenological stages – Vitis vinifera L.

INTRODUCTIONThe success of viticultural production depends highly on weather parameters. The effects of

climate change are already visible in the phenology of several varieties. Modelling the startingdates of budbreak and full bloom is very important because the success of plant protection andtechnology techniques scheduling is depending mainly on phenological information. Moreover,several risk factors can be traced back to the connection of weather and phenological timing. Thiskind of research is of even greater importance nowadays when usual phenological timing ischanging.

3 - 65



The budbreak date is the first phase of vegetation period. It depends on weather parameters(like soil and air temperatures, thermo parameters of winter and spring), variety, physiologicalstage of the vine-stock, maturity of buds, etc. The budbreak starts off, when the necessary(critical) biologically effective heat sum is reached. The beginning of budbreak was recorded,when the broken buds reach the proportion of 50 %.

The blooming of grapevine usually occurs in Hungary between end of May and middle of June.However, climate change and weather anomalies in the last decades and in the future may causethe change of phenology timing. First of all, temperature and relative air humidity define thebeginning of blooming. There are appreciable deviations between blooming periods of Vitisvinifera L., North-American or East-Asian species, and between the early and late ripeningvarieties. The ideal temperature for grapevine blooming is between 20°C and 26°C. Duringbloom dry weather with low air humidity is unfavourable as well as heavy rainfall isdisadvantageous. We set the beginning of bloom when 4-5 % of grapevine flowers opened. Thefull bloom is defined by 60-70 % of flowers opened.

The starting dates of budbreak and full bloom are investigated with a biologically effective daydegreee joint model which depends on lower and upper base temperatures and also on the startingdate of heat accumulation.With the help of the phenology model as well as of the RegCM3.1regional climate model the expected shifts of budbreak and full bloom are calculated.

MATERIALS AND METHODSThe phenology data of Central Agricultural Office (CAO) we used were recorded between 2000

and 2004 in Helvécia (South Great Hungarian Plain Region, Hungary), when budbreak and fullbloom periods started. Five white wine grape varieties (Chardonnay, Szürkebarát (Pinot gris),Pinot blanc, Riesling, Hárslevelű) and their clone varieties (Chardonnay 75 and 96, Riesling 239, 378, 391 and 49, Hárslevelű P.41 and K.9., Pinot blanc 54, 55 and D55, Szürkebarát 34 and 52) were investigated. In this region the soil is sandy with very low humus content (Pernesz,2004).The number of yearly sunny hours is between 2000-2500 hours and the vegetation periodis highly variable (Fig. 1). The main risk factors are frost and drought.

0 50 100 150 200 250 300 350Julian day

Vegetation periods

2004

2003

2002

2001

2000

Figure 1 The lengths of the vegetation periods of the years 2000-2004 in Helvécia

A phenology model for the estimation of budbreak and full bloom dates

The method of calculating the sum of daily mean temperatures as “degree days”, is based on theobservation that the plants are able to utilise cumulatively - in growth and development - thetemperature above a lower and under an upper base temperature (Tomasi et al., 2005).

For grapevines (Vitis vinifera L.), 10 °C is widely accepted as (lower) base temperature (Jones,2003, Jones et al., 2005). However, we decided to calculate the base temperatures of grapevinewith optimization method for budbreak and flowering starting dates separately. The optimizationwas based on the least standard deviation in days as well as on the least average absolutedeviation in days and on the least maximum deviation in days. The thermal time wasaccumulated from the average daily temperature above the lower base temperature and, in case offlowering starting date estimation, with a ceiling of the upper base temperature if the averageexceeded it. Though the most widely used starting date of thermal accumulation for budbreakdate models is the 1st of January (Riou, 1994, Bindi et al., 1997 a,b,), after optimization we havechosen a later starting date which has improved our budbreak date estimation. The optimizedstarting date can be considered as the statistical end of endodormancy (the period when buds aredormant due to physiological conditions) and the starting date of ecodormancy (when budsremain dormant just because of unfavourable environmental conditions (Lang, 1987, Cesaraccio,2004). Judging by the quantity and quality of the available data we decided to use a daily scaledlinear model.

The joint model was then applied to study the impact of climate change on budbreak and fullbloom starting dates. To this we took the RegCM3.1 (regional) climate model with 10 kmresolution referring to 2021-2050 and with reference period 1961-90, supposed the SRESscenario A1B. The original climate change model was developed by Giorgi al. (1993) and wasdownscaled at Eötvös Loránd University, Department of Meteorology, Budapest, Hungary(Bartholy et. al., 2009, Torma et. al., 2008). We calculated the expected shifts of budbreak andfull bloom and proved that the changes are significant.

RESULTS AND DISCUSSIONWe determined the optimal (lower) base temperature as 6 °C and the optimal starting date as

the 41st Julian day of the year for the budbreak. (It means that the statistically calculated date ofthe end of the endodormancy is the 10th of February.) The optima, however, has not changedwhen the upper base temperature was built in the model. The reason of it is that there were nosuch high average daily temperature between 10th of February and budbreak in these years whichcould significantly change the value of degree days. It means that between the end ofendodormancy and budbreak the heat was as high as the plant could almost totally benefit it.

The optima are corresponding to the ones in the literature based on physiological reasons(Gladstones, 2000). Table 1 represents the accumulated heat sums (°C) of the different varietiesin the time period 2000-2004. The critical values of heat sums (°C) for the budbreak of thefifteen white wine varieties during the five years 2000-2004 are also summarized in Table 1where the average values for the certain varieties can be seen below, in separate lines. We cansee that the varieties needed the less heat sum in 2003, the most heat sum in 2004 for theirbudbreak.

3 - 66



The budbreak date is the first phase of vegetation period. It depends on weather parameters(like soil and air temperatures, thermo parameters of winter and spring), variety, physiologicalstage of the vine-stock, maturity of buds, etc. The budbreak starts off, when the necessary(critical) biologically effective heat sum is reached. The beginning of budbreak was recorded,when the broken buds reach the proportion of 50 %.

The blooming of grapevine usually occurs in Hungary between end of May and middle of June.However, climate change and weather anomalies in the last decades and in the future may causethe change of phenology timing. First of all, temperature and relative air humidity define thebeginning of blooming. There are appreciable deviations between blooming periods of Vitisvinifera L., North-American or East-Asian species, and between the early and late ripeningvarieties. The ideal temperature for grapevine blooming is between 20°C and 26°C. Duringbloom dry weather with low air humidity is unfavourable as well as heavy rainfall isdisadvantageous. We set the beginning of bloom when 4-5 % of grapevine flowers opened. Thefull bloom is defined by 60-70 % of flowers opened.

The starting dates of budbreak and full bloom are investigated with a biologically effective daydegreee joint model which depends on lower and upper base temperatures and also on the startingdate of heat accumulation.With the help of the phenology model as well as of the RegCM3.1regional climate model the expected shifts of budbreak and full bloom are calculated.

MATERIALS AND METHODSThe phenology data of Central Agricultural Office (CAO) we used were recorded between 2000

and 2004 in Helvécia (South Great Hungarian Plain Region, Hungary), when budbreak and fullbloom periods started. Five white wine grape varieties (Chardonnay, Szürkebarát (Pinot gris),Pinot blanc, Riesling, Hárslevelű) and their clone varieties (Chardonnay 75 and 96, Riesling 239, 378, 391 and 49, Hárslevelű P.41 and K.9., Pinot blanc 54, 55 and D55, Szürkebarát 34 and 52) were investigated. In this region the soil is sandy with very low humus content (Pernesz,2004).The number of yearly sunny hours is between 2000-2500 hours and the vegetation periodis highly variable (Fig. 1). The main risk factors are frost and drought.

0 50 100 150 200 250 300 350Julian day

Vegetation periods

2004

2003

2002

2001

2000

Figure 1 The lengths of the vegetation periods of the years 2000-2004 in Helvécia

A phenology model for the estimation of budbreak and full bloom dates

The method of calculating the sum of daily mean temperatures as “degree days”, is based on theobservation that the plants are able to utilise cumulatively - in growth and development - thetemperature above a lower and under an upper base temperature (Tomasi et al., 2005).

For grapevines (Vitis vinifera L.), 10 °C is widely accepted as (lower) base temperature (Jones,2003, Jones et al., 2005). However, we decided to calculate the base temperatures of grapevinewith optimization method for budbreak and flowering starting dates separately. The optimizationwas based on the least standard deviation in days as well as on the least average absolutedeviation in days and on the least maximum deviation in days. The thermal time wasaccumulated from the average daily temperature above the lower base temperature and, in case offlowering starting date estimation, with a ceiling of the upper base temperature if the averageexceeded it. Though the most widely used starting date of thermal accumulation for budbreakdate models is the 1st of January (Riou, 1994, Bindi et al., 1997 a,b,), after optimization we havechosen a later starting date which has improved our budbreak date estimation. The optimizedstarting date can be considered as the statistical end of endodormancy (the period when buds aredormant due to physiological conditions) and the starting date of ecodormancy (when budsremain dormant just because of unfavourable environmental conditions (Lang, 1987, Cesaraccio,2004). Judging by the quantity and quality of the available data we decided to use a daily scaledlinear model.

The joint model was then applied to study the impact of climate change on budbreak and fullbloom starting dates. To this we took the RegCM3.1 (regional) climate model with 10 kmresolution referring to 2021-2050 and with reference period 1961-90, supposed the SRESscenario A1B. The original climate change model was developed by Giorgi al. (1993) and wasdownscaled at Eötvös Loránd University, Department of Meteorology, Budapest, Hungary(Bartholy et. al., 2009, Torma et. al., 2008). We calculated the expected shifts of budbreak andfull bloom and proved that the changes are significant.

RESULTS AND DISCUSSIONWe determined the optimal (lower) base temperature as 6 °C and the optimal starting date as

the 41st Julian day of the year for the budbreak. (It means that the statistically calculated date ofthe end of the endodormancy is the 10th of February.) The optima, however, has not changedwhen the upper base temperature was built in the model. The reason of it is that there were nosuch high average daily temperature between 10th of February and budbreak in these years whichcould significantly change the value of degree days. It means that between the end ofendodormancy and budbreak the heat was as high as the plant could almost totally benefit it.

The optima are corresponding to the ones in the literature based on physiological reasons(Gladstones, 2000). Table 1 represents the accumulated heat sums (°C) of the different varietiesin the time period 2000-2004. The critical values of heat sums (°C) for the budbreak of thefifteen white wine varieties during the five years 2000-2004 are also summarized in Table 1where the average values for the certain varieties can be seen below, in separate lines. We cansee that the varieties needed the less heat sum in 2003, the most heat sum in 2004 for theirbudbreak.

3 - 67



Table 1 The accumulated observed heat sums (°C) above the (lower) base temperature of 6 °Cfor different varieties in the time period 2000-2004 together with their averages (called critical

heat sums)Varieties

YearsCh Ch_75 Ch_96 Szb Szb_34 Szb_52 Pb_54 Pb_55 Pb_D55 Rr_239 Rr_378 Rr_391 Rr_49 Hl_P41 Hl_K9 Average

2000 160,75 160,75 188,75 238,25 238,25 238,25 188,75 238,25 238,25 238,25 238,25 238,25 238,25 238,25 238,25 221,322001 204,25 216,75 204,25 234,50 224,75 246,00 195,25 204,25 204,25 234,50 246,00 246,00 246,00 234,50 234,50 225,052002 202,00 202,00 208,00 245,50 245,50 256,50 202,00 208,00 193,00 214,50 221,50 208,00 230,00 221,50 221,50 218,632003 160,50 160,50 160,50 182,00 199,50 182,00 182,00 182,00 171,00 182,00 182,00 182,00 182,00 182,00 171,00 177,402004 215,00 223,50 223,50 201,00 223,50 215,00 207,00 234,50 223,50 278,00 266,50 266,50 223,50 244,00 215,00 230,67

Average 188,50 192,70 197,00 220,25 226,30 227,55 195,00 213,40 206,00 229,45 230,85 228,15 223,95 224,05 216,05

Table 2 The errors of the estimations of budbreak

Ch Ch_75 Ch_96 Szb Szb_34 Szb_52 Pb_54 Pb_55 Pb_D55 Rr_239 Rr_378 Rr_391 Rr_49 Hl_P41 Hl_K9

Yearlyaverage

of theabsolutevalues

2000 2 3 1 -1 0 0 1 -1 -2 0 0 0 0 0 -1 0,802001 -2 -3 0 -1 1 -1 0 2 1 0 -1 -1 -2 -1 -2 1,202002 -1 -1 -1 -3 -2 -3 0 1 2 2 1 3 0 1 0 1,402003 3 3 3 3 2 3 1 3 3 3 4 3 3 3 4 2,932004 -3 -4 -3 3 1 2 -1 -2 -2 -4 -3 -3 1 -1 1 2,27

Averageof the

absolutevalues

2,20 2,80 1,60 2,20 1,20 1,80 0,60 1,80 2,00 1,80 1,80 2,00 1,20 1,20 1,60 1,72

The (normalized) sum of squares error, the average (absolute) error and the maximal error are2,09, 1,72 and 4 days for budbreak. Considering Table 2 we can see that the budbreak dates ofvariety Chardonnay_75 (Ch_75) can be predicted with the largest average absolute error (2,8days). The model estimations gave the smallest errors for the clone variety Pinot blanc 54(Pb_54), its average absolute error was 0,6 day. The less predictable year was 2003 with anaverage absolute error of 2,93 days. The averages of the absolute errors of the years 2000-2002were all under 2 days.

Moreover, we set 10,5 °C and 26,5 °C as lower and upper optimal base temperatures for fullbloom. The (normalized) sum of squares error, the average (absolute) error and the maximal errorare 2,22, 1,76 and 6 days for full bloom.

Table 3 represents the accumulated heat sums (°C) of the different varieties in the time period2000-2004 while Table 4 shows the errors of the estimations (days).

Table 3 The accumulated observed heat sums (°C) of the different varieties between budbreakand bloom in the time period 2000-2004 together with their averages (called critical heat sums)

VarietiesYears

Ch Ch_75 Ch_96 Szb Szb_34 Szb_52 Pb_54 Pb_55 Pb_D55 Rr_239 Rr_378 Rr_391 Rr_49 Hl_P41 Hl_K9 Average

2000 273,20 262,60 262,60 253,50 245,90 240,80 262,60 253,50 256,00 245,90 247,50 245,90 245,90 298,50 306,10 260,032001 247,05 247,05 240,45 248,55 248,05 244,95 254,75 250,15 252,25 244,95 248,05 244,95 251,65 302,15 291,15 254,412002 209,70 285,90 252,50 292,40 267,90 267,90 284,90 278,70 267,60 274,00 263,80 267,90 274,00 285,70 288,30 270,752003 248,78 248,78 248,78 219,35 227,58 227,58 248,78 227,58 235,68 227,58 213,98 227,58 227,58 219,35 219,35 231,222004 268,60 268,60 267,00 235,00 220,30 208,20 242,30 250,70 240,20 255,60 255,60 270,80 255,60 279,90 247,10 251,03

Average 249,47 262,59 254,27 249,76 241,95 237,89 258,67 252,13 250,35 249,61 245,79 251,43 250,95 277,12 270,40

Table 4 The deviations of the estimations from the observed dates of full bloomVarieties

YearsCh Ch_75 Ch_96 Szb Szb_34 Szb_52 Pb_54 Pb_55 Pb_D55 Rr_239 Rr_378 Rr_391 Rr_49 Hl_P41 Hl_K9

Yearlyaverage


2000 -2 0 -1 0 0 0 -1 0 -1 2 -1 2 2 -1 -2 1,002001 1 5 3 1 -2 -2 2 1 0 2 0 2 0 -1 -1 1,532002 4 -2 1 -6 -2 -3 -3 -3 -1 -3 -2 -1 -3 -1 -3 2,532003 1 2 1 3 2 1 1 3 2 2 3 3 3 6 5 2,532004 -1 0 -1 2 2 3 2 1 1 0 0 -2 0 0 3 1,20

Averageof the

absolutevalues

1,8 1,8 1,4 2,4 1,6 1,8 1,8 1,6 1 1,8 1,2 2 1,6 1,8 2,8 1,76

Analyzing the deviations of the estimations from the observed dates (Table 4) we can see thatthe starting date of full bloom of Hárslevelű K.9 clone was the most difficult to forecast. The most varieties however, have their absolute error near the average (1,76 days) which indicatesthe relative high stability of the model. The least average absolute error of the model wasresulted for the clone variety Pinot blanc D55 (1 day).

105

110

115

120

125

2000 2001 2002 2003 2004

day

year

Chardonnay

Ch__o

Ch__p

Ch_75_o

Ch_75_p

Ch_96_o

Ch_96_p

105

130

155

2000 2001 2002 2003 2004

day

year

Chardonnay

Ch__o

Ch__p

Ch_75_o

Ch_75_p

Ch_96_o

Ch_96_p

105

110

115

120

125

2000 2001 2002 2003 2004

day

year

Szürkebarát (Pinot gris)

Szb_o

Szb_p

Szb34_o

Szb34_p

Szb52_o

Szb52_p

105

130

155

2000 2001 2002 2003 2004

day

year


Szb_o

Szb_p

Szb34_o

Szb34_p

Szb52_o

Szb52_p

105

110

115

120

125

2000 2001 2002 2003 2004

day

year

Pinot blanc

Pb54_o

Pb54_p

Pb55_o

Pb55_p

Pb_D55_o

Pb_D55_p

105

130

155

2000 2001 2002 2003 2004

day

year

Pinot blanc

Pb54_o

Pb54_p

Pb55_o

Pb55_p

Pb_D55_o

Pb_D55_p

105

130

155

2000 2001 2002 2003 2004

day

year

Rizling (Riesling)

Rr239_o

Rr_239_p

Rr__378_o

Rr_378_p

Rr_391_o

Rr_391_p

Rr49_o

Rr49_p

105

130

155

2000 2001 2002 2003 2004

day

year

Rizling (Riesling)

Rr239_o

Rr_239_p

Rr__378_o

Rr_378_p

Rr_391_o

Rr_391_p

Rr49_o

Rr49_p

105

110

115

120

125

2000 2001 2002 2003 2004

day

year

Hárslevelű

Hl_P41_o

Hl_P41_p

Hl_K9_o

Hl_K9_p

105

130

155

2000 2001 2002 2003 2004

day

year

Hárslevelű

Hl_P41_o

Hl_P41_p

Hl_K9_o

Hl_K9_p

Figure 2 The observed (_o) and the predicted (_p) bud break (left) and full blooming (right)starting dates of the five white vine varieties in the time period 2000-2004.

3 - 68



Table 1 The accumulated observed heat sums (°C) above the (lower) base temperature of 6 °Cfor different varieties in the time period 2000-2004 together with their averages (called critical

heat sums)Varieties

YearsCh Ch_75 Ch_96 Szb Szb_34 Szb_52 Pb_54 Pb_55 Pb_D55 Rr_239 Rr_378 Rr_391 Rr_49 Hl_P41 Hl_K9 Average

2000 160,75 160,75 188,75 238,25 238,25 238,25 188,75 238,25 238,25 238,25 238,25 238,25 238,25 238,25 238,25 221,322001 204,25 216,75 204,25 234,50 224,75 246,00 195,25 204,25 204,25 234,50 246,00 246,00 246,00 234,50 234,50 225,052002 202,00 202,00 208,00 245,50 245,50 256,50 202,00 208,00 193,00 214,50 221,50 208,00 230,00 221,50 221,50 218,632003 160,50 160,50 160,50 182,00 199,50 182,00 182,00 182,00 171,00 182,00 182,00 182,00 182,00 182,00 171,00 177,402004 215,00 223,50 223,50 201,00 223,50 215,00 207,00 234,50 223,50 278,00 266,50 266,50 223,50 244,00 215,00 230,67

Average 188,50 192,70 197,00 220,25 226,30 227,55 195,00 213,40 206,00 229,45 230,85 228,15 223,95 224,05 216,05

Table 2 The errors of the estimations of budbreak

Ch Ch_75 Ch_96 Szb Szb_34 Szb_52 Pb_54 Pb_55 Pb_D55 Rr_239 Rr_378 Rr_391 Rr_49 Hl_P41 Hl_K9

Yearlyaverage


2000 2 3 1 -1 0 0 1 -1 -2 0 0 0 0 0 -1 0,802001 -2 -3 0 -1 1 -1 0 2 1 0 -1 -1 -2 -1 -2 1,202002 -1 -1 -1 -3 -2 -3 0 1 2 2 1 3 0 1 0 1,402003 3 3 3 3 2 3 1 3 3 3 4 3 3 3 4 2,932004 -3 -4 -3 3 1 2 -1 -2 -2 -4 -3 -3 1 -1 1 2,27

Averageof the

absolutevalues

2,20 2,80 1,60 2,20 1,20 1,80 0,60 1,80 2,00 1,80 1,80 2,00 1,20 1,20 1,60 1,72

The (normalized) sum of squares error, the average (absolute) error and the maximal error are2,09, 1,72 and 4 days for budbreak. Considering Table 2 we can see that the budbreak dates ofvariety Chardonnay_75 (Ch_75) can be predicted with the largest average absolute error (2,8days). The model estimations gave the smallest errors for the clone variety Pinot blanc 54(Pb_54), its average absolute error was 0,6 day. The less predictable year was 2003 with anaverage absolute error of 2,93 days. The averages of the absolute errors of the years 2000-2002were all under 2 days.

Moreover, we set 10,5 °C and 26,5 °C as lower and upper optimal base temperatures for fullbloom. The (normalized) sum of squares error, the average (absolute) error and the maximal errorare 2,22, 1,76 and 6 days for full bloom.

Table 3 represents the accumulated heat sums (°C) of the different varieties in the time period2000-2004 while Table 4 shows the errors of the estimations (days).

Table 3 The accumulated observed heat sums (°C) of the different varieties between budbreakand bloom in the time period 2000-2004 together with their averages (called critical heat sums)

VarietiesYears

Ch Ch_75 Ch_96 Szb Szb_34 Szb_52 Pb_54 Pb_55 Pb_D55 Rr_239 Rr_378 Rr_391 Rr_49 Hl_P41 Hl_K9 Average

2000 273,20 262,60 262,60 253,50 245,90 240,80 262,60 253,50 256,00 245,90 247,50 245,90 245,90 298,50 306,10 260,032001 247,05 247,05 240,45 248,55 248,05 244,95 254,75 250,15 252,25 244,95 248,05 244,95 251,65 302,15 291,15 254,412002 209,70 285,90 252,50 292,40 267,90 267,90 284,90 278,70 267,60 274,00 263,80 267,90 274,00 285,70 288,30 270,752003 248,78 248,78 248,78 219,35 227,58 227,58 248,78 227,58 235,68 227,58 213,98 227,58 227,58 219,35 219,35 231,222004 268,60 268,60 267,00 235,00 220,30 208,20 242,30 250,70 240,20 255,60 255,60 270,80 255,60 279,90 247,10 251,03

Average 249,47 262,59 254,27 249,76 241,95 237,89 258,67 252,13 250,35 249,61 245,79 251,43 250,95 277,12 270,40

Table 4 The deviations of the estimations from the observed dates of full bloomVarieties

YearsCh Ch_75 Ch_96 Szb Szb_34 Szb_52 Pb_54 Pb_55 Pb_D55 Rr_239 Rr_378 Rr_391 Rr_49 Hl_P41 Hl_K9

Yearlyaverage


2000 -2 0 -1 0 0 0 -1 0 -1 2 -1 2 2 -1 -2 1,002001 1 5 3 1 -2 -2 2 1 0 2 0 2 0 -1 -1 1,532002 4 -2 1 -6 -2 -3 -3 -3 -1 -3 -2 -1 -3 -1 -3 2,532003 1 2 1 3 2 1 1 3 2 2 3 3 3 6 5 2,532004 -1 0 -1 2 2 3 2 1 1 0 0 -2 0 0 3 1,20

Averageof the

absolutevalues

1,8 1,8 1,4 2,4 1,6 1,8 1,8 1,6 1 1,8 1,2 2 1,6 1,8 2,8 1,76

Analyzing the deviations of the estimations from the observed dates (Table 4) we can see thatthe starting date of full bloom of Hárslevelű K.9 clone was the most difficult to forecast. The most varieties however, have their absolute error near the average (1,76 days) which indicatesthe relative high stability of the model. The least average absolute error of the model wasresulted for the clone variety Pinot blanc D55 (1 day).

105

110

115

120

125

2000 2001 2002 2003 2004

day

year

Chardonnay

Ch__o

Ch__p

Ch_75_o

Ch_75_p

Ch_96_o

Ch_96_p

105

130

155

2000 2001 2002 2003 2004

day

year

Chardonnay

Ch__o

Ch__p

Ch_75_o

Ch_75_p

Ch_96_o

Ch_96_p

105

110

115

120

125

2000 2001 2002 2003 2004

day

year


Szb_o

Szb_p

Szb34_o

Szb34_p

Szb52_o

Szb52_p

105

130

155

2000 2001 2002 2003 2004

day

year


Szb_o

Szb_p

Szb34_o

Szb34_p

Szb52_o

Szb52_p

105

110

115

120

125

2000 2001 2002 2003 2004

day

year

Pinot blanc

Pb54_o

Pb54_p

Pb55_o

Pb55_p

Pb_D55_o

Pb_D55_p

105

130

155

2000 2001 2002 2003 2004

day

year

Pinot blanc

Pb54_o

Pb54_p

Pb55_o

Pb55_p

Pb_D55_o

Pb_D55_p

105

130

155

2000 2001 2002 2003 2004

day

year

Rizling (Riesling)

Rr239_o

Rr_239_p

Rr__378_o

Rr_378_p

Rr_391_o

Rr_391_p

Rr49_o

Rr49_p

105

130

155

2000 2001 2002 2003 2004

day

year

Rizling (Riesling)

Rr239_o

Rr_239_p

Rr__378_o

Rr_378_p

Rr_391_o

Rr_391_p

Rr49_o

Rr49_p

105

110

115

120

125

2000 2001 2002 2003 2004

day

year

Hárslevelű

Hl_P41_o

Hl_P41_p

Hl_K9_o

Hl_K9_p

105

130

155

2000 2001 2002 2003 2004

day

year

Hárslevelű

Hl_P41_o

Hl_P41_p

Hl_K9_o

Hl_K9_p

Figure 2 The observed (_o) and the predicted (_p) bud break (left) and full blooming (right)starting dates of the five white vine varieties in the time period 2000-2004.

3 - 69



112157

Chardonnay

115

152

113157

Pinot gris

119

152

111157

Pinot blanc

117

152

115158

Rajnai rizling

120

153

113160

Hárslevelű

119

156

Figure 3 The results of the phenological model in the time period 2000-2004 (external circle),respectively by the RegCM3.1 regional climate model predicted dates in 2021-2050 (internal

circle). The numbers mean the starting date of budburst and full boom (in Julian Day).

After having determined the parameters of the model, based on the regional climate modelRegCM3.1 (Bartholy, et al. 2009, Torma, et al. 2008), we examined what the model predicts tothe time period between 2021 and 2050. We illustrated our results on Figure 3. The modelpredicted the beginning of budbrake (red) about five days earlier, the starting date of full bloom(green) about five days later in the examined period compared to the observed data.

CONCLUSIONSPhenology modelling is a useful tool to predict the expected changes in scheduling, caused by

climate change. In case long-term phenology data together with weather records are available,the above introduced model can be improved. For this

1. Further spatial and temporal validity study is necessary. Other varieties should also beinvolved.

2. Other parameters as chilling sum during the dormancy as well as precipitation/humidity datashould be inserted in the model to make it more accurate.

ACKNOWLEDGMENTSOur work was supported by the projects No. OM-00265/2008 and OTKA K 63065/2006.

BIBLIOGRAPHYBartholy, J., Pongracz, R., Torma, Cs., Pieczka, I., Kardos, P., Hunyady, A., (2009): Analysis of

regional climate change modelling experiments for the Carpathian basin. International Journalof Global Warming 1, 238-252.

Bindi, M., Miglietta, F., Gozzini, B., Orlandini, S., Seghi, L. (1997a) A simple model forsimulation of growth and development in grapevine (Vitis vinifera L.). I. Model description.Vitis 36(2):67–71.

Bindi, M., Miglietta, F., Gozzini, B., Orlandini, S., Seghi, L. (1997b) A simple model forsimulation of growth and development in grapevine (Vitis vinifera L.). II. Model validation.Vitis 36(2):73–76.

Cesaraccio, C., Spano, D., Snyder, R. L., Duce, P. (2004) Chilling and forcing model to predictbud-burst of crop and forest species. Agric For Meteorol 126:1–13,doi:10.1016/j.agrformet.2004.03.010.

Giorgi, F., M. R. Marinucci, and G. T. Bates (1993): Development of a second generationregional climate model (RegCM2) i: Boundary layer and radiative transfer processes, Mon.Wea. Rev., 121, 2794–2813.

Gladstones, J. (2000) Past and future climatic indices for viticulture. Proc. 5th Intl. Symp. CoolClimate Vitic. Oenol., Melbourne, Australia. 10 pp.

Jones, G. V. (2003) Winegrape phenology. In: Schwartz MD (ed) Phenology: an integrativeenvironmental science. Kluwer, Milwaukee, pp 523–540.

Jones, G. V., Duchene, E., Tomasi, D., Yuste, J., Braslavksa, O., Schultz, H., Martinez, C., Boso,S., Langellier, F., Perruchot, C., Guimberteau, G. (2005) Changes in European winegrapephenology and relationships with climate. In: Proceedings of XIV International GESCOViticulture Congress, Geisenheim, Germany, 23–27 August, 2005, pp 55–62.

Lang, G. A., Early, J. D., Martin, G. C., Darnell, R. L. (1987) Endo-, para-, and ecodormancy:physiological terminology and classification for dormancy research. HortScience 22(3):371–377.

Moncur, M. W., Rattigan, K., Mackenzie, D. H., McIntyre, G. N. (1989) Base temperatures forbudbreak and leaf appearance of grapevines. Am J Enol Vitic 40(1):21–26.

Pernesz, Gy. (2004) New resistant table grape cultivars bred in Hungary Proceedings of the FirstInternational Symposium on Grapewine Growing, Commerce and Research, Acta HorticulturaeNumber 652 p 321.Tomasi, D., Pascarella, G., Sivilotti, P., Gardiman, M., Pitacco, A. (2005) Grape bud burst:

thermal heat requirement and bud antagonism. ISHS Acta Horticulturae 754: InternationalWorkshop on Advances in Grapevine and Wine Research. Sept. 14-17, Venosa.

Torma, Cs., Bartholy, J., Pongracz, R., Barcza, Z., Coppola, E., Giorgi, F., (2008): Adaptationand validation of the RegCM3 climate model for the Carpathian Basin. Idojaras 112, 233-247.

Riou, C. (1994) The effect of climate on grape ripening: application to the zoning of sugarcontent in the european community. CECACEE- CECA, Luxembourg.

3 - 70



112157

Chardonnay

115

152

113157

Pinot gris

119

152

111157

Pinot blanc

117

152

115158

Rajnai rizling

120

153

113160

Hárslevelű

119

156

Figure 3 The results of the phenological model in the time period 2000-2004 (external circle),respectively by the RegCM3.1 regional climate model predicted dates in 2021-2050 (internal

circle). The numbers mean the starting date of budburst and full boom (in Julian Day).

After having determined the parameters of the model, based on the regional climate modelRegCM3.1 (Bartholy, et al. 2009, Torma, et al. 2008), we examined what the model predicts tothe time period between 2021 and 2050. We illustrated our results on Figure 3. The modelpredicted the beginning of budbrake (red) about five days earlier, the starting date of full bloom(green) about five days later in the examined period compared to the observed data.

CONCLUSIONSPhenology modelling is a useful tool to predict the expected changes in scheduling, caused by

climate change. In case long-term phenology data together with weather records are available,the above introduced model can be improved. For this

1. Further spatial and temporal validity study is necessary. Other varieties should also beinvolved.

2. Other parameters as chilling sum during the dormancy as well as precipitation/humidity datashould be inserted in the model to make it more accurate.

ACKNOWLEDGMENTSOur work was supported by the projects No. OM-00265/2008 and OTKA K 63065/2006.

BIBLIOGRAPHYBartholy, J., Pongracz, R., Torma, Cs., Pieczka, I., Kardos, P., Hunyady, A., (2009): Analysis of

regional climate change modelling experiments for the Carpathian basin. International Journalof Global Warming 1, 238-252.

Bindi, M., Miglietta, F., Gozzini, B., Orlandini, S., Seghi, L. (1997a) A simple model forsimulation of growth and development in grapevine (Vitis vinifera L.). I. Model description.Vitis 36(2):67–71.

Bindi, M., Miglietta, F., Gozzini, B., Orlandini, S., Seghi, L. (1997b) A simple model forsimulation of growth and development in grapevine (Vitis vinifera L.). II. Model validation.Vitis 36(2):73–76.

Cesaraccio, C., Spano, D., Snyder, R. L., Duce, P. (2004) Chilling and forcing model to predictbud-burst of crop and forest species. Agric For Meteorol 126:1–13,doi:10.1016/j.agrformet.2004.03.010.

Giorgi, F., M. R. Marinucci, and G. T. Bates (1993): Development of a second generationregional climate model (RegCM2) i: Boundary layer and radiative transfer processes, Mon.Wea. Rev., 121, 2794–2813.

Gladstones, J. (2000) Past and future climatic indices for viticulture. Proc. 5th Intl. Symp. CoolClimate Vitic. Oenol., Melbourne, Australia. 10 pp.

Jones, G. V. (2003) Winegrape phenology. In: Schwartz MD (ed) Phenology: an integrativeenvironmental science. Kluwer, Milwaukee, pp 523–540.

Jones, G. V., Duchene, E., Tomasi, D., Yuste, J., Braslavksa, O., Schultz, H., Martinez, C., Boso,S., Langellier, F., Perruchot, C., Guimberteau, G. (2005) Changes in European winegrapephenology and relationships with climate. In: Proceedings of XIV International GESCOViticulture Congress, Geisenheim, Germany, 23–27 August, 2005, pp 55–62.

Lang, G. A., Early, J. D., Martin, G. C., Darnell, R. L. (1987) Endo-, para-, and ecodormancy:physiological terminology and classification for dormancy research. HortScience 22(3):371–377.

Moncur, M. W., Rattigan, K., Mackenzie, D. H., McIntyre, G. N. (1989) Base temperatures forbudbreak and leaf appearance of grapevines. Am J Enol Vitic 40(1):21–26.

Pernesz, Gy. (2004) New resistant table grape cultivars bred in Hungary Proceedings of the FirstInternational Symposium on Grapewine Growing, Commerce and Research, Acta HorticulturaeNumber 652 p 321.Tomasi, D., Pascarella, G., Sivilotti, P., Gardiman, M., Pitacco, A. (2005) Grape bud burst:

thermal heat requirement and bud antagonism. ISHS Acta Horticulturae 754: InternationalWorkshop on Advances in Grapevine and Wine Research. Sept. 14-17, Venosa.

Torma, Cs., Bartholy, J., Pongracz, R., Barcza, Z., Coppola, E., Giorgi, F., (2008): Adaptationand validation of the RegCM3 climate model for the Carpathian Basin. Idojaras 112, 233-247.

Riou, C. (1994) The effect of climate on grape ripening: application to the zoning of sugarcontent in the european community. CECACEE- CECA, Luxembourg.

3 - 71



1

ANALYSIS OF THE DAILY MINIMUM TEMPERATURES VARIABILITY IN THE CASABLANCA VALLEY, CHILE

Carlo Montes

Centro de Estudios Avanzados en Zonas Áridas (CEAZA) Raúl Bitrán S/N, La Serena, Chile

[email protected]

ABSTRACT The Casablanca Valley (CV) has a complex topography and is located near the Pacific

Ocean. These factors generate important climatic differences in relation to other wine producing zones of Central Chile. The air temperature is one of the most important atmospheric variables in viticulture by its influence on the vine development and the quality of the grapes and wines. In this work, the minimum temperature has been studied using a set of meteorological stations to make a comparative climatology between the CV and surrounding viticultural zones, and also with data from an agrometeorological network inside the CV, to make a local comparison applying the Principal Component Analysis. The synoptic configurations were analyzed for the higher and lower minimum temperatures. The comparison with the surrounding zones shows that the CV has differences in the annual cycle of the minimum temperatures (amplitude and extremes values). Its minimum temperature anomalies are less correlated with the more continental stations, and the differences are statistically more marked and are increasing with growing season. The analysis inside de CV shows low differences, with a 93% of the variance explained by the first principal component, but some oceanic influence exists. The analysis shows that the valley has a well differentiated regime of minimum temperatures compared with other wine-producing zones, noticeable in the warm period. Inside the CV there is a low spatial variability, with an important synoptic control, and it is possible to describe some gradient along the ocean proximity.

KEYWORD Minimum temperature – temperature variability – terroir – viticultural zoning INTRODUCTION The optimal physiological behavior of grapevines requires certain weather conditions, and

temperature is considered one of the main ones, given its role in the rate of maturation and harvest date of grapes, as well as its effect on grape composition and wine quality. In a daily scale, daytime temperatures are important for processes such as primary metabolism, while nighttime temperatures are important both as a risk factor (frosts) as well as its influence on processes like the secondary metabolism, responsible for the synthesis of compounds associated with wine quality (e.g. anthocyanins and tannins) (Jackson and Lombard, 1993). The importance of the minimum temperatures on wine production was the aim for doing the study of its variability in an important white wine-producing region of Chile. The study was performed in the Casablanca Valley, which corresponds to the main area of production of grape varieties for white wines, given its cold weather condition compared to other Chilean viticultural valleys. The valley has a complex topography, with higher altitudes reaching about 1000 m and lower points of about 100 m above sea level. Its area is close to 20 km NS

2

and 17 km EW, and has a distance of about 18 km from the Pacific Ocean. The main objective of this work is to characterize the minimum temperature variability in the Casablanca Valley and make a comparison with other regions of Central Chile.

MATERIALS AND METHODS Daily minimum temperatures data of four weather stations were used for the period between

January 2001 to December 2007, one located in the Casablanca Valley and three in surrounding zones outside the valley. These stations are: Casablanca (33.32S 71.44W), Santiago (33.43S 70.68W), La Platina (southern Santiago, 33.52S 70.62W) and Codigua (33.46S 71.20W) (Fig.1). Also data from an agrometeorological network of 9 stations at the Casablanca Valley were used for the period September 2006 to July 2008, showed in Fig.2. A non parametric statistical test and a Principal Components Analysis were performed for the time series comparison. The synoptic configurations were studied, using data of sea level pressure and wind vectors from the European Center of Medium-Range Weather Forecast reanalysis (Simmons and Gibson, 2000).

Longitude (º)

Latit

ude

(º)

Cod

St

Lp

Csb

-71.6 -71.2 -70.8 -70.4 -70-34,0

-33,8

-33,6

-33,4

-33,2

-33,0

-72 -71 -70-36

-35

-34

-33

-32

-31

-30

-29

Longitude (º)

Latit

ude

(º)

Figure 1. Position of the stations Casablanca (Csb), Codigua (Co), Santiago (St) and La Platina (Lp). The white square shows the location of the Casablanca Valley, and gray outline corresponds to the city of Santiago.

Longitude (º)

Latit

ude

(º)

E1

E2

E3

E4

E5

E6E7E8

E9

-71.6 -71.5 -71.4 -71.3 -71.2-33.5

-33.4

-33.3

-33.2

-72 -71 -70-36

-35

-34

-33

-32

-31

-30

-29

Longitude (º)

Latit

ude

(º)

Figure 2. Position of the agrometeorological stations located in the Casablanca Valley.

RESULTS AND DISCUSSION The first analysis was done for the minimum temperature series of the Casablanca Valley

and the three weather stations in nearby areas (Fig.1). In Fig.3 the time series and the annual cycle obtained by fitting Fourier series are showed. In the curves it can be seen that, for the

3 - 72



1

ANALYSIS OF THE DAILY MINIMUM TEMPERATURES VARIABILITY IN THE CASABLANCA VALLEY, CHILE

Carlo Montes

Centro de Estudios Avanzados en Zonas Áridas (CEAZA) Raúl Bitrán S/N, La Serena, Chile

[email protected]

ABSTRACT The Casablanca Valley (CV) has a complex topography and is located near the Pacific

Ocean. These factors generate important climatic differences in relation to other wine producing zones of Central Chile. The air temperature is one of the most important atmospheric variables in viticulture by its influence on the vine development and the quality of the grapes and wines. In this work, the minimum temperature has been studied using a set of meteorological stations to make a comparative climatology between the CV and surrounding viticultural zones, and also with data from an agrometeorological network inside the CV, to make a local comparison applying the Principal Component Analysis. The synoptic configurations were analyzed for the higher and lower minimum temperatures. The comparison with the surrounding zones shows that the CV has differences in the annual cycle of the minimum temperatures (amplitude and extremes values). Its minimum temperature anomalies are less correlated with the more continental stations, and the differences are statistically more marked and are increasing with growing season. The analysis inside de CV shows low differences, with a 93% of the variance explained by the first principal component, but some oceanic influence exists. The analysis shows that the valley has a well differentiated regime of minimum temperatures compared with other wine-producing zones, noticeable in the warm period. Inside the CV there is a low spatial variability, with an important synoptic control, and it is possible to describe some gradient along the ocean proximity.

KEYWORD Minimum temperature – temperature variability – terroir – viticultural zoning INTRODUCTION The optimal physiological behavior of grapevines requires certain weather conditions, and

temperature is considered one of the main ones, given its role in the rate of maturation and harvest date of grapes, as well as its effect on grape composition and wine quality. In a daily scale, daytime temperatures are important for processes such as primary metabolism, while nighttime temperatures are important both as a risk factor (frosts) as well as its influence on processes like the secondary metabolism, responsible for the synthesis of compounds associated with wine quality (e.g. anthocyanins and tannins) (Jackson and Lombard, 1993). The importance of the minimum temperatures on wine production was the aim for doing the study of its variability in an important white wine-producing region of Chile. The study was performed in the Casablanca Valley, which corresponds to the main area of production of grape varieties for white wines, given its cold weather condition compared to other Chilean viticultural valleys. The valley has a complex topography, with higher altitudes reaching about 1000 m and lower points of about 100 m above sea level. Its area is close to 20 km NS

2

and 17 km EW, and has a distance of about 18 km from the Pacific Ocean. The main objective of this work is to characterize the minimum temperature variability in the Casablanca Valley and make a comparison with other regions of Central Chile.

MATERIALS AND METHODS Daily minimum temperatures data of four weather stations were used for the period between

January 2001 to December 2007, one located in the Casablanca Valley and three in surrounding zones outside the valley. These stations are: Casablanca (33.32S 71.44W), Santiago (33.43S 70.68W), La Platina (southern Santiago, 33.52S 70.62W) and Codigua (33.46S 71.20W) (Fig.1). Also data from an agrometeorological network of 9 stations at the Casablanca Valley were used for the period September 2006 to July 2008, showed in Fig.2. A non parametric statistical test and a Principal Components Analysis were performed for the time series comparison. The synoptic configurations were studied, using data of sea level pressure and wind vectors from the European Center of Medium-Range Weather Forecast reanalysis (Simmons and Gibson, 2000).

Longitude (º)

Latit

ude

(º)

Cod

St

Lp

Csb

-71.6 -71.2 -70.8 -70.4 -70-34,0

-33,8

-33,6

-33,4

-33,2

-33,0

-72 -71 -70-36

-35

-34

-33

-32

-31

-30

-29

Longitude (º)

Latit

ude

(º)

Figure 1. Position of the stations Casablanca (Csb), Codigua (Co), Santiago (St) and La Platina (Lp). The white square shows the location of the Casablanca Valley, and gray outline corresponds to the city of Santiago.

Longitude (º)

Latit

ude

(º)

E1

E2

E3

E4

E5

E6E7E8

E9

-71.6 -71.5 -71.4 -71.3 -71.2-33.5

-33.4

-33.3

-33.2

-72 -71 -70-36

-35

-34

-33

-32

-31

-30

-29

Longitude (º)

Latit

ude

(º)

Figure 2. Position of the agrometeorological stations located in the Casablanca Valley.

RESULTS AND DISCUSSION The first analysis was done for the minimum temperature series of the Casablanca Valley

and the three weather stations in nearby areas (Fig.1). In Fig.3 the time series and the annual cycle obtained by fitting Fourier series are showed. In the curves it can be seen that, for the

3 - 73



3

entire period, the Casablanca serie is the one with the lowest average values of minimum temperature, which rarely exceed the average of 10ºC, and the lower amplitude in the annual cycle. Such differences are more marked with the stations Santiago and La Platina, which are located in a more continental position, due to their higher distance from the Ocean and its influence. Taking the differences between the minimum temperature series of each station and their respective fitted Fourier series, the minimum temperature anomalies (MTa) were obtained. Having removed the annual cycle of the minimum temperature it is possible to do a correlation between them.

2001 2002 2003 2004 2005 2006 2007-5

0

5

10

15

20

Year

Tem

pera

ture

(ºC)

Casablanca

2001 2002 2003 2004 2005 2006 2007-5

0

5

10

15

20

YearTe

mpe

ratu

re (º

C)

Santiago

2001 2002 2003 2004 2005 2006 2007-5

0

5

10

15

20

Year

Tem

pera

ture

(ºC)

Codigua

2001 2002 2003 2004 2005 2006 2007-5

0

5

10

15

20

Year

Tem

pera

ture

(ºC)

La Platina

Figure 3. Minimum temperature series (gray) for Casablanca, Codigua, Santiago and La Platina. The black curve represents the annual cycle obtained by fitting Fourier series.

The Pearson correlation coefficient between the MTa of Casablanca and Santiago (0.57),

and between Casablanca and La Platina (0.55) is nearly the same. The correlation between the MTa of Casablanca and Codigua is higher (0.76), which shows the difference between zones near the coastline and inland, and the different thermal regulation in the basin of Santiago and the zones near the ocean, which also can be seen in the high correlation between Santiago and La Platina (0.81) and between Casablanca and Codigua (0.76). It is likely that the spatial and temporal patterns are both strongly influenced by the synoptic conditions, observable in the high correlation between the MTa of the nearest stations, as the influence of local factors like the proximity to the ocean from each geographical area.

Given the different importance of minimum temperatures in the annual cycle in viticulture, a comparative analysis in different periods was done. The annual average series of minimum temperatures from 2001 to 2007 were calculated (Fig.4). Casablanca and Codigua have an average difference of nearly 2.5°C throughout the year, while between Casablanca and La Platina-Santiago the main differences are extended to the warm season. From the point of view of wine production, these differences indicate a condition of grapes growing at Casablanca which can take a maturation process under nighttime temperatures described as favorable for the optimal quality factors expression (Jackson and Lombard, 1993). This highlights the difference with the Central Valley area of Chile, but it also indicates that during the winter months the chill accumulation, needed for breaking dormancy, can be met in both

4

areas. It is likely that the geographical position of Codigua station, which is located near the mouth of the Maipo River, gets more direct influence of air from the Pacific Ocean, which would result in higher minimum temperatures.

Jan Apr Jul Oct Jan0

5

10

15

Month

Tem

pera

ture

(ºC

)

Csb Stg


5

10

15

Month

Tem

pera

ture

(ºC

)

Csb Cod


5

10

15

Month

Tem

pera

ture

(ºC

)

Csb Lp

Figure 4. Average annual series of minimum temperature and Fourier fitting. Csb: Casablanca, Stg: Santiago; Cod: Codigua; Lp: La Platina.

As mentioned, the differences described above are clear for spring-summer time, but during

autumn-winter such differences tend to decrease. To estimate the time when the temperatures are statistically different or not, the Wilcoxon-Mann-Whitney nonparametric test (Wilks, 1995) was performed for a comparison of a 30-day mobile period for each station and Casablanca. Thus, the period 1 to 30 January, 2 to 31 January, 3 January to 1 February, and so on was compared successively. Fig.5 shows the temporal change of the p value, which indicates the statistical significance of the differences between the ranges of 30 days being compared, in addition to the limit of p = 0.05 used as criteria for statistical significance. There is a clear difference between the three comparisons. So between Casablanca and Codigua the minimum temperatures are statistically different in the entire year, except for a short period in autumn when the p value exceeds 0.05. The minimum temperatures between Casablanca and the Santiago-La Platina stations remain statistically different for the whole period of spring and summer (p<0.05), but for a long period in autumn-winter they are not statistically different except for a short period, which could respond to the greater synoptic scale variability induced by the passage of frontal systems in central Chile, which are more active during the winter months, and increases the variability of this time scale (Falvey and Garreaud, 2007).

A local comparison was done throughout the Principal Component Analysis (PCA) and the associated Empirical Orthogonal Functions (EOF) (Wilks, 1995), using the minimum temperatures anomalies for the data from the agrometeorological network inside the Casablanca Valley (Fig.2). For the temporal variability, the first mode explains a 93.4% of the total variance with a large decrease for the subsequent modes. The high variance explained by the first PC could be related to the proximity of each station and to the topographical configuration of the valley, which allows a common synoptic control that regulates temperature behavior. The EOF’s associated with each PC show some spatial structure of the main modes of variation. In Table 1 the values of the EOF1 and EOF2 are presented. The structure of the EOF1 shows differences between the stations located near the coast and the others, except station E1, which has a different behavior. These results show that, despite the low temporal variability that exists in the valley, there is a spatial pattern of the MTa that varies with the sea proximity. The EOF2 has a different pattern that could respond to some

3 - 74



3

entire period, the Casablanca serie is the one with the lowest average values of minimum temperature, which rarely exceed the average of 10ºC, and the lower amplitude in the annual cycle. Such differences are more marked with the stations Santiago and La Platina, which are located in a more continental position, due to their higher distance from the Ocean and its influence. Taking the differences between the minimum temperature series of each station and their respective fitted Fourier series, the minimum temperature anomalies (MTa) were obtained. Having removed the annual cycle of the minimum temperature it is possible to do a correlation between them.

2001 2002 2003 2004 2005 2006 2007-5

0

5

10

15

20

Year

Tem

pera

ture

(ºC)

Casablanca

2001 2002 2003 2004 2005 2006 2007-5

0

5

10

15

20

Year

Tem

pera

ture

(ºC)

Santiago

2001 2002 2003 2004 2005 2006 2007-5

0

5

10

15

20

Year

Tem

pera

ture

(ºC)

Codigua

2001 2002 2003 2004 2005 2006 2007-5

0

5

10

15

20

Year

Tem

pera

ture

(ºC)

La Platina

Figure 3. Minimum temperature series (gray) for Casablanca, Codigua, Santiago and La Platina. The black curve represents the annual cycle obtained by fitting Fourier series.

The Pearson correlation coefficient between the MTa of Casablanca and Santiago (0.57),

and between Casablanca and La Platina (0.55) is nearly the same. The correlation between the MTa of Casablanca and Codigua is higher (0.76), which shows the difference between zones near the coastline and inland, and the different thermal regulation in the basin of Santiago and the zones near the ocean, which also can be seen in the high correlation between Santiago and La Platina (0.81) and between Casablanca and Codigua (0.76). It is likely that the spatial and temporal patterns are both strongly influenced by the synoptic conditions, observable in the high correlation between the MTa of the nearest stations, as the influence of local factors like the proximity to the ocean from each geographical area.

Given the different importance of minimum temperatures in the annual cycle in viticulture, a comparative analysis in different periods was done. The annual average series of minimum temperatures from 2001 to 2007 were calculated (Fig.4). Casablanca and Codigua have an average difference of nearly 2.5°C throughout the year, while between Casablanca and La Platina-Santiago the main differences are extended to the warm season. From the point of view of wine production, these differences indicate a condition of grapes growing at Casablanca which can take a maturation process under nighttime temperatures described as favorable for the optimal quality factors expression (Jackson and Lombard, 1993). This highlights the difference with the Central Valley area of Chile, but it also indicates that during the winter months the chill accumulation, needed for breaking dormancy, can be met in both

4

areas. It is likely that the geographical position of Codigua station, which is located near the mouth of the Maipo River, gets more direct influence of air from the Pacific Ocean, which would result in higher minimum temperatures.


5

10

15

Month

Tem

pera

ture

(ºC

)

Csb Stg


5

10

15

Month

Tem

pera

ture

(ºC

)

Csb Cod


5

10

15

Month

Tem

pera

ture

(ºC

)

Csb Lp

Figure 4. Average annual series of minimum temperature and Fourier fitting. Csb: Casablanca, Stg: Santiago; Cod: Codigua; Lp: La Platina.

As mentioned, the differences described above are clear for spring-summer time, but during

autumn-winter such differences tend to decrease. To estimate the time when the temperatures are statistically different or not, the Wilcoxon-Mann-Whitney nonparametric test (Wilks, 1995) was performed for a comparison of a 30-day mobile period for each station and Casablanca. Thus, the period 1 to 30 January, 2 to 31 January, 3 January to 1 February, and so on was compared successively. Fig.5 shows the temporal change of the p value, which indicates the statistical significance of the differences between the ranges of 30 days being compared, in addition to the limit of p = 0.05 used as criteria for statistical significance. There is a clear difference between the three comparisons. So between Casablanca and Codigua the minimum temperatures are statistically different in the entire year, except for a short period in autumn when the p value exceeds 0.05. The minimum temperatures between Casablanca and the Santiago-La Platina stations remain statistically different for the whole period of spring and summer (p<0.05), but for a long period in autumn-winter they are not statistically different except for a short period, which could respond to the greater synoptic scale variability induced by the passage of frontal systems in central Chile, which are more active during the winter months, and increases the variability of this time scale (Falvey and Garreaud, 2007).

A local comparison was done throughout the Principal Component Analysis (PCA) and the associated Empirical Orthogonal Functions (EOF) (Wilks, 1995), using the minimum temperatures anomalies for the data from the agrometeorological network inside the Casablanca Valley (Fig.2). For the temporal variability, the first mode explains a 93.4% of the total variance with a large decrease for the subsequent modes. The high variance explained by the first PC could be related to the proximity of each station and to the topographical configuration of the valley, which allows a common synoptic control that regulates temperature behavior. The EOF’s associated with each PC show some spatial structure of the main modes of variation. In Table 1 the values of the EOF1 and EOF2 are presented. The structure of the EOF1 shows differences between the stations located near the coast and the others, except station E1, which has a different behavior. These results show that, despite the low temporal variability that exists in the valley, there is a spatial pattern of the MTa that varies with the sea proximity. The EOF2 has a different pattern that could respond to some

3 - 75



5

influence of other physical factors, since the pattern is reversed, but with a low explained variance (3.7%).


0.10.20.30.40.5

p va

lue

Time

Casablanca - Santiago


0.10.20.30.40.5

p va

lue

Time

Casablanca - Codigua


0.10.20.30.40.5

p va

lue

Time

Casablanca - La Platina

Figure 5. Temporal change of the p value of the Wilcoxon-Mann-Whitney test. The dotted line shows the limit of p=0.05 for statistically significant differences. Table 1. Empirical Orthogonal Functions associated to the Principal Component 1 and 2.

Stations EOF 1 EOF 2 E1 0.886 0.426 E2 0.985 0.089 E3 0.986 0.055 E4 0.985 0.054 E5 0.984 0.002 E6 0.978 -0.175 E7 0.972 -0.204 E8 0.956 -0.255 E9 0.957 0.040

To explore the synoptic configurations associated with the extreme minimum temperature at

the valley, the anomalies of sea level pressure (SLP) and wind vectors corresponding to days of the 10% upper and lower limit of the PC1 were taken. The days with lower minimum temperatures (Fig.6a) are associated with a regional positive SLP anomaly in the Pacific Ocean, which allows a greater surface cooling, and with a negative anomaly in the Atlantic region, with a wind field that allows a cold air intrusion from the polar regions as a geostrophic response, what has been described in a generalized way (Garreaud, 2000). The days with upper minimum temperatures are associated with a weakening of the dominant anticyclonic activity (Fig.6b), with a negative SLP anomaly next to the Chilean coasts, and a wind field with an important meridional component that allows the arrival of equatorial air. This analysis allows to infer an important regional synoptic influence that regulates the minimum temperatures variability in a scale of the whole valley, given the weakening or the intensification of the SLP and the advection of cooler or warmer air than Casablanca.

6

-7

-7

-7

-6

-6

-5

-5

-4

-4

-4

-3

-3

-3

-2

-2

-2

-2

-1

-1

-1

-1

-1

0

0

0

0

0

0

1

11

1

1

2

2

2

2

2

3

3

3

3

3

4

4

4

4

4

5

55

5

5

6

6

6

7

7

7

8

8

8

-8

-8

-8

9

9

9

-9

-9

-9-10

-10

-11

-11

6

6

-6-5

0

-4

7

-3

10

7

Longitud (º)

Latit

ud (º

)

-110 -90 -70 -50 -30-60

-50

-40

-30

-20

-10

0

-10

-5

0

5

10

-4

-4

-3

-3

-2

-2

-2

-2

-2

-2

-2

-1

-1

-1

-1 -1

-1-1

-1

-1

0

0

0

0

0

-3-3

-3

-3

-4

-4

-4

-4

0

0

-5

-5

-5

-6

-6

00

-7

-5

0

0

Longitude (º)

Latit

ude

(º)

-110 -90 -70 -50 -30-60

-50

-40

-30

-20

-10

0

-12

-10

-8

-6

-4

-2

0

250 260 270 280 290 300 310 320 330-60

-50

-40

-30

-20

-10

06 m/s

250 260 270 280 290 300 310 320 330-60

-50

-40

-30

-20

-10

010 m/s

Figure 6. Composites of sea level pressure (hPa) and wind vectors (m/s) anomalies for: a) lower and b) upper limit of the 10% in the PC1. White star shows the Casablanca Valley location.

CONCLUSIONS The variability of the minimum temperatures in the Casablanca Valley was analyzed. The

comparison shows important differences with the more continental zones, in a regional scale, with differences in the annual cycle that are marked mainly in the growing season and not in winter and autumn time. This is significant from the point of view of the occurrence of sprimg frosts, which are more frequent in the Casablanca Valley, and for the lowest temperatures at which the plants are exposed during the growing season. The local comparison shows a low temporal variability, with minimum temperature anomalies that don’t change significatively at one point or another. The associated EOF shows that, despite the low temporal variability, there is a spatial pattern that shows a sea proximity orientation. The synoptic patterns associated with the extreme minimum temperature anomalies shows an average pressure and wind field associated to the cold or warm air advection and the surface process that can induce a reinforced nocturnal cooling.

ACKNOWLEDGMENTS The author gratefully acknowledges the Asociación de Empresarios Vitivinícolas del Valle

de Casablanca and the Dirección Meteorológica de Chile for the meteorological data. BIBLIOGRAPHY

Falvey, M. and R. Garreaud. 2007. Wintertime precipitation episodes in Central Chile: associated meteorological conditions and orographic influences. J. of Hydrometeor. 8: 171-193. Jackson D.I. and P.B. Lombard. 1993. Environmental and management practices affecting grape composition and wine quality: a review. Am. J. Enol. Vitic. 4, 409–430. Simmons, A. and J. Gibson. 2000. The ERA-40 Project Plan. ERA-40 Project Report Series No. 1. Wilks, D. 1995. Statistical methods in the atmospheric sciences. Academic Press. 604p.

3 - 76



5

influence of other physical factors, since the pattern is reversed, but with a low explained variance (3.7%).


0.10.20.30.40.5

p va

lue

Time

Casablanca - Santiago


0.10.20.30.40.5

p va

lue

Time

Casablanca - Codigua


0.10.20.30.40.5

p va

lue

Time

Casablanca - La Platina

Figure 5. Temporal change of the p value of the Wilcoxon-Mann-Whitney test. The dotted line shows the limit of p=0.05 for statistically significant differences. Table 1. Empirical Orthogonal Functions associated to the Principal Component 1 and 2.

Stations EOF 1 EOF 2 E1 0.886 0.426 E2 0.985 0.089 E3 0.986 0.055 E4 0.985 0.054 E5 0.984 0.002 E6 0.978 -0.175 E7 0.972 -0.204 E8 0.956 -0.255 E9 0.957 0.040

To explore the synoptic configurations associated with the extreme minimum temperature at

the valley, the anomalies of sea level pressure (SLP) and wind vectors corresponding to days of the 10% upper and lower limit of the PC1 were taken. The days with lower minimum temperatures (Fig.6a) are associated with a regional positive SLP anomaly in the Pacific Ocean, which allows a greater surface cooling, and with a negative anomaly in the Atlantic region, with a wind field that allows a cold air intrusion from the polar regions as a geostrophic response, what has been described in a generalized way (Garreaud, 2000). The days with upper minimum temperatures are associated with a weakening of the dominant anticyclonic activity (Fig.6b), with a negative SLP anomaly next to the Chilean coasts, and a wind field with an important meridional component that allows the arrival of equatorial air. This analysis allows to infer an important regional synoptic influence that regulates the minimum temperatures variability in a scale of the whole valley, given the weakening or the intensification of the SLP and the advection of cooler or warmer air than Casablanca.

6

-7

-7

-7

-6

-6

-5

-5

-4

-4

-4

-3

-3

-3

-2

-2

-2

-2

-1

-1

-1

-1

-1

0

0

0

0

0

0

1

11

1

1

2

2

2

2

2

3

3

3

3

3

4

4

4

4

4

5

55

5

5

6

6

6

7

7

7

8

8

8

-8

-8

-8

9

99

-9

-9

-9-10

-10

-11

-11

6

6

-6-5

0

-4

7

-3

10

7

Longitud (º)

Latit

ud (º

)

-110 -90 -70 -50 -30-60

-50

-40

-30

-20

-10

0

-10

-5

0

5

10

-4

-4

-3

-3

-2-2

-2

-2

-2

-2

-2

-1

-1

-1

-1 -1

-1-1

-1

-1

0

0

0

0

0

-3-3

-3

-3

-4

-4

-4

-4

0

0

-5

-5

-5

-6

-6

00

-7

-5

0

0

Longitude (º)

Latit

ude

(º)

-110 -90 -70 -50 -30-60

-50

-40

-30

-20

-10

0

-12

-10

-8

-6

-4

-2

0

250 260 270 280 290 300 310 320 330-60

-50

-40

-30

-20

-10

06 m/s

250 260 270 280 290 300 310 320 330-60

-50

-40

-30

-20

-10

010 m/s

Figure 6. Composites of sea level pressure (hPa) and wind vectors (m/s) anomalies for: a) lower and b) upper limit of the 10% in the PC1. White star shows the Casablanca Valley location.

CONCLUSIONS The variability of the minimum temperatures in the Casablanca Valley was analyzed. The

comparison shows important differences with the more continental zones, in a regional scale, with differences in the annual cycle that are marked mainly in the growing season and not in winter and autumn time. This is significant from the point of view of the occurrence of sprimg frosts, which are more frequent in the Casablanca Valley, and for the lowest temperatures at which the plants are exposed during the growing season. The local comparison shows a low temporal variability, with minimum temperature anomalies that don’t change significatively at one point or another. The associated EOF shows that, despite the low temporal variability, there is a spatial pattern that shows a sea proximity orientation. The synoptic patterns associated with the extreme minimum temperature anomalies shows an average pressure and wind field associated to the cold or warm air advection and the surface process that can induce a reinforced nocturnal cooling.

ACKNOWLEDGMENTS The author gratefully acknowledges the Asociación de Empresarios Vitivinícolas del Valle

de Casablanca and the Dirección Meteorológica de Chile for the meteorological data. BIBLIOGRAPHY

Falvey, M. and R. Garreaud. 2007. Wintertime precipitation episodes in Central Chile: associated meteorological conditions and orographic influences. J. of Hydrometeor. 8: 171-193. Jackson D.I. and P.B. Lombard. 1993. Environmental and management practices affecting grape composition and wine quality: a review. Am. J. Enol. Vitic. 4, 409–430. Simmons, A. and J. Gibson. 2000. The ERA-40 Project Plan. ERA-40 Project Report Series No. 1. Wilks, D. 1995. Statistical methods in the atmospheric sciences. Academic Press. 604p.

3 - 77



ANALYSIS OF CLIMATE SPATIO-TEMPORAL VARIABILITY IN THE

CONEGLIANO-VALDOBBIADENE DOCG WINE DISTRICT

G. Fila, F. Meggio, L.M. Veilleux, A.PitaccoUniversity of Padova, Department of Environmental Agronomy and Crop Science

I-35020 Legnaro (PD), Italy

[email protected]

ABSTRACT

Local climate characterization is fundamental in terroir description, yet global change

perspectives raise questions about its feasibility, since temporal stability cannot be no more

assumed for the forthcoming years.

The objective of this work was to gain a better understanding of the climatic spatio-temporal

variability of a grapevine growing area, and how this has changed during recent times.

Using as a case-study the Conegliano-Valdobbiadene DOCG wine district in North-Eastern

Italy, we developed a methodology to downscale daily mean air temperature from the European

Climate Assessment gridded dataset (E-OBS), to derive daily temperature surfaces at 500m

spatial resolution. This allowed to analyse how the spatio-temporal variability affected grapevine

phenology in the last 60 years.

The main results showed that, respect to the 1950-1979 period, the average Winkler index

between 1980 and 2008 showed a +184 °C increase, with little spatial variation, as well as for the

estimated dates for the main phenological events, which showed a generalized anticipation of

about 2 to 5 days. More pronounced changes were observed on the interannual variability, which

showed increases in both the average values and pattern of distribution.

KEYWORDS

Grapevine, Climate Change, Temperature, Phenology, Downscaling, Spatial Interpolation

INTRODUCTION

Climate is one of the most important factor in the description of terroir, since the amount and

temporal distribution of weather variables influences greatly wine production under both the

quantitative and qualitative aspects (Jones and Davis, 2000). It is generally always possible to

describe the principal traits of a local climate, provided that temporal stability and a relative

spatial homogeneity can be assumed. These two conditions are implicit in the notion of terroir,

which admits also the year to year climatic variability, which generates the vintage effect, as long

as it can be considered a fluctuation within a stable range of variation.

The climate stability assumption is being challenged by the forthcoming global changes

predicted by climatologists during the 21st century. Adaptation to these changes represent a major

challenge for viticulturists worldwide, and among the various issues which need to be addressed,

one is to understand to what extent the characterization of terroirs will remain feasible in a

context of climatic changes.

In this work we have undertaken the task by studying the climatology of a wine production

district in North-Eastern Italy, under both the spatial and temporal variability.

The analyses of trends in climate variability requires long-term weather observations taken at

high temporal and spatial resolution. For the area under study, weather data were available only

for a limited time extent, with wide gaps in the recording series. Global or continental climatic

datasets which covers much longer time extents could be used, but they are typically elaborated

at low spatial resolutions, which made them suitable only for large scale studies.

Part of the work was dedicated therefore to the development of a downscaling procedure for

reconstruct local past weather from a European gridded dataset, and to interpolate it at a high

spatial resolution. The result was the construction of complete daily weather database which

allowed to analyse how grapevine phenology has changed within the area under study during the

last sixty years, on both the spatial and time scale.


Study site - The Conegliano-Valdobbiadene DOCG wine district is located in the Veneto region

(45.95N, 12.17E). It is a prevailing hilly area which comprises 15 municipalities, extending for

approximately 35 and 25 km in the East-West, and Norh-South directions respectively. Elevation

ranges from 30 to 460m a.s.l., and 72.7% of the surface covered by vineyards varies between 100

and 300 m. The cultivated area is 8017 ha wide, and the principal production is the Prosecco di

Conegliano-Valdobbiadene wine, obtained mainly from the native grape cultivar Glera (formerly

“Prosecco”).

Weather data collection - Historical records of daily maximum and minimum temperature were

obtained from ARPAV the regional agency for environmental prevention and protection, and

from CODITV, the local provincial agency for crop protection. In total, data from 1989 to 2008

for 20 sites were available.

Another utilized source of data was the European Climate Assessment (ECA) dataset,

developed as part of the European Union Framework 6 ENSEMBLES project, (Haylock et al.

2008, Klok & Klein Tank, 2009). This dataset has been elaborated into a gridded version, the E-

OBS gridded dataset, which includes data at 0.25° spatial resolution for maximum, minimum and

average daily air temperature, as well as rainfall. The whole gridded dataset was downloaded

from the web site http://eca.knmi.nl.

E-OBS data were used to reconstruct local time series through a downscaling procedure, while

locally measured data were used to calibrate and validate the procedure.

Soil cover data - Information about soil cover was used to focus the analysis only on the surface

actually covered by vineyards. The location of all vineyards plots was extracted from the 1:10000

soil cover map published in shapefile format by Veneto Region in the year 2009.

Downscaling approach - The spatial distribution of the surface air temperature at a given

moment is determined by the interactions between the state of the atmosphere and the terrain

features, e.g. elevation, slope or the presence of water bodies, whose effects changes depending

on the type of weather conditions. As an example, the lapse rate induced by elevation gradients

are known to vary throughout the year, following seasonal patterns. Since it is reasonable to think

that similar weather conditions should correspond to similar patterns of atmosphere-terrain

interaction, it was hypothesised that if a finite number of “weather types” could be defined

according to some classification criterion, then each type should correspond to a specific pattern

3 - 78



ANALYSIS OF CLIMATE SPATIO-TEMPORAL VARIABILITY IN THE

CONEGLIANO-VALDOBBIADENE DOCG WINE DISTRICT

G. Fila, F. Meggio, L.M. Veilleux, A.PitaccoUniversity of Padova, Department of Environmental Agronomy and Crop Science

I-35020 Legnaro (PD), Italy

[email protected]

ABSTRACT

Local climate characterization is fundamental in terroir description, yet global change

perspectives raise questions about its feasibility, since temporal stability cannot be no more

assumed for the forthcoming years.

The objective of this work was to gain a better understanding of the climatic spatio-temporal

variability of a grapevine growing area, and how this has changed during recent times.

Using as a case-study the Conegliano-Valdobbiadene DOCG wine district in North-Eastern

Italy, we developed a methodology to downscale daily mean air temperature from the European

Climate Assessment gridded dataset (E-OBS), to derive daily temperature surfaces at 500m

spatial resolution. This allowed to analyse how the spatio-temporal variability affected grapevine

phenology in the last 60 years.

The main results showed that, respect to the 1950-1979 period, the average Winkler index

between 1980 and 2008 showed a +184 °C increase, with little spatial variation, as well as for the

estimated dates for the main phenological events, which showed a generalized anticipation of

about 2 to 5 days. More pronounced changes were observed on the interannual variability, which

showed increases in both the average values and pattern of distribution.

KEYWORDS

Grapevine, Climate Change, Temperature, Phenology, Downscaling, Spatial Interpolation

INTRODUCTION

Climate is one of the most important factor in the description of terroir, since the amount and

temporal distribution of weather variables influences greatly wine production under both the

quantitative and qualitative aspects (Jones and Davis, 2000). It is generally always possible to

describe the principal traits of a local climate, provided that temporal stability and a relative

spatial homogeneity can be assumed. These two conditions are implicit in the notion of terroir,

which admits also the year to year climatic variability, which generates the vintage effect, as long

as it can be considered a fluctuation within a stable range of variation.

The climate stability assumption is being challenged by the forthcoming global changes

predicted by climatologists during the 21st century. Adaptation to these changes represent a major

challenge for viticulturists worldwide, and among the various issues which need to be addressed,

one is to understand to what extent the characterization of terroirs will remain feasible in a

context of climatic changes.

In this work we have undertaken the task by studying the climatology of a wine production

district in North-Eastern Italy, under both the spatial and temporal variability.

The analyses of trends in climate variability requires long-term weather observations taken at

high temporal and spatial resolution. For the area under study, weather data were available only

for a limited time extent, with wide gaps in the recording series. Global or continental climatic

datasets which covers much longer time extents could be used, but they are typically elaborated

at low spatial resolutions, which made them suitable only for large scale studies.

Part of the work was dedicated therefore to the development of a downscaling procedure for

reconstruct local past weather from a European gridded dataset, and to interpolate it at a high

spatial resolution. The result was the construction of complete daily weather database which

allowed to analyse how grapevine phenology has changed within the area under study during the

last sixty years, on both the spatial and time scale.


Study site - The Conegliano-Valdobbiadene DOCG wine district is located in the Veneto region

(45.95N, 12.17E). It is a prevailing hilly area which comprises 15 municipalities, extending for

approximately 35 and 25 km in the East-West, and Norh-South directions respectively. Elevation

ranges from 30 to 460m a.s.l., and 72.7% of the surface covered by vineyards varies between 100

and 300 m. The cultivated area is 8017 ha wide, and the principal production is the Prosecco di

Conegliano-Valdobbiadene wine, obtained mainly from the native grape cultivar Glera (formerly

“Prosecco”).

Weather data collection - Historical records of daily maximum and minimum temperature were

obtained from ARPAV the regional agency for environmental prevention and protection, and

from CODITV, the local provincial agency for crop protection. In total, data from 1989 to 2008

for 20 sites were available.

Another utilized source of data was the European Climate Assessment (ECA) dataset,

developed as part of the European Union Framework 6 ENSEMBLES project, (Haylock et al.

2008, Klok & Klein Tank, 2009). This dataset has been elaborated into a gridded version, the E-

OBS gridded dataset, which includes data at 0.25° spatial resolution for maximum, minimum and

average daily air temperature, as well as rainfall. The whole gridded dataset was downloaded

from the web site http://eca.knmi.nl.

E-OBS data were used to reconstruct local time series through a downscaling procedure, while

locally measured data were used to calibrate and validate the procedure.

Soil cover data - Information about soil cover was used to focus the analysis only on the surface

actually covered by vineyards. The location of all vineyards plots was extracted from the 1:10000

soil cover map published in shapefile format by Veneto Region in the year 2009.

Downscaling approach - The spatial distribution of the surface air temperature at a given

moment is determined by the interactions between the state of the atmosphere and the terrain

features, e.g. elevation, slope or the presence of water bodies, whose effects changes depending

on the type of weather conditions. As an example, the lapse rate induced by elevation gradients

are known to vary throughout the year, following seasonal patterns. Since it is reasonable to think

that similar weather conditions should correspond to similar patterns of atmosphere-terrain

interaction, it was hypothesised that if a finite number of “weather types” could be defined

according to some classification criterion, then each type should correspond to a specific pattern

3 - 79



of atmosphere/terrain interaction, and ultimately to a specific pattern of temperature distribution.

Furthermore, if for each weather type, a given site could be characterized by a quantitative

relationship between its temperature, and the average zone temperature, then we would have a

convenient method to estimate the site time-series of temperature from the knowledge of the

weather type sequence and the time-series of the average zone temperature.

The idea was to use the E-OBS gridded dataset to individuate groups of days with similar

meteorology. As a classification criterion we used the minimum and maximum E-OBS

temperature values in the two grid-knots covering to the study area. These four daily temperature

time-series were subjected to k-means cluster analysis. The optimal number of groups was

searched in the interval 5-60. At the end, 45 groups gave the best results. For each group, the

mean difference between the measured and the E-OBS values, was calculated for each measuring

station in the interval where measurements were available. These differences were then used as

correction factor to estimate the station mean temperature from E-OBS data in the periods where

measurements were not available. The correction factors were calculated on half of the measures

dataset (odd days), and validated on the other half (even days). The accuracy of estimation was

assessed by calculating the Root Mean Square Error (RMSE) between estimates and

measurements.

Spatial interpolation of the reconstructed weather data - Once the whole climatic series for the

period 1950-2008 for each measuring station was reconstructed, daily mean temperature was

spatially extrapolated for the whole vineyard-covered area, at a 500 m spatial resolution.

Interpolation was carried out by multiple linear regression, using as predictors the elevation,

latitude, longitude, average slope and aspect, derived from a digital terrain model with ordinary

GIS procedures. The residuals between the estimates and the reconstructed data were then

interpolated with Inverse Distance Weighting, and used to correct the estimates.

Phenology modelling - The estimated temperature were used to run a phenology model (Fila et

al., 2010), predicting end of dormancy, bud break, flowering and veraison in the Glera cv. The

model was an extension of the Unified Model developed by Chuine (2000), which was calibrated

on 18 years of observations.

Analysis outline and statistical assessment - The interval from 1950 to 2008 was divided into

two sub periods: from 1950 to 1979 (period 1), and from 1980 to 2008 (period 2). With respect to

these two periods, the temperature variation in the vineyard-covered area was studied by

analysing the Winkler index (WI), the sum of the active temperature from 1 April to 31 October

(Σ(Tavg-10)) (Winkler, 1962). WI spatial variability was studied by calculating its distribution

frequencies across the studied area, while the interannual variability was quantified by means of

the Coefficient of Variation (CV), the ratio between standard deviation and the mean. The same

evaluation scheme was applied to the estimated dates for the principal phenological events: end

of dormancy, bud break, flowering and veraison.


E-OBS data underwent a preliminary explorative analysis to outline the main variation trends

(Fig.1). With respect to the average values calculated for the two sub periods, WI increased from

1485°C (period 1) to 1699°C (period 2).

1950 1960 1970 1980 1990 2000 2010

1200

1400

1600

1800

2000 Winkler index

annual means

average1950-1979

average1980-2008

Figure 1 – Winkler index annual means, calculated from the E -OBS gridded dataset. The

dashed horizontal lines represent the average for the two periods 1950-1979 and 1980-2008.

The downscaling procedure allowed to produce estimates of mean daily temperature with an

RMSE of 0.96 °C calculated on the validation subset. The 81.2 % of the residuals varied between

± 1°C and 92.5% between ± 1.5°C. This accuracy was considered sufficient to reconstruct the

whole time series, and to interpolate them at the spatial resolution of 500 m.

The spatial distribution of the estimated WI across the cultivated area is reported in Figure 2.

The index varied from 1100 to 2100 °C during period 1, and from about 1200 to 2200 °C in

period 2. WI increased homogeneously in the whole cultivated area, without relevant changes in

the shape of spatial distribution. The coefficient of variation, quantifying interannual variability,

increased from 7.5% to 8.9%. In this case, a modification of the distribution curve is appreciable:

in period 2 the mass of the distribution shifted towards the right, raising the curve skewness.

Winkler index (°C)

1000 1200 1400 1600 1800 2000 2200

% o

f vin

eyar

d-c

over

ed a

rea

0

10

20

30

1950-1979

1980-2008

Coefficient of Variation (%)

6 7 8 9 10 11

% o

f vin

eyar

d-c

over

ed a

rea

0

5

10

15

20

Figure 2 – Spatial distribution of the Winkler

index in the vineyard-covered area. Closed

symbols: 1950-1979 average; open symbols:

1980-2008 average.

Figure 3 - Spatial distribution of the

Coefficient of Variation, quantifying

interannual variability of the Winkler index

Closed and open symbols as in Fig. 2.

1950-1979

1980-2008

The generalized temperature increase had an impact on the estimated phenological behaviour.

The dates for the end of dormancy and bud break showed very little changes in their spatial

3 - 80



of atmosphere/terrain interaction, and ultimately to a specific pattern of temperature distribution.

Furthermore, if for each weather type, a given site could be characterized by a quantitative

relationship between its temperature, and the average zone temperature, then we would have a

convenient method to estimate the site time-series of temperature from the knowledge of the

weather type sequence and the time-series of the average zone temperature.

The idea was to use the E-OBS gridded dataset to individuate groups of days with similar

meteorology. As a classification criterion we used the minimum and maximum E-OBS

temperature values in the two grid-knots covering to the study area. These four daily temperature

time-series were subjected to k-means cluster analysis. The optimal number of groups was

searched in the interval 5-60. At the end, 45 groups gave the best results. For each group, the

mean difference between the measured and the E-OBS values, was calculated for each measuring

station in the interval where measurements were available. These differences were then used as

correction factor to estimate the station mean temperature from E-OBS data in the periods where

measurements were not available. The correction factors were calculated on half of the measures

dataset (odd days), and validated on the other half (even days). The accuracy of estimation was

assessed by calculating the Root Mean Square Error (RMSE) between estimates and

measurements.

Spatial interpolation of the reconstructed weather data - Once the whole climatic series for the

period 1950-2008 for each measuring station was reconstructed, daily mean temperature was

spatially extrapolated for the whole vineyard-covered area, at a 500 m spatial resolution.

Interpolation was carried out by multiple linear regression, using as predictors the elevation,

latitude, longitude, average slope and aspect, derived from a digital terrain model with ordinary

GIS procedures. The residuals between the estimates and the reconstructed data were then

interpolated with Inverse Distance Weighting, and used to correct the estimates.

Phenology modelling - The estimated temperature were used to run a phenology model (Fila et

al., 2010), predicting end of dormancy, bud break, flowering and veraison in the Glera cv. The

model was an extension of the Unified Model developed by Chuine (2000), which was calibrated

on 18 years of observations.

Analysis outline and statistical assessment - The interval from 1950 to 2008 was divided into

two sub periods: from 1950 to 1979 (period 1), and from 1980 to 2008 (period 2). With respect to

these two periods, the temperature variation in the vineyard-covered area was studied by

analysing the Winkler index (WI), the sum of the active temperature from 1 April to 31 October

(Σ(Tavg-10)) (Winkler, 1962). WI spatial variability was studied by calculating its distribution

frequencies across the studied area, while the interannual variability was quantified by means of

the Coefficient of Variation (CV), the ratio between standard deviation and the mean. The same

evaluation scheme was applied to the estimated dates for the principal phenological events: end

of dormancy, bud break, flowering and veraison.


E-OBS data underwent a preliminary explorative analysis to outline the main variation trends

(Fig.1). With respect to the average values calculated for the two sub periods, WI increased from

1485°C (period 1) to 1699°C (period 2).

1950 1960 1970 1980 1990 2000 2010

1200

1400

1600

1800

2000 Winkler index

annual means

average1950-1979

average1980-2008

Figure 1 – Winkler index annual means, calculated from the E -OBS gridded dataset. The

dashed horizontal lines represent the average for the two periods 1950-1979 and 1980-2008.

The downscaling procedure allowed to produce estimates of mean daily temperature with an

RMSE of 0.96 °C calculated on the validation subset. The 81.2 % of the residuals varied between

± 1°C and 92.5% between ± 1.5°C. This accuracy was considered sufficient to reconstruct the

whole time series, and to interpolate them at the spatial resolution of 500 m.

The spatial distribution of the estimated WI across the cultivated area is reported in Figure 2.

The index varied from 1100 to 2100 °C during period 1, and from about 1200 to 2200 °C in

period 2. WI increased homogeneously in the whole cultivated area, without relevant changes in

the shape of spatial distribution. The coefficient of variation, quantifying interannual variability,

increased from 7.5% to 8.9%. In this case, a modification of the distribution curve is appreciable:

in period 2 the mass of the distribution shifted towards the right, raising the curve skewness.

Winkler index (°C)

1000 1200 1400 1600 1800 2000 2200

% o

f vin

eyar

d-c

over

ed a

rea

0

10

20

30

1950-1979

1980-2008

Coefficient of Variation (%)

6 7 8 9 10 11

% o

f vin

eyar

d-c

over

ed a

rea

0

5

10

15

20

Figure 2 – Spatial distribution of the Winkler

index in the vineyard-covered area. Closed

symbols: 1950-1979 average; open symbols:

1980-2008 average.

Figure 3 - Spatial distribution of the

Coefficient of Variation, quantifying

interannual variability of the Winkler index

Closed and open symbols as in Fig. 2.

1950-1979

1980-2008

The generalized temperature increase had an impact on the estimated phenological behaviour.

The dates for the end of dormancy and bud break showed very little changes in their spatial

3 - 81



distribution and in their multi-year average. An anticipation of about 2 and 5 days was evident for

the occurrence of flowering and veraison respectively, without changes in the shape of spatial

distribution (Fig.4). It is interesting to note that the generalised temperature increase had a very

low impact on the fulfilment of grapevine chilling requirement.

-40 -20 0 20 40

% o

f vin

eyar

d-c

over

ed a

rea

0

10

20

30

40

50

1950-1979

1980-2008

end of

dormancy

70 80 90 100 110 120 130

budbreak

130 140 150 160 170 180 190

flowering

day of the year

180 200 220 240 260

veraison

Figure 4 – Spatial distribution of the estimated dates for the main phenological events. Closed and

open symbols as in Fig. 2.

The interannual variability showed more evident changes in both averages and spatial

distributions (Fig. 5). All phenological events showed an increase in the average CV from period

1 to period 2, except for the end of dormancy, where it decreased from 7.8 to 7.5 %, due to a

higher concentration of areas around the distribution peak. Bud break showed an average increase

of the CV, but the area maintained a variability range between 4 and 10%. Flowering and

veraison showed the least interannual variation, which was associated to a very little spatial

variation.

4 6 8 10% o

f vin

eyar

d-c

over

ed a

rea

0

10

20

30

40

50

60

4 6 8 10 2 3 4 5 6

coefficient of variation (%)

2 3 4

end of

dormancybudbreak flowering veraison

Figure 5 – Spatial distribution of the coefficient of variation for the estimated dates for the main

phenological eve nts. Closed and open symbols are as in Fig. 2.

1950-1979

1980-2008

CONCLUSIONS

An analysis of the spatio-temporal variation of daily temperature in a wine producing zone was

carried out by combining a downscaling and a spatial interpolation procedure to an European

gridded dataset. The spatial variation was analysed with reference to the periods 1950-1979 and

1980-2008. In general, the second period was characterized by an increase in the average

temperature and in the interannual variability.

The climatic change had little effects on the end of dormancy and bud break, while it induced

an anticipation of flowering and veraison.

More important changes were observed in the interannual variability, which were also

associated with variation in the spatial distribution across the vineyards.

In the time interval studied, the temperature changes do not seem to have relevance in the

zoning of the district. The increased interannual variability on the other hand, is more likely to

have an impact on the operational standpoint, making necessary to adjust the organization of

vineyard management.

ACKNOWLEDGEMENTS

We acknowledge the E-OBS dataset from the EU-FP6 project ENSEMBLES (http://ensembles-

eu.metoffice.com) and the data providers in the ECA&D project (http://eca.knmi.nl).

REFERENCES

1. Chuine I., 2000. A Unified Model for Budburst of Trees. Journal of Theoretical Biology

207(3): 337-347.

2. Fila G., Belvini P., Meggio F., Pitacco A. 2010. Validation of phenological models for

grapevine in the Veneto region. In: These Proceedings.3. Haylock M.R., Hofstra N., Klein Tank A.M.G., Klok E.J., Jones P.D., New M. 2008. A European daily

high-resolution gridded data set of surface temperature and precipitation for 1950–2006. Journal of

Geophysical Research (Atmospheres), 113, D20119, doi:10.1029/2008JD10201

4. Jones G.V., Davis R.E. 2000. Climate influences on grapevine phenology, grape

composition, and wine production and quality for Bordeaux, France. American Journal of

Enology and Viticulture, 51(3):249-261

5. Klok E. J., Klein Tank A. M. G. 2009. Updated and extended European dataset of daily

climate observations. International Journal of Climatology, 29: 1182–1191.

6. Winkler A., Cook J.A., Kliewer W.M., Lider L.A. 1974. General Viticulture. 2nd

revised edition. Berkeley and Los Angeles, California. University of California Press.

3 - 82



distribution and in their multi-year average. An anticipation of about 2 and 5 days was evident for

the occurrence of flowering and veraison respectively, without changes in the shape of spatial

distribution (Fig.4). It is interesting to note that the generalised temperature increase had a very

low impact on the fulfilment of grapevine chilling requirement.

-40 -20 0 20 40

% o

f vin

eyar

d-c

over

ed a

rea

0

10

20

30

40

50

1950-1979

1980-2008

end of

dormancy

70 80 90 100 110 120 130

budbreak

130 140 150 160 170 180 190

flowering

day of the year

180 200 220 240 260

veraison

Figure 4 – Spatial distribution of the estimated dates for the main phenological events. Closed and

open symbols as in Fig. 2.

The interannual variability showed more evident changes in both averages and spatial

distributions (Fig. 5). All phenological events showed an increase in the average CV from period

1 to period 2, except for the end of dormancy, where it decreased from 7.8 to 7.5 %, due to a

higher concentration of areas around the distribution peak. Bud break showed an average increase

of the CV, but the area maintained a variability range between 4 and 10%. Flowering and

veraison showed the least interannual variation, which was associated to a very little spatial

variation.

4 6 8 10% o

f vin

eyar

d-c

over

ed a

rea

0

10

20

30

40

50

60

4 6 8 10 2 3 4 5 6

coefficient of variation (%)

2 3 4

end of

dormancybudbreak flowering veraison

Figure 5 – Spatial distribution of the coefficient of variation for the estimated dates for the main

phenological eve nts. Closed and open symbols are as in Fig. 2.

1950-1979

1980-2008

CONCLUSIONS

An analysis of the spatio-temporal variation of daily temperature in a wine producing zone was

carried out by combining a downscaling and a spatial interpolation procedure to an European

gridded dataset. The spatial variation was analysed with reference to the periods 1950-1979 and

1980-2008. In general, the second period was characterized by an increase in the average

temperature and in the interannual variability.

The climatic change had little effects on the end of dormancy and bud break, while it induced

an anticipation of flowering and veraison.

More important changes were observed in the interannual variability, which were also

associated with variation in the spatial distribution across the vineyards.

In the time interval studied, the temperature changes do not seem to have relevance in the

zoning of the district. The increased interannual variability on the other hand, is more likely to

have an impact on the operational standpoint, making necessary to adjust the organization of

vineyard management.

ACKNOWLEDGEMENTS

We acknowledge the E-OBS dataset from the EU-FP6 project ENSEMBLES (http://ensembles-

eu.metoffice.com) and the data providers in the ECA&D project (http://eca.knmi.nl).

REFERENCES

1. Chuine I., 2000. A Unified Model for Budburst of Trees. Journal of Theoretical Biology

207(3): 337-347.

2. Fila G., Belvini P., Meggio F., Pitacco A. 2010. Validation of phenological models for

grapevine in the Veneto region. In: These Proceedings.3. Haylock M.R., Hofstra N., Klein Tank A.M.G., Klok E.J., Jones P.D., New M. 2008. A European daily

high-resolution gridded data set of surface temperature and precipitation for 1950–2006. Journal of

Geophysical Research (Atmospheres), 113, D20119, doi:10.1029/2008JD10201

4. Jones G.V., Davis R.E. 2000. Climate influences on grapevine phenology, grape


Enology and Viticulture, 51(3):249-261

5. Klok E. J., Klein Tank A. M. G. 2009. Updated and extended European dataset of daily

climate observations. International Journal of Climatology, 29: 1182–1191.

6. Winkler A., Cook J.A., Kliewer W.M., Lider L.A. 1974. General Viticulture. 2nd

revised edition. Berkeley and Los Angeles, California. University of California Press.

3 - 83



IMPORTANZA DEL MONITORAGGIO MICRO-METEOROLOGICO NELLA CARATTERIZZAZIONE DEL TERROIR

A. Matese(1), F. Di Gennaro(2) , L. Genesio(1) , F. P. Vaccari(1), F. Sabatini(1), M. Pieri (2) (1) Consiglio Nazionale delle Ricerche - Istituto of Biometeorologia (CNR-IBIMET)

Via G. Caproni, 8 50145 Firenze (Italia) [email protected]

(2) Società Consortile Tuscania S.r.l. – Piazza Strozzi, 1 50100 Firenze (Italia) [email protected]

RIASSUNTO Le variabili meteorologiche e micro-meteorologiche ricoprono un importante ruolo sulla

risposta vegeto-produttiva della vite e di conseguenza sulla qualità delle produzioni. Utilizzando una rete wireless di sensori sono stati monitorati i parametri meteorologici e micro-meteorologici di 4 vigneti del territorio toscano e in differenti condizioni di gestione agronomica. La comparazione di Land Indicators (indici calcolati a partire dal dato meteo territoriale proveniente da una stazione meteo tradizionale situata al di fuori del vigneto) e Proximity Indicator (indici calcolati dal dato meteo prossimale rilevato all’interno del vigneto) fa emergere come le due scale di indagine offrano una caratterizzazione del terroir significativamente diversa, in particolare per quanto concerne il ciclo giornaliero della temperatura del grappolo. Lo studio dell’impatto delle diverse pratiche di gestione della chioma sul micro-clima, ha evidenziato differenze tra le tesi defogliate e non, soprattutto nei valori di temperature massime e radiazione misurate a livello del grappolo. Questo studio evidenzia come il monitoraggio micro-meteorologico sia uno strumento efficace per ottenere delle sotto-zonazioni dei vigneti soprattutto in territori caratterizzati da morfologia eterogenea e quindi da grande variabilità spaziale dei parametri ambientali.

PAROLE CHIAVE Parametri micro-meteorologici, gestione della chioma, indicatori territoriali e prossimali.

ABSTRACT The micro-meteorological and meteorological variables play an important role on the

vegetative-productive response of the grapevine and consequently on quality products. Using a wireless sensor network, meteorological and micro-meteorological parameters of four Tuscany vineyards have been monitored and in different conditions of agronomic management. The comparison of Land Indicators (territorial data from a traditional weather station located outside the vineyard) and Proximity Indicators (proximal data monitored inside the vineyard) highlighted large differences especially with regard to the diurnal course of bunch temperature. The impact of different management practices on canopy microclimate pointed out significative differences between defoliated and non-thesis, especially in maximum temperature and solar radiation at bunch level. Present study emphasize the role of micro-meteorological monitoring in providing a reliable picture of vineyard sub-zones that can be useful in those areas characterized by an heterogeneous morphology and hence by a large spatial variability of environmental parameters.

KEYWORD Micro-meteorological parameters, canopy management, Land and Proximity indicators. INTRODUZIONE I vigneti si presentano come ambienti caratterizzati da una elevata eterogeneità dovuta a fattori

strutturali quali la morfologia e il tipo di suolo ed altri dinamici quali le pratiche colturali e l’andamento meteorologico stagionale. Le variabili meteorologiche e micro-meteorologiche ricoprono un importante ruolo sulla risposta vegeto-produttiva della vite e di conseguenza sulla qualità delle produzioni. Lo studio eseguito nell’ambito del progetto di ricerca “Consorzio Tuscania”, ha sviluppato una rete wireless di sensori (NAV-Network Avanzato per il Vigneto - Matese et al., 2009), al fine di monitorare i parametri meteorologici e micro-meteorologici di 3 tipiche zone a vocazione vitivinicola della Toscana. I vigneti sperimentali sono situati presso le aziende di Barone Ricasoli (Castello di Brolio Chianti Classico DOCG), Donna Olimpia 1898 (Bolgheri DOC) e Tenuta Le Mortelle (Monteregio di Massa Marittima DOC).

L’esposizione dei grappoli e la loro temperatura è molto importante per la composizione e il metabolismo degli acini (Spayd et al., 2002). Infatti la radiazione solare diretta sui grappoli aumenta la loro temperatura che, durante il giorno, può raggiungere valori più elevati di 11°C (o più) rispetto a quella di grappoli non esposti (Kliewer, Lider, 1968; Spayd et al., 2002). Poca radiazione intercettata dai grappoli può limitare l’accumulo degli antociani (Dokoozlian, Kliewer, 1996; Bergqvist et al., 2001; Spayd et al., 2002). Queste premesse confermano l’importanza del monitoraggio meteorologico, ma soprattutto micro-meteorologico nella caratterizzazione di terroirs di qualità.

Il primo obiettivo di questa ricerca è il confronto di Land Indicator e Proximity Indicator per capire quanto il dato territoriale proveniente da una stazione meteo tradizionale situata al di fuori del vigneto possa considerarsi rappresentativo delle condizioni che si verificano all'interno della parete vegetale in termini di valori assoluti e di dinamica giornaliera e come gli scostamenti varino tra diversi vigneti. La sperimentazione del Consorzio Tuscania ha previsto inoltre, all'interno dei vigneti sperimentali, la disposizione di 14 blocchi sperimentali distribuiti su zone omogenee di Vigore Vegetativo (mappe NDVI) a loro volta individuate per mezzo dell'analisi multi spettrale di foto aeree. All'interno di ciascun Blocco Sperimentale sono state definite n. 8 Tesi Sperimentali, relative alle differenti combinazioni delle modalità di gestione della chioma indagate (Carica di gemme, Sfogliatura precoce, Diradamento dei grappoli e loro relative interazioni), tutto questo con l’obiettivo di capire l’impatto delle diverse pratiche di gestione della chioma sul micro-clima e sulla qualità finale dell’uva

MATERIALI E METODI La struttura del sistema NAV si basa su una rete di stazioni meteorologiche wireless, costituita

da una stazione agrometeorologica base (Unità Master) posta al di fuori del vigneto e una serie di stazioni periferiche (Unità Slave) per la raccolta dei parametri micro-meteorologici poste all’interno dei filari (Fig.1). La stazione Master è equipaggiata con la sensoristica necessaria alla misura dei principali parametri agrometeorologici quali: temperatura ed umidità dell’aria, pressione atmosferica, radiazione globale, precipitazione, velocità e direzione del vento.

Le unità periferiche sono state progettate per essere collocate all’interno della chioma, così da rilevare i parametri micro-meteorologici sulla pianta.

Lo studio qui descritto si concentra sui parametri di radiazione e temperatura del grappolo.

3 - 84



IMPORTANZA DEL MONITORAGGIO MICRO-METEOROLOGICO NELLA CARATTERIZZAZIONE DEL TERROIR

A. Matese(1), F. Di Gennaro(2) , L. Genesio(1) , F. P. Vaccari(1), F. Sabatini(1), M. Pieri (2) (1) Consiglio Nazionale delle Ricerche - Istituto of Biometeorologia (CNR-IBIMET)

Via G. Caproni, 8 50145 Firenze (Italia) [email protected]

(2) Società Consortile Tuscania S.r.l. – Piazza Strozzi, 1 50100 Firenze (Italia) [email protected]

RIASSUNTO Le variabili meteorologiche e micro-meteorologiche ricoprono un importante ruolo sulla

risposta vegeto-produttiva della vite e di conseguenza sulla qualità delle produzioni. Utilizzando una rete wireless di sensori sono stati monitorati i parametri meteorologici e micro-meteorologici di 4 vigneti del territorio toscano e in differenti condizioni di gestione agronomica. La comparazione di Land Indicators (indici calcolati a partire dal dato meteo territoriale proveniente da una stazione meteo tradizionale situata al di fuori del vigneto) e Proximity Indicator (indici calcolati dal dato meteo prossimale rilevato all’interno del vigneto) fa emergere come le due scale di indagine offrano una caratterizzazione del terroir significativamente diversa, in particolare per quanto concerne il ciclo giornaliero della temperatura del grappolo. Lo studio dell’impatto delle diverse pratiche di gestione della chioma sul micro-clima, ha evidenziato differenze tra le tesi defogliate e non, soprattutto nei valori di temperature massime e radiazione misurate a livello del grappolo. Questo studio evidenzia come il monitoraggio micro-meteorologico sia uno strumento efficace per ottenere delle sotto-zonazioni dei vigneti soprattutto in territori caratterizzati da morfologia eterogenea e quindi da grande variabilità spaziale dei parametri ambientali.

PAROLE CHIAVE Parametri micro-meteorologici, gestione della chioma, indicatori territoriali e prossimali.

ABSTRACT The micro-meteorological and meteorological variables play an important role on the

vegetative-productive response of the grapevine and consequently on quality products. Using a wireless sensor network, meteorological and micro-meteorological parameters of four Tuscany vineyards have been monitored and in different conditions of agronomic management. The comparison of Land Indicators (territorial data from a traditional weather station located outside the vineyard) and Proximity Indicators (proximal data monitored inside the vineyard) highlighted large differences especially with regard to the diurnal course of bunch temperature. The impact of different management practices on canopy microclimate pointed out significative differences between defoliated and non-thesis, especially in maximum temperature and solar radiation at bunch level. Present study emphasize the role of micro-meteorological monitoring in providing a reliable picture of vineyard sub-zones that can be useful in those areas characterized by an heterogeneous morphology and hence by a large spatial variability of environmental parameters.

KEYWORD Micro-meteorological parameters, canopy management, Land and Proximity indicators. INTRODUZIONE I vigneti si presentano come ambienti caratterizzati da una elevata eterogeneità dovuta a fattori

strutturali quali la morfologia e il tipo di suolo ed altri dinamici quali le pratiche colturali e l’andamento meteorologico stagionale. Le variabili meteorologiche e micro-meteorologiche ricoprono un importante ruolo sulla risposta vegeto-produttiva della vite e di conseguenza sulla qualità delle produzioni. Lo studio eseguito nell’ambito del progetto di ricerca “Consorzio Tuscania”, ha sviluppato una rete wireless di sensori (NAV-Network Avanzato per il Vigneto - Matese et al., 2009), al fine di monitorare i parametri meteorologici e micro-meteorologici di 3 tipiche zone a vocazione vitivinicola della Toscana. I vigneti sperimentali sono situati presso le aziende di Barone Ricasoli (Castello di Brolio Chianti Classico DOCG), Donna Olimpia 1898 (Bolgheri DOC) e Tenuta Le Mortelle (Monteregio di Massa Marittima DOC).

L’esposizione dei grappoli e la loro temperatura è molto importante per la composizione e il metabolismo degli acini (Spayd et al., 2002). Infatti la radiazione solare diretta sui grappoli aumenta la loro temperatura che, durante il giorno, può raggiungere valori più elevati di 11°C (o più) rispetto a quella di grappoli non esposti (Kliewer, Lider, 1968; Spayd et al., 2002). Poca radiazione intercettata dai grappoli può limitare l’accumulo degli antociani (Dokoozlian, Kliewer, 1996; Bergqvist et al., 2001; Spayd et al., 2002). Queste premesse confermano l’importanza del monitoraggio meteorologico, ma soprattutto micro-meteorologico nella caratterizzazione di terroirs di qualità.

Il primo obiettivo di questa ricerca è il confronto di Land Indicator e Proximity Indicator per capire quanto il dato territoriale proveniente da una stazione meteo tradizionale situata al di fuori del vigneto possa considerarsi rappresentativo delle condizioni che si verificano all'interno della parete vegetale in termini di valori assoluti e di dinamica giornaliera e come gli scostamenti varino tra diversi vigneti. La sperimentazione del Consorzio Tuscania ha previsto inoltre, all'interno dei vigneti sperimentali, la disposizione di 14 blocchi sperimentali distribuiti su zone omogenee di Vigore Vegetativo (mappe NDVI) a loro volta individuate per mezzo dell'analisi multi spettrale di foto aeree. All'interno di ciascun Blocco Sperimentale sono state definite n. 8 Tesi Sperimentali, relative alle differenti combinazioni delle modalità di gestione della chioma indagate (Carica di gemme, Sfogliatura precoce, Diradamento dei grappoli e loro relative interazioni), tutto questo con l’obiettivo di capire l’impatto delle diverse pratiche di gestione della chioma sul micro-clima e sulla qualità finale dell’uva

MATERIALI E METODI La struttura del sistema NAV si basa su una rete di stazioni meteorologiche wireless, costituita

da una stazione agrometeorologica base (Unità Master) posta al di fuori del vigneto e una serie di stazioni periferiche (Unità Slave) per la raccolta dei parametri micro-meteorologici poste all’interno dei filari (Fig.1). La stazione Master è equipaggiata con la sensoristica necessaria alla misura dei principali parametri agrometeorologici quali: temperatura ed umidità dell’aria, pressione atmosferica, radiazione globale, precipitazione, velocità e direzione del vento.

Le unità periferiche sono state progettate per essere collocate all’interno della chioma, così da rilevare i parametri micro-meteorologici sulla pianta.

Lo studio qui descritto si concentra sui parametri di radiazione e temperatura del grappolo.

3 - 85



Il sensore di temperatura del grappolo (termocoppia tipo T) viene installato in un grappolo campione sul rachide a circa metà della lunghezza dello stesso fissato in modo tale da non creare una strozzatura sul rachide. La scelta di questo posizionamento è stata valutata in base alla necessità di ottenere una misura della temperatura che fosse rappresentativa delle condizioni termiche medie all’interno del grappolo. Il sensore di radiazione solare è stato realizzato utilizzando una fotocellula inserita in un diffusore di teflon, in modo da intercettare sia la radiazione incidente che quella diffusa, ed è stato posizionato al di sopra del grappolo campione.

Fig.1: Sensori delle stazioni Slave (sinistra) e Master (destra).

Il grappolo campione è stato scelto come rappresentativo di una determinata pratica di gestione e non come massima espressione della pratica stessa. Quindi, per esempio, per la tesi sfogliata non è stato scelto il grappolo più esposto, ma quello rappresentativo della media esposizione nella area sfogliata.

RISULTATI E DISCUSSIONE Il sistema ha dimostrato interessanti risultati in termini di caratterizzazione climatica del

vigneto. Il Land Indicator è stato utilizzato per il calcolo dei principali indici bioclimatici utilizzati nella descrizione di un terroir viticolo, quali Indice di Winkler (WI), l’Indice di Huglin (HI), la somma delle escursioni termiche (SET), la somma delle precipitazioni nel periodo vegetativo aprile-settembre (Papr-set).

Fig. 2: Caratterizzazione climatica dei terroirs nelle annate 2008 e 2009.

Utilizzando i dati raccolti da ciascuna stazione Master, è stato possibile ottenere un quadro dettagliato della variabilità del dato meteo tra i vigneti sperimentali delle 3 zone viticole prese in esame (Fig.2). L’annata 2009 è risultata generalmente più calda del 2008 in tutti i vigneti come si

vede per gli indici IH e IW. Particolarmente elevato è risultato l’indice di Winkler per il 2009, circa 170 Gradi Giorno a Brolio rispetto al 2008. Per quanto concerne l’indice IH anche qui con un delta positivo a Brolio di circa 259 Gradi Giorno per il 2009 rispetto al 2008. Le escursioni termiche sono state simili nei due anni. Confrontando invece i vari vigneti si notano valori più alti a Brolio per quanto riguarda la sommatorie delle escursioni termiche che raggiungono circa 1860 gradi Giorno, mentre il vigneto di Le Mortelle presenta gli indici IW e IH più alti (2721 per il 2008 e 2833 per il 2009). Interessante notare come il vigneto di Donna Olimpia presenta escursioni termiche minori (1623 per il 2008 e 1690 per il 2009) probabilmente a causa della locazione che risente dell’effetto del mare. Le Precipitazioni di Aprile-Settembre maggiori si sono avute a Brolio nel 2008 con circa 336 mm, mentre è pochissima la differenza tra le piogge cadute nei vigneti nel 2009 (circa 260 mm)

Per quanto riguarda il Proximity Indicator, il sistema di monitoraggio multipuntuale svolto dalle stazioni Slave, permette una caratterizzazione micro-meteorologica che consente di individuare la variabilità interna al vigneto, e nello specifico sulle tesi di gestione della canopy . I risultati sono riferiti alle medie dei valori nelle tesi sui parametri di temperatura del grappolo.

Fig.3: Temperature del grappolo massime, medie e minime delle tesi defogliate e non defogliate nei periodi di preinvaiatura e nell’invaiatura per il vigneto del Castello di Brolio (Fig.3a) e della Tenuta Le Mortelle (Fig.3b). Nelle Fig.3c (Castello di Brolio) e Fig.3d (Tenuta Le Mortelle),

sono delle espresse le temperature del grappolo delle tesi potate a 1 e 3 gemme.

Nella Fig. 3a vengono riportate le temperature del grappolo massime, medie e minime delle tesi defogliate e non defogliate nei due periodi presi in esame, ossia nella preinvaiatura (15 Giugno – 31 Luglio) e dall’invaiatura alla raccolta (1 Agosto – 15 Settembre) nel vigneto di Brolio. Si osservano temperature superiori nel secondo periodo di circa 5°C. Il trattamento di sfogliatura causa un incremento delle temperature di circa 2°C per quanto riguarda le temperature massime mentre scarsa è la sua influenza sulle temperature medie o minime, questo a causa della maggior incidenza della pratica nelle ore centrali del giorno che ricevono un maggior irraggiamento. Nella Fig. 3b sono rappresentati i dati di temperatura del grappolo relativi al vigneto di Le Mortelle; nella tesi sfogliata le temperature massime sono più alte di oltre 1.5°C, mentre per quanto riguarda le temperature minime, in particolare dopo l’invaiatura, queste si presentano inferiori presumibilmente a causa dell’effetto di schermatura della radiazione uscente e della riduzione di

3 - 86



Il sensore di temperatura del grappolo (termocoppia tipo T) viene installato in un grappolo campione sul rachide a circa metà della lunghezza dello stesso fissato in modo tale da non creare una strozzatura sul rachide. La scelta di questo posizionamento è stata valutata in base alla necessità di ottenere una misura della temperatura che fosse rappresentativa delle condizioni termiche medie all’interno del grappolo. Il sensore di radiazione solare è stato realizzato utilizzando una fotocellula inserita in un diffusore di teflon, in modo da intercettare sia la radiazione incidente che quella diffusa, ed è stato posizionato al di sopra del grappolo campione.

Fig.1: Sensori delle stazioni Slave (sinistra) e Master (destra).

Il grappolo campione è stato scelto come rappresentativo di una determinata pratica di gestione e non come massima espressione della pratica stessa. Quindi, per esempio, per la tesi sfogliata non è stato scelto il grappolo più esposto, ma quello rappresentativo della media esposizione nella area sfogliata.

RISULTATI E DISCUSSIONE Il sistema ha dimostrato interessanti risultati in termini di caratterizzazione climatica del

vigneto. Il Land Indicator è stato utilizzato per il calcolo dei principali indici bioclimatici utilizzati nella descrizione di un terroir viticolo, quali Indice di Winkler (WI), l’Indice di Huglin (HI), la somma delle escursioni termiche (SET), la somma delle precipitazioni nel periodo vegetativo aprile-settembre (Papr-set).

Fig. 2: Caratterizzazione climatica dei terroirs nelle annate 2008 e 2009.

Utilizzando i dati raccolti da ciascuna stazione Master, è stato possibile ottenere un quadro dettagliato della variabilità del dato meteo tra i vigneti sperimentali delle 3 zone viticole prese in esame (Fig.2). L’annata 2009 è risultata generalmente più calda del 2008 in tutti i vigneti come si

vede per gli indici IH e IW. Particolarmente elevato è risultato l’indice di Winkler per il 2009, circa 170 Gradi Giorno a Brolio rispetto al 2008. Per quanto concerne l’indice IH anche qui con un delta positivo a Brolio di circa 259 Gradi Giorno per il 2009 rispetto al 2008. Le escursioni termiche sono state simili nei due anni. Confrontando invece i vari vigneti si notano valori più alti a Brolio per quanto riguarda la sommatorie delle escursioni termiche che raggiungono circa 1860 gradi Giorno, mentre il vigneto di Le Mortelle presenta gli indici IW e IH più alti (2721 per il 2008 e 2833 per il 2009). Interessante notare come il vigneto di Donna Olimpia presenta escursioni termiche minori (1623 per il 2008 e 1690 per il 2009) probabilmente a causa della locazione che risente dell’effetto del mare. Le Precipitazioni di Aprile-Settembre maggiori si sono avute a Brolio nel 2008 con circa 336 mm, mentre è pochissima la differenza tra le piogge cadute nei vigneti nel 2009 (circa 260 mm)

Per quanto riguarda il Proximity Indicator, il sistema di monitoraggio multipuntuale svolto dalle stazioni Slave, permette una caratterizzazione micro-meteorologica che consente di individuare la variabilità interna al vigneto, e nello specifico sulle tesi di gestione della canopy . I risultati sono riferiti alle medie dei valori nelle tesi sui parametri di temperatura del grappolo.

Fig.3: Temperature del grappolo massime, medie e minime delle tesi defogliate e non defogliate nei periodi di preinvaiatura e nell’invaiatura per il vigneto del Castello di Brolio (Fig.3a) e della Tenuta Le Mortelle (Fig.3b). Nelle Fig.3c (Castello di Brolio) e Fig.3d (Tenuta Le Mortelle),

sono delle espresse le temperature del grappolo delle tesi potate a 1 e 3 gemme.

Nella Fig. 3a vengono riportate le temperature del grappolo massime, medie e minime delle tesi defogliate e non defogliate nei due periodi presi in esame, ossia nella preinvaiatura (15 Giugno – 31 Luglio) e dall’invaiatura alla raccolta (1 Agosto – 15 Settembre) nel vigneto di Brolio. Si osservano temperature superiori nel secondo periodo di circa 5°C. Il trattamento di sfogliatura causa un incremento delle temperature di circa 2°C per quanto riguarda le temperature massime mentre scarsa è la sua influenza sulle temperature medie o minime, questo a causa della maggior incidenza della pratica nelle ore centrali del giorno che ricevono un maggior irraggiamento. Nella Fig. 3b sono rappresentati i dati di temperatura del grappolo relativi al vigneto di Le Mortelle; nella tesi sfogliata le temperature massime sono più alte di oltre 1.5°C, mentre per quanto riguarda le temperature minime, in particolare dopo l’invaiatura, queste si presentano inferiori presumibilmente a causa dell’effetto di schermatura della radiazione uscente e della riduzione di

3 - 87



circolazione operata dalla canopy. Per quanto riguarda invece l’effetto del trattamento di potatura sulle dinamiche della temperatura del grappolo, si può osservare nelle Fig. 3c e 3d, che le differenze tra tesi in entrambi i due periodi sono minime, nonostante il periodo dell’invaiatura continui ad essere caratterizzato da temperature fino a 4°C superiori, sia per il vigneto di Brolio che per Le Mortelle.

Nella Fig. 4 sono mostrati i cicli giornalieri della temperatura media del grappolo, ossia le variazioni della temperatura nell’arco di una giornata, in funzione del trattamento di sfogliatura, per i due periodi considerati nei vigneti di Brolio e Donna Olimpia durante l’annata 2009. Analizzando il ciclo giornaliero si nota che la mattina il Land Indicator (Master) ha temperature più alte di circa 2°C del Proximity, questo probabilmente è imputabile a piccoli fenomeni di inversione termica che si presentano negli strati più bassi dove appunto sono installati i sensori nel grappolo, mentre i sensori di temperatura delle stazioni meteo Master (Land Indicator) sono a circa 2 m di altezza. Di giorno invece, i Proximity Indicator presentano temperature più alte a causa della ridotta circolazione dovuta alla barriera costituita dalla vegetazione. Nel vigneto Brolio l’escursione termica maggiore tra Proximity e Land si presenta nel periodo dell’invaiatura presentando un delta termico di circa 5°C nelle ore centrali del giorno, mentre per quanto riguarda Donna Olimpia si hanno solo piccole differenze nei due periodi. Le tesi sfogliate presentano temperature medie del grappolo superiori di circa 2°C rispetto alle tesi non sfogliate nel vigneto di Brolio nelle ore centrali del giorno, mentre nel vigneto di Donna Olimpia queste differenze sono più contenute. In generale si nota un più rapido incremento delle temperature al mattino nelle tesi sfogliate.

S F O G L IA T A N O N S F O G L IA T AS F O G L IA T A N O N S F O G L IA T A

Fig.4: Cicli giornalieri della temperatura media del grappolo nei vigneti di Brolio e Donna Olimpia, per la tesi defogliata, non defogliata e per la Master.

Analizzando il parametro della radiazione solare globale all’interno dei vigneti sperimentali, è emerso che l’effetto delle pratiche di gestione della canopy ha avuto un ruolo rilevante.

Castello di Brolio (Siena) - 2009

800

900

1000

1100

1200

1300

sfogliato non sfogliato 1 gemma 3 gemme

Mj/m

2

preinvaiatura invaiatura

Fig.5: Cumulata radiativa del vigneto di Brolio nell’annata 2009

Nella Fig. 5 è portata ad esempio la caratterizzazione radiativa del vigneto di Brolio nell’annata 2009. In questo grafico si presenta il dato della radiazione cumulata incidente sul grappolo, per tesi sfogliate e non sfogliate, e per tesi potate a 1 gemma o a 3 gemme, distinto nei due periodi analizzati. Il periodo di invaiatura presenta una radiazione giornaliera che aumenta di circa 250 MJ/m2 rispetto al periodo di preinvaiatura. In entrambi i periodi analizzati la tesi defogliata è caratterizzata da una radiazione solare superiore di circa il 10%. Per quanto riguarda invece la potatura non si hanno grosse differenze, probabilmente perché questa pratica non influenza in modo evidente l’intercettazione della radiazione incidente.

CONCLUSIONI I principali studi di zonazione viticola analizzano le caratteristiche climatiche del territorio,

sulla base di una o poche stazioni meteorologiche presenti nella zona del vigneto e spesso poco rappresentative. Inoltre, la risoluzione temporale dei dati acquisiti non tiene conto delle piccole variazioni che si hanno durante il ciclo giornaliero. Utilizzando i parametri meteorologici e micro-meteorologici monitorati nei vigneti sperimentali è stato possibile caratterizzare dal punto di vista climatico le differenti zone viticole e indagare come alcune pratiche viticole normalmente attuate dai viticoltori per la gestione della chioma (carica di gemme, sfogliatura precoce, diradamento del grappolo), possano influenzare direttamente ed indirettamente (attraverso l’interazione con altri fattori) il microclima. In generale abbiamo osservato come il Proximity indicator riesca a rilevare gli effetti delle pratiche colturali di gestione della chioma. Questo inoltre è risultato più sensibile nel rilevare alcuni indici e differenze termiche utili per la valutazione della fisiologia della pianta come nel caso dell’escursione termica, che è risultata maggiore nella rilevazione del Proximity Indicator ripetto a quella del Land Indicator.

RINGRAZIAMENTI Si ringraziano Lorenzo Albanese, Giacomo Tagliaferri, Piero Toscano e Alessandro Zaldei di

CNR – IBIMET di Firenze per il loro supporto in alcune fasi del progetto, Stefano Di Blasi, Alessandra Biondi Bartolini lo staff della “Società Consortile Tuscania S.r.l.".

Progetto coordinato e finanziato da: Piazza Strozzi 1 – Firenze

BIBLIOGRAFIA Bergqvist J., Dokoozlian N., Ebisuda N., 2001. Sunlight exposure and temperature effects on berry

growth and composition of ‘Cabernet Sauvignon’ and ‘Grenache’ in the Central San Joaquin Valley of California. Am. J. Enol. Vitic., 52, 1, 1-7.

Dokoozlian N.K., Kliewer W.M., 1996. Influence of light on grape berry growth and composition varies during fruit development. J. Amer. Soc. Hort. Sci., 121, 5, 869-874.

Kliewer W.M., Lider L.A., 1968. Influence of cluster exposure to the sun on the composition of ‘Thompson Seedless’ fruit. Am. J. Enol. Vitic., 19, 175-184.

Matese A., Di Gennaro S.F., Zaldei A., Genesio L., Vaccari F.P., 2009. A Wireless sensor network for precision viticulture: The NAV system. Computers and Electronics in Agriculture 69, 51-58.

Spayd S.E., Tarara J.M., Mee D.L., Ferguson J.C., 2002. Separation of sunlight and temperature effects on the composition of Vitis vinifera ‘Merlot’ berries. Am. J. Enol. Vitic., 53, 3, 171-182.

3 - 88



circolazione operata dalla canopy. Per quanto riguarda invece l’effetto del trattamento di potatura sulle dinamiche della temperatura del grappolo, si può osservare nelle Fig. 3c e 3d, che le differenze tra tesi in entrambi i due periodi sono minime, nonostante il periodo dell’invaiatura continui ad essere caratterizzato da temperature fino a 4°C superiori, sia per il vigneto di Brolio che per Le Mortelle.

Nella Fig. 4 sono mostrati i cicli giornalieri della temperatura media del grappolo, ossia le variazioni della temperatura nell’arco di una giornata, in funzione del trattamento di sfogliatura, per i due periodi considerati nei vigneti di Brolio e Donna Olimpia durante l’annata 2009. Analizzando il ciclo giornaliero si nota che la mattina il Land Indicator (Master) ha temperature più alte di circa 2°C del Proximity, questo probabilmente è imputabile a piccoli fenomeni di inversione termica che si presentano negli strati più bassi dove appunto sono installati i sensori nel grappolo, mentre i sensori di temperatura delle stazioni meteo Master (Land Indicator) sono a circa 2 m di altezza. Di giorno invece, i Proximity Indicator presentano temperature più alte a causa della ridotta circolazione dovuta alla barriera costituita dalla vegetazione. Nel vigneto Brolio l’escursione termica maggiore tra Proximity e Land si presenta nel periodo dell’invaiatura presentando un delta termico di circa 5°C nelle ore centrali del giorno, mentre per quanto riguarda Donna Olimpia si hanno solo piccole differenze nei due periodi. Le tesi sfogliate presentano temperature medie del grappolo superiori di circa 2°C rispetto alle tesi non sfogliate nel vigneto di Brolio nelle ore centrali del giorno, mentre nel vigneto di Donna Olimpia queste differenze sono più contenute. In generale si nota un più rapido incremento delle temperature al mattino nelle tesi sfogliate.

S F O G L IA T A N O N S F O G L IA T AS F O G L IA T A N O N S F O G L IA T A

Fig.4: Cicli giornalieri della temperatura media del grappolo nei vigneti di Brolio e Donna Olimpia, per la tesi defogliata, non defogliata e per la Master.

Analizzando il parametro della radiazione solare globale all’interno dei vigneti sperimentali, è emerso che l’effetto delle pratiche di gestione della canopy ha avuto un ruolo rilevante.

Castello di Brolio (Siena) - 2009

800

900

1000

1100

1200

1300

sfogliato non sfogliato 1 gemma 3 gemme

Mj/m

2

preinvaiatura invaiatura

Fig.5: Cumulata radiativa del vigneto di Brolio nell’annata 2009

Nella Fig. 5 è portata ad esempio la caratterizzazione radiativa del vigneto di Brolio nell’annata 2009. In questo grafico si presenta il dato della radiazione cumulata incidente sul grappolo, per tesi sfogliate e non sfogliate, e per tesi potate a 1 gemma o a 3 gemme, distinto nei due periodi analizzati. Il periodo di invaiatura presenta una radiazione giornaliera che aumenta di circa 250 MJ/m2 rispetto al periodo di preinvaiatura. In entrambi i periodi analizzati la tesi defogliata è caratterizzata da una radiazione solare superiore di circa il 10%. Per quanto riguarda invece la potatura non si hanno grosse differenze, probabilmente perché questa pratica non influenza in modo evidente l’intercettazione della radiazione incidente.

CONCLUSIONI I principali studi di zonazione viticola analizzano le caratteristiche climatiche del territorio,

sulla base di una o poche stazioni meteorologiche presenti nella zona del vigneto e spesso poco rappresentative. Inoltre, la risoluzione temporale dei dati acquisiti non tiene conto delle piccole variazioni che si hanno durante il ciclo giornaliero. Utilizzando i parametri meteorologici e micro-meteorologici monitorati nei vigneti sperimentali è stato possibile caratterizzare dal punto di vista climatico le differenti zone viticole e indagare come alcune pratiche viticole normalmente attuate dai viticoltori per la gestione della chioma (carica di gemme, sfogliatura precoce, diradamento del grappolo), possano influenzare direttamente ed indirettamente (attraverso l’interazione con altri fattori) il microclima. In generale abbiamo osservato come il Proximity indicator riesca a rilevare gli effetti delle pratiche colturali di gestione della chioma. Questo inoltre è risultato più sensibile nel rilevare alcuni indici e differenze termiche utili per la valutazione della fisiologia della pianta come nel caso dell’escursione termica, che è risultata maggiore nella rilevazione del Proximity Indicator ripetto a quella del Land Indicator.

RINGRAZIAMENTI Si ringraziano Lorenzo Albanese, Giacomo Tagliaferri, Piero Toscano e Alessandro Zaldei di

CNR – IBIMET di Firenze per il loro supporto in alcune fasi del progetto, Stefano Di Blasi, Alessandra Biondi Bartolini lo staff della “Società Consortile Tuscania S.r.l.".

Progetto coordinato e finanziato da: Piazza Strozzi 1 – Firenze

BIBLIOGRAFIA Bergqvist J., Dokoozlian N., Ebisuda N., 2001. Sunlight exposure and temperature effects on berry

growth and composition of ‘Cabernet Sauvignon’ and ‘Grenache’ in the Central San Joaquin Valley of California. Am. J. Enol. Vitic., 52, 1, 1-7.

Dokoozlian N.K., Kliewer W.M., 1996. Influence of light on grape berry growth and composition varies during fruit development. J. Amer. Soc. Hort. Sci., 121, 5, 869-874.

Kliewer W.M., Lider L.A., 1968. Influence of cluster exposure to the sun on the composition of ‘Thompson Seedless’ fruit. Am. J. Enol. Vitic., 19, 175-184.

Matese A., Di Gennaro S.F., Zaldei A., Genesio L., Vaccari F.P., 2009. A Wireless sensor network for precision viticulture: The NAV system. Computers and Electronics in Agriculture 69, 51-58.

Spayd S.E., Tarara J.M., Mee D.L., Ferguson J.C., 2002. Separation of sunlight and temperature effects on the composition of Vitis vinifera ‘Merlot’ berries. Am. J. Enol. Vitic., 53, 3, 171-182.

3 - 89



FUTURE SCENARIOS FOR VITICULTURAL CLIMATIC ZONING

IN EUROPE

A. C. Malheiro (1)

, J. A. Santos (1)

, H. Fraga (1)

, J. G. Pinto (2)

(1) Centre for Research and Technology of Agro-Environment and Biological Sciences (CITAB), University of Trás-

os-Montes e Alto Douro, 5001-801 Vila Real, Portugal ([email protected]) (2)

Institut für Geophysik und Meteorologie, Universität zu Köln, Kerpener Str. 13, 50923 Köln, Germany

ABSTRACT

Climate is one of the main conditioning factors of winemaking. In this context, bioclimatic

indices are a useful zoning tool, allowing the description of the suitability of a particular region

for wine production. In this study, we compute climatic indices for Europe, characterize regions

with different viticultural aptitude, and assess possible variations in these regions under a future

climate conditions using a state-of-the-art regional climate model. The indices are calculated

from climatic variables (mostly daily temperatures and precipitation) obtained from the regional

climate model COSMO-CLM for recent and future climate conditions. Maps of theses indices for

recent decades (1961-2000) and for the XXI century (following the SRES A1B scenario) are

considered to identify possible changes. Results show that climate change is projected to have a

significant negative impact in wine quality by increased dryness and cumulative thermal effects

during growing seasons in Southern European regions (e.g. Portugal, Spain and Italy). These

changes represent an important constraint to grapevine growth and development, making crucial

adaptation/mitigation strategies to be adopted. On the other hand, regions of western and central

Europe (e.g. southern Britain, northern France and Germany) will benefit from this scenario both

in wine quality, and in new potential areas for viticulture. This approach provides a macro-

characterization of European areas where grapevines may preferentially grow, as well as their

projected changes, and is thus a valuable tool for viticultural zoning in a changing climate.

KEYWORD

Viticultural zoning – scenarios – Europe – climate change – CLM

INTRODUCTION

Winemaking has a predominant economic, social and environmental relevance in Europe.

Studies addressing the influence of climate variability and change in viticulture are particularly

pertinent, as climate is the dominant factor from grapevine yield and quality (e.g. van Leeuwen et

al., 2004) to its geographic worldwide distribution (Jones, 2006). In fact, Vitis vinifera has well-

defined climatic requirements; it is a heat demanding crop, needing proper high radiation

intensities and temperatures, not only during its vegetative growth and development, but also for

berry maturation and it is highly sensitive to frost occurrences in winter and spring (Magalhães,

2008). Due to these very selective requirements, most wine-producing areas are geographically

located between latitudes 30-50º of the Northern Hemisphere (Spellman, 1999), where the “warm

temperate climates” (Kottek et al., 2006), including the Mediterranean type, are found. These

climates roughly correspond to the belt limited by the 10-20ºC annual isotherms (Spellman,

1999) or, as more recently defined, to the April-October 12-22°C isotherms (Jones, 2006).

In the next decades, a significant temperature increase is expected over Europe, particularly for

the summer periods, while precipitation is projected to diminish significantly (Meehl et al.,

2007). Further, the recent climatic trends recorded are in line with the future projections (Meehl

et al., 2007). Therefore, some wine regions, e.g. in Southern Europe, may be already near the

limit of ideal conditions for high-quality wine production (e.g. Jones et al., 2005). Conversely, a

changed environment may allow some of the northern regions (e.g. Southern England) to become

more relevant in terms of viticulture. Thus, climate change may have both positive and negative

impacts on viticulture, depending on the geographical area (e.g. Jones et al., 2005).

Taking into account the interactions between wine-grape climatic needs and its growing cycle,

some climate-based indices (“bioclimatic indices”) have been developed. One of the earliest

indices was the heat unit concept, using a growing degrees base of 10°C (Amerine and Winkler,

1944). Other indices also include radiation (e.g. Branas, 1974) or precipitation (e.g. Branas et al.,

1946), besides thermal information. In this context, the bioclimatic indices are a widely used tool

in viticultural zoning by allowing the assessment of the potential suitability of a particular region

for an economically-sustained wine production.

The main objectives of the present study are threefold: i) to determine the spatial patterns in

Europe of a set of appropriate bioclimatic indices for the recent-past period (1961-2000) and a

future periods in the XXI century considering regional climate model datasets, ii) to characterise

regions with diverse grape suitability and, iii) to identify possible future geographical variations

in these regions by analysing significant changes in the index patterns.


To analyse the effect of predicted global warming on the geography of European winegrape

growing zoning, four bioclimatic indices were selected: 1) Latitude-temperature index (LTI)

(Kenny and Shao, 1992), 2) Winkler index (WI) from April to October (Winkler et al., 1974), 3)

Branas Heliothermic index (BHI) (Branas, 1974) and, 4) Hydrothermic index of Branas, Bernon

and Levadoux (HyI) (Branas et al., 1946).

The indices are calculated using simulated daily maximum and minimum temperatures and

daily precipitation totals obtained from the regional model COSMO-CLM (Consortium for Small

Scale Modelling - Climate version of the Lokal-Modell; Böhm et al., 2006). Gridded climatic

data is available with a spatial resolution of 0.165º latitude-longitude (grid size of about 20 km).

Astronomical insolation for BHI computation is estimated by using the solar declination and the

time of the year (Allen et al., 1998). All bioclimatic indices are calculated on a daily basis. A

two-member ensemble simulation of the past climate (1960-2000) and for the XXI century

(2001-2100; A1B) following the SRES A1B scenario (Nakienovi et al., 2000) are considered.

Wine indices over the European sector (35-60ºN; 10ºW-35ºE) are compared for the recent-past

(1960-2000) versus two future periods (2011-2040 and 2041-2070).


Different bioclimatic indices were used to map macroclimate winegrape suitability across

Europe for a recent-past period and for future periods under a climate change scenario (Fig. 1, 2).

The patterns of LTI are scaled in such a way that the intervals correspond to the climatic groups

defined by Kenny and Shao (1992). A region is considered unsuitable for vine growing when LTI

3 - 90



FUTURE SCENARIOS FOR VITICULTURAL CLIMATIC ZONING

IN EUROPE

A. C. Malheiro (1)

, J. A. Santos (1)

, H. Fraga (1)

, J. G. Pinto (2)

(1) Centre for Research and Technology of Agro-Environment and Biological Sciences (CITAB), University of Trás-

os-Montes e Alto Douro, 5001-801 Vila Real, Portugal ([email protected]) (2)

Institut für Geophysik und Meteorologie, Universität zu Köln, Kerpener Str. 13, 50923 Köln, Germany

ABSTRACT

Climate is one of the main conditioning factors of winemaking. In this context, bioclimatic

indices are a useful zoning tool, allowing the description of the suitability of a particular region

for wine production. In this study, we compute climatic indices for Europe, characterize regions

with different viticultural aptitude, and assess possible variations in these regions under a future

climate conditions using a state-of-the-art regional climate model. The indices are calculated

from climatic variables (mostly daily temperatures and precipitation) obtained from the regional

climate model COSMO-CLM for recent and future climate conditions. Maps of theses indices for

recent decades (1961-2000) and for the XXI century (following the SRES A1B scenario) are

considered to identify possible changes. Results show that climate change is projected to have a

significant negative impact in wine quality by increased dryness and cumulative thermal effects

during growing seasons in Southern European regions (e.g. Portugal, Spain and Italy). These

changes represent an important constraint to grapevine growth and development, making crucial

adaptation/mitigation strategies to be adopted. On the other hand, regions of western and central

Europe (e.g. southern Britain, northern France and Germany) will benefit from this scenario both

in wine quality, and in new potential areas for viticulture. This approach provides a macro-

characterization of European areas where grapevines may preferentially grow, as well as their

projected changes, and is thus a valuable tool for viticultural zoning in a changing climate.

KEYWORD

Viticultural zoning – scenarios – Europe – climate change – CLM

INTRODUCTION

Winemaking has a predominant economic, social and environmental relevance in Europe.

Studies addressing the influence of climate variability and change in viticulture are particularly

pertinent, as climate is the dominant factor from grapevine yield and quality (e.g. van Leeuwen et

al., 2004) to its geographic worldwide distribution (Jones, 2006). In fact, Vitis vinifera has well-

defined climatic requirements; it is a heat demanding crop, needing proper high radiation

intensities and temperatures, not only during its vegetative growth and development, but also for

berry maturation and it is highly sensitive to frost occurrences in winter and spring (Magalhães,

2008). Due to these very selective requirements, most wine-producing areas are geographically

located between latitudes 30-50º of the Northern Hemisphere (Spellman, 1999), where the “warm

temperate climates” (Kottek et al., 2006), including the Mediterranean type, are found. These

climates roughly correspond to the belt limited by the 10-20ºC annual isotherms (Spellman,

1999) or, as more recently defined, to the April-October 12-22°C isotherms (Jones, 2006).

In the next decades, a significant temperature increase is expected over Europe, particularly for

the summer periods, while precipitation is projected to diminish significantly (Meehl et al.,

2007). Further, the recent climatic trends recorded are in line with the future projections (Meehl

et al., 2007). Therefore, some wine regions, e.g. in Southern Europe, may be already near the

limit of ideal conditions for high-quality wine production (e.g. Jones et al., 2005). Conversely, a

changed environment may allow some of the northern regions (e.g. Southern England) to become

more relevant in terms of viticulture. Thus, climate change may have both positive and negative

impacts on viticulture, depending on the geographical area (e.g. Jones et al., 2005).

Taking into account the interactions between wine-grape climatic needs and its growing cycle,

some climate-based indices (“bioclimatic indices”) have been developed. One of the earliest

indices was the heat unit concept, using a growing degrees base of 10°C (Amerine and Winkler,

1944). Other indices also include radiation (e.g. Branas, 1974) or precipitation (e.g. Branas et al.,

1946), besides thermal information. In this context, the bioclimatic indices are a widely used tool

in viticultural zoning by allowing the assessment of the potential suitability of a particular region

for an economically-sustained wine production.

The main objectives of the present study are threefold: i) to determine the spatial patterns in

Europe of a set of appropriate bioclimatic indices for the recent-past period (1961-2000) and a

future periods in the XXI century considering regional climate model datasets, ii) to characterise

regions with diverse grape suitability and, iii) to identify possible future geographical variations

in these regions by analysing significant changes in the index patterns.


To analyse the effect of predicted global warming on the geography of European winegrape

growing zoning, four bioclimatic indices were selected: 1) Latitude-temperature index (LTI)

(Kenny and Shao, 1992), 2) Winkler index (WI) from April to October (Winkler et al., 1974), 3)

Branas Heliothermic index (BHI) (Branas, 1974) and, 4) Hydrothermic index of Branas, Bernon

and Levadoux (HyI) (Branas et al., 1946).

The indices are calculated using simulated daily maximum and minimum temperatures and

daily precipitation totals obtained from the regional model COSMO-CLM (Consortium for Small

Scale Modelling - Climate version of the Lokal-Modell; Böhm et al., 2006). Gridded climatic

data is available with a spatial resolution of 0.165º latitude-longitude (grid size of about 20 km).

Astronomical insolation for BHI computation is estimated by using the solar declination and the

time of the year (Allen et al., 1998). All bioclimatic indices are calculated on a daily basis. A

two-member ensemble simulation of the past climate (1960-2000) and for the XXI century

(2001-2100; A1B) following the SRES A1B scenario (Nakienovi et al., 2000) are considered.

Wine indices over the European sector (35-60ºN; 10ºW-35ºE) are compared for the recent-past

(1960-2000) versus two future periods (2011-2040 and 2041-2070).


Different bioclimatic indices were used to map macroclimate winegrape suitability across

Europe for a recent-past period and for future periods under a climate change scenario (Fig. 1, 2).

The patterns of LTI are scaled in such a way that the intervals correspond to the climatic groups

defined by Kenny and Shao (1992). A region is considered unsuitable for vine growing when LTI

3 - 91



is < 380, which for current climate conditions, roughly corresponds to regions northward of the

52ºN parallel (Fig. 1). For 2041-2070, this northern limit of wine production suitability is

displaced northward to 55ºN. Similar expansions can also be found in the other groups, with

obvious exceptions over mountainous areas. Therefore, the projected climate change is expected

to have a positive thermal effect on grapevine growing and wine quality over most of Europe,

with the clearest exception of high-latitude or high-altitude regions, where thermal conditions

will remain far below the minimum climatic needs for an adequate vegetative development.

Fig. 1 On the left: Latitude-temperature index for: 1960-2000 (top panel); b) 2011-2040 (middle

panel) and c) 2041-2070 (bottom panel). On the right: the same for the Winkler index.

These results are in agreement with Kenny and Harrison (1992) and Jones et al. (2005). This

index, however, may be not suitable for evaluating seasonal differences in heat accumulation

(Jackson, 2001). Conversely, LTI is particularly functional for discerning between cool climate

areas, particularly those in Region I of Winkler classification. The classical Winkler Index, WI

(Winkler et al., 1974), which integrates heat accumulation variations, is generally in agreement

with the LTI results, though with better geographical definition, showing much higher values

over Southern Europe, particularly over south-western Iberia and parts of Italy (Fig. 1). In fact,

WI has been found to be accurate for comparing seasons, e.g. for discriminating intermediate

through hot regions (e.g. southern Europe) (Jackson, 2001).

The scale in Fig. 1 corresponds to the five climatic regions defined by Amerine and Winkler

(1944): from Region I (Cool) to Region V (Very hot). High values of WI reveal suitable areas

for grapevine with late maturation, while low values fit for early maturation varieties. With some

few exceptions, all classes but Region I (< 1372ºC) can be found over Southern Europe in present

climate conditions. For future climate conditions, a slight northward displacement of these

categories is found, e.g. for some areas in Central Europe, while the coolest region remains

largely unchanged. Most latitudes above 50ºN remain keyed as having “cool” climates, being

thus mostly unfavourable to wine production. Conversely, climate regions over Southern Europe

are expected to change dramatically to warmer classifications (cf. Fig. 1). These predicted

changes are indeed new challenges that can be a real threat for this sector if suitable adaptation

measures are not timely planned. For instance, it will be necessary to implement vineyard

irrigation in hot and dry climates for most of Southern Europe during the XXI century (Schultz,

2000). Further, some regions of Southern Europe may become excessively warm and dry for

producing high-quality wines (Kenny and Harrison, 1992; Jones et al., 2005).

Fig. 2 As in Fig. 1, but now for the Branas Heliothermic index (on the left) and for the

Hydrothermic index of Branas, Bernon and Levadoux (on the right).

The Branas Heliothermic index (BHI) has the advantage of integrating radiation (astronomical

insolation) besides temperature (Branas, 1974), both variables having a strong influence on grape

development and quality (e.g. Spellman, 1999). Values below 2.6 ºC.h are considered the lower

threshold defining the northern limit of wine production viability in Western Europe

(Carbonneau, 2003). In fact, BHI depicts values below this threshold over most of central and

northern Europe, both in recent decades and the XXI century (Fig. 2), whereas it reveals high

values over large areas of southern Europe (mainly over some parts of southern Iberia, Italy and

the Balkans). Nevertheless, the latitudinal increase in day length (higher astronomical insolation)

leads to an important weakening of the north-south contrast in the classification when compared

3 - 92



is < 380, which for current climate conditions, roughly corresponds to regions northward of the

52ºN parallel (Fig. 1). For 2041-2070, this northern limit of wine production suitability is

displaced northward to 55ºN. Similar expansions can also be found in the other groups, with

obvious exceptions over mountainous areas. Therefore, the projected climate change is expected

to have a positive thermal effect on grapevine growing and wine quality over most of Europe,

with the clearest exception of high-latitude or high-altitude regions, where thermal conditions

will remain far below the minimum climatic needs for an adequate vegetative development.

Fig. 1 On the left: Latitude-temperature index for: 1960-2000 (top panel); b) 2011-2040 (middle

panel) and c) 2041-2070 (bottom panel). On the right: the same for the Winkler index.

These results are in agreement with Kenny and Harrison (1992) and Jones et al. (2005). This

index, however, may be not suitable for evaluating seasonal differences in heat accumulation

(Jackson, 2001). Conversely, LTI is particularly functional for discerning between cool climate

areas, particularly those in Region I of Winkler classification. The classical Winkler Index, WI

(Winkler et al., 1974), which integrates heat accumulation variations, is generally in agreement

with the LTI results, though with better geographical definition, showing much higher values

over Southern Europe, particularly over south-western Iberia and parts of Italy (Fig. 1). In fact,

WI has been found to be accurate for comparing seasons, e.g. for discriminating intermediate

through hot regions (e.g. southern Europe) (Jackson, 2001).

The scale in Fig. 1 corresponds to the five climatic regions defined by Amerine and Winkler

(1944): from Region I (Cool) to Region V (Very hot). High values of WI reveal suitable areas

for grapevine with late maturation, while low values fit for early maturation varieties. With some

few exceptions, all classes but Region I (< 1372ºC) can be found over Southern Europe in present

climate conditions. For future climate conditions, a slight northward displacement of these

categories is found, e.g. for some areas in Central Europe, while the coolest region remains

largely unchanged. Most latitudes above 50ºN remain keyed as having “cool” climates, being

thus mostly unfavourable to wine production. Conversely, climate regions over Southern Europe

are expected to change dramatically to warmer classifications (cf. Fig. 1). These predicted

changes are indeed new challenges that can be a real threat for this sector if suitable adaptation

measures are not timely planned. For instance, it will be necessary to implement vineyard

irrigation in hot and dry climates for most of Southern Europe during the XXI century (Schultz,

2000). Further, some regions of Southern Europe may become excessively warm and dry for

producing high-quality wines (Kenny and Harrison, 1992; Jones et al., 2005).

Fig. 2 As in Fig. 1, but now for the Branas Heliothermic index (on the left) and for the

Hydrothermic index of Branas, Bernon and Levadoux (on the right).

The Branas Heliothermic index (BHI) has the advantage of integrating radiation (astronomical

insolation) besides temperature (Branas, 1974), both variables having a strong influence on grape

development and quality (e.g. Spellman, 1999). Values below 2.6 ºC.h are considered the lower

threshold defining the northern limit of wine production viability in Western Europe

(Carbonneau, 2003). In fact, BHI depicts values below this threshold over most of central and

northern Europe, both in recent decades and the XXI century (Fig. 2), whereas it reveals high

values over large areas of southern Europe (mainly over some parts of southern Iberia, Italy and

the Balkans). Nevertheless, the latitudinal increase in day length (higher astronomical insolation)

leads to an important weakening of the north-south contrast in the classification when compared

3 - 93



with WI. In fact, at mid-latitudes, this index reveals grapevine growing suitability over several

areas that were not identified by the WI, including northern and central parts of France and some

parts of south-western Germany that are renowned wine producing areas. For future climate

conditions, a significant northward extension of the regions with wine-producing potential is

projected, being particularly meaningful over large areas of southern England, Belgium, the

Netherlands, Germany, Czech Republic and southern Poland. These results are especially

significant when taking into account the efficiency of the BHI in viticultural zoning.

Changes in precipitation might also have an important impact on viticultural zoning: for

example, excessive dryness usually compels irrigation, while excessive precipitation, air

humidity and soil moisture can expose grapevines to diseases and pests. Therefore, bioclimatic

indices were developed that combine temperature and precipitation, e.g. for estimations of risk

diseases such as downy mildew (Carbonneau, 2003). We consider the Hydrothermic index of

Branas, Bernon and Levadoux, HyI (Branas et al., 1946). The risk of contamination by downy

mildew is considered low when HyI has values below 2500ºC.mm and high when is higher than

5100ºC.mm. The patterns of HyI show that all areas with a Mediterranean-type climate present

low risks of contamination, both for recent and future conditions, whilst high- and mid-latitude

Europe present moderate to high risks, with an increasing trend over Central and Eastern Europe.

This situation is particularly relevant for mid-latitude areas, where thermal conditions tend to be

gradually favourable to wine production. Here, the climate warming is not accompanied by a

beneficial decrease in precipitation, which increases the risk of grapevine attack by this disease.

This can be a limitation to wine production, in spite of the general improvement in the thermal

conditions. The projected enhancement of dryness in Southern Europe further decreases this risk,

but might force the development of expansive and environmentally unsustainable irrigation

techniques in several regions where they are not currently implemented.

CONCLUSIONS

Macroclimate maps show that under changed climate significant shifts and/or expansions in the

European viticultural zoning can be expected. Negative impacts on wine quality by increased

cumulative thermal and dryness effects during growing seasons in Southern European regions

(e.g. Portugal, Spain and Italy) may happen. This is an important constraint to grapevine growth

and development, making the development of adequate adaptation and/or mitigation measures of

vital importance for this key economical sector. On the other hand, regions of western and central

Europe (e.g. southern Britain, northern France and Germany) will benefit from changed

conditions, both in terms of wine quality and in emerging new areas for viticulture. This

approach also provides a macro-characterization of the European areas, where grapevines may

grow in an economically-sustained manner, as well as their projected changes under

anthropogenic forcing. Thus, these indices are a valuable tool for viticultural zoning in a

changing global climate. Further bioclimatic indices will be analysed to improve this study.

ACKNOWLEDGMENTS

Part of this study was supported by the project SUVIDUR – Sustentabilidade da Viticultura de

Encosta nas Regiões do Douro e do Duero. Programa Operacional de Cooperação

Transfronteiriça Espanha-Portugal (POCTEP). We thank the German Federal Environment

Agency, the WDCC/CERA database and the COSMO-CLM consortium for the CLM data.

BIBLIOGRAPHY

Allen R.G., Pereira L.S., Raes D., Smith M., 1998. Crop evapotranspiration: guidelines for

computing crop water requirements. FAO irrigation and drainage paper 56. FAO, Rome.

Amerine M.A., Winkler A.J., 1944. Composition and quality of musts and wines of California

grapes. Hilgardia 15: 493-675.

Böhm U., Kücken M., Ahrens W., Block A., Hauffe D., Keuler K., Rockel B., Will A., 2006.

CLM- The Climate Version of LM: Brief Description and Long- Term Applications. COSMO

Newsletter 6: 225–235.

Branas J., 1974. Viticulture. Imp. Dehan, Montpellier.

Branas J., Bernon G., Levadoux L., 1946. Eléments de viticulture générale. Imp. Dehan,

Montpellier.

Carbonneau A., 2003. Ecophysiologie de la vigne et terroir. In: Terroir, zonazione, viticoltura.

Trattato internazionale, M. Fregoni, D. Schuster, A. Paoletti (eds.). Ed. Phytoline, 61-102.

Jackson D., 2001. Climate, monographs in cool climate viticulture – 2. Daphne Brasell

Associates Ltd.

Jones G.V., White M.A., Cooper O.R., Storchmann K., 2005. Climate change and global wine

quality. Clim. Change 73: 319-343.

Jones G.V., 2006. Climate and terroir: Impacts of climate variability and change on wine. In Fine

Wine and Terroir - The Geoscience Perspective. Macqueen, R.W., and Meinert, L.D., (eds.),

Geoscience Canada Reprint Series Number 9, Geological Association of Canada, St. John’s,

Newfoundland.

Kenny G.J., Harrison P.A., 1992. The effects of climate variability and change on grape

suitability in Europe. J. Wine Res. 3: 163-183.

Kenny G.J., Shao J., 1992. An assessment of a latitude-temperature index for predicting climate

suitability for grapes in Europe. J. Hortic. Sci. 67(2): 239-246.

Kottek M., Grieser J., Beck C., Rudolf B., Rubel F., 2006. World Map of the Köppen-Geiger

climate classification updated. Meteorologische Zeitschrift 15(3): 259-263.

Magalhães N.P., 2008. Tratado de Viticultura - A videira, a vinha e o “terroir”. Chaves Ferreira

Publicações, Lisboa.

Meehl G.A. et al., 2007. Global Climate Projections. In: Climate Change 2007: The Physical

Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the

Intergovernmental Panel on Climate Change (eds. Solomon, S., Qin, D., Manning, M., Chen,

Z., Marquis, M., Averyt, K. B., Tignor, M. and Miller, H. L.). Cambridge University Press,

Cambridge, United Kingdom and New York, NY, USA.

Nakienovi N. et al., 2000. IPCC Special Report on Emissions Scenarios. Cambridge University

Press, Cambridge, United Kingdom and New York, NY, USA.

Schultz H.R., 2000. Climate change and viticulture: a European perspective on climatology,

carbon dioxide and UV-B effects. Aust. J. Grape Wine Res. 6: 2-12.

Spellman G., 1999. Wine, weather and climate. Weather 54: 230-239.

van Leeuwen C., Friant P., Choné X., Tregoat O., Koundouras S., Dubourdieu D., 2004.

Influence of climate, soil, and cultivar on terroir. Am. J. Enol. Viticult. 55: 207-217.

Winkler A.J., Cook J.A., Kliwer W.M., Lider L.A., 1974. General viticulture. University of

California Press.

3 - 94



with WI. In fact, at mid-latitudes, this index reveals grapevine growing suitability over several

areas that were not identified by the WI, including northern and central parts of France and some

parts of south-western Germany that are renowned wine producing areas. For future climate

conditions, a significant northward extension of the regions with wine-producing potential is

projected, being particularly meaningful over large areas of southern England, Belgium, the

Netherlands, Germany, Czech Republic and southern Poland. These results are especially

significant when taking into account the efficiency of the BHI in viticultural zoning.

Changes in precipitation might also have an important impact on viticultural zoning: for

example, excessive dryness usually compels irrigation, while excessive precipitation, air

humidity and soil moisture can expose grapevines to diseases and pests. Therefore, bioclimatic

indices were developed that combine temperature and precipitation, e.g. for estimations of risk

diseases such as downy mildew (Carbonneau, 2003). We consider the Hydrothermic index of

Branas, Bernon and Levadoux, HyI (Branas et al., 1946). The risk of contamination by downy

mildew is considered low when HyI has values below 2500ºC.mm and high when is higher than

5100ºC.mm. The patterns of HyI show that all areas with a Mediterranean-type climate present

low risks of contamination, both for recent and future conditions, whilst high- and mid-latitude

Europe present moderate to high risks, with an increasing trend over Central and Eastern Europe.

This situation is particularly relevant for mid-latitude areas, where thermal conditions tend to be

gradually favourable to wine production. Here, the climate warming is not accompanied by a

beneficial decrease in precipitation, which increases the risk of grapevine attack by this disease.

This can be a limitation to wine production, in spite of the general improvement in the thermal

conditions. The projected enhancement of dryness in Southern Europe further decreases this risk,

but might force the development of expansive and environmentally unsustainable irrigation

techniques in several regions where they are not currently implemented.

CONCLUSIONS

Macroclimate maps show that under changed climate significant shifts and/or expansions in the

European viticultural zoning can be expected. Negative impacts on wine quality by increased

cumulative thermal and dryness effects during growing seasons in Southern European regions

(e.g. Portugal, Spain and Italy) may happen. This is an important constraint to grapevine growth

and development, making the development of adequate adaptation and/or mitigation measures of

vital importance for this key economical sector. On the other hand, regions of western and central

Europe (e.g. southern Britain, northern France and Germany) will benefit from changed

conditions, both in terms of wine quality and in emerging new areas for viticulture. This

approach also provides a macro-characterization of the European areas, where grapevines may

grow in an economically-sustained manner, as well as their projected changes under

anthropogenic forcing. Thus, these indices are a valuable tool for viticultural zoning in a

changing global climate. Further bioclimatic indices will be analysed to improve this study.

ACKNOWLEDGMENTS

Part of this study was supported by the project SUVIDUR – Sustentabilidade da Viticultura de

Encosta nas Regiões do Douro e do Duero. Programa Operacional de Cooperação

Transfronteiriça Espanha-Portugal (POCTEP). We thank the German Federal Environment

Agency, the WDCC/CERA database and the COSMO-CLM consortium for the CLM data.

BIBLIOGRAPHY

Allen R.G., Pereira L.S., Raes D., Smith M., 1998. Crop evapotranspiration: guidelines for

computing crop water requirements. FAO irrigation and drainage paper 56. FAO, Rome.

Amerine M.A., Winkler A.J., 1944. Composition and quality of musts and wines of California

grapes. Hilgardia 15: 493-675.

Böhm U., Kücken M., Ahrens W., Block A., Hauffe D., Keuler K., Rockel B., Will A., 2006.

CLM- The Climate Version of LM: Brief Description and Long- Term Applications. COSMO

Newsletter 6: 225–235.

Branas J., 1974. Viticulture. Imp. Dehan, Montpellier.

Branas J., Bernon G., Levadoux L., 1946. Eléments de viticulture générale. Imp. Dehan,

Montpellier.

Carbonneau A., 2003. Ecophysiologie de la vigne et terroir. In: Terroir, zonazione, viticoltura.

Trattato internazionale, M. Fregoni, D. Schuster, A. Paoletti (eds.). Ed. Phytoline, 61-102.

Jackson D., 2001. Climate, monographs in cool climate viticulture – 2. Daphne Brasell

Associates Ltd.

Jones G.V., White M.A., Cooper O.R., Storchmann K., 2005. Climate change and global wine

quality. Clim. Change 73: 319-343.

Jones G.V., 2006. Climate and terroir: Impacts of climate variability and change on wine. In Fine

Wine and Terroir - The Geoscience Perspective. Macqueen, R.W., and Meinert, L.D., (eds.),

Geoscience Canada Reprint Series Number 9, Geological Association of Canada, St. John’s,

Newfoundland.

Kenny G.J., Harrison P.A., 1992. The effects of climate variability and change on grape

suitability in Europe. J. Wine Res. 3: 163-183.

Kenny G.J., Shao J., 1992. An assessment of a latitude-temperature index for predicting climate

suitability for grapes in Europe. J. Hortic. Sci. 67(2): 239-246.

Kottek M., Grieser J., Beck C., Rudolf B., Rubel F., 2006. World Map of the Köppen-Geiger

climate classification updated. Meteorologische Zeitschrift 15(3): 259-263.

Magalhães N.P., 2008. Tratado de Viticultura - A videira, a vinha e o “terroir”. Chaves Ferreira

Publicações, Lisboa.

Meehl G.A. et al., 2007. Global Climate Projections. In: Climate Change 2007: The Physical

Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the

Intergovernmental Panel on Climate Change (eds. Solomon, S., Qin, D., Manning, M., Chen,

Z., Marquis, M., Averyt, K. B., Tignor, M. and Miller, H. L.). Cambridge University Press,

Cambridge, United Kingdom and New York, NY, USA.

Nakienovi N. et al., 2000. IPCC Special Report on Emissions Scenarios. Cambridge University

Press, Cambridge, United Kingdom and New York, NY, USA.

Schultz H.R., 2000. Climate change and viticulture: a European perspective on climatology,

carbon dioxide and UV-B effects. Aust. J. Grape Wine Res. 6: 2-12.

Spellman G., 1999. Wine, weather and climate. Weather 54: 230-239.

van Leeuwen C., Friant P., Choné X., Tregoat O., Koundouras S., Dubourdieu D., 2004.

Influence of climate, soil, and cultivar on terroir. Am. J. Enol. Viticult. 55: 207-217.

Winkler A.J., Cook J.A., Kliwer W.M., Lider L.A., 1974. General viticulture. University of

California Press.

3 - 95



1

THE ADAPTATIVE CAPACITY OF A VITICULTURAL AREA (VALLE TELESINA, SOUTHERN ITALY) TO CLIMATE CHANGES

A. Bonfante(1), A. Basile (1), F. De Lorenzi (1), G. Langella (1), F. Terribile (2), M. Menenti (3)

(1) Institute for Mediterranean Agricultural and Forest Systems (ISAFOM-CNR), Ercolano (NA), Italy ([email protected])

(2)University of Naples Federico II, Portici (NA). Italy (3) Delft University of Technology, Delft, The Netherlands

ABSTRACT The viticulture aiming at the production of high quality wine is very important for the landscape

conservation, because it allows to combine high farmer income with soil conservation. The quality of grape and wine is variety-specific and it depends significantly on the pedoclimatic conditions. The evolution of climate may thus endanger not only yield (IPCC, 2007) but, more significantly, the sustainability of current varieties. Adaptation of current production systems may be feasible, but requires a timely evaluation of whether adaptation to climate evolution might be limited to improving crop and soil management or should involve replacement of cvs or species altogether.

This study addressed this question by evaluating the adaptive capacity of a 20000 ha viticultural area in the “Valle Telesina” (Campania Region, Southern Italy). This area has a long tradition in the production of high quality wines (DOC and DOCG) and it is characterized by a complex geomorphology with a large soil and climate variability.

Two climate periods were considered: “past” (1984-1996) and “present” (2000-2009), which show a pattern of climate variability. The periods were taken as an example of different scenarios generated by climate changes.

The Amerine & Winkler index was calculated in each climate period and compared with the thermal requirements of a set of grapevine cvs, including the ones currently cultivated in the area.

Due to the observed trend of temperature increase from the “past” to the “present” period, differences were detected in the A&W index’s values and spatial distribution. When compared with the A&W indexes of the grape varieties the temperature increase resulted in a considerable increase of the area eligible to some varieties (Guarnaccia and Forastera) and a strong reduction of the area suitable for some of the most important current varieties (Aglianico and Falanghina).

Moreover, the hydrological model SWAP was applied to estimate the Crop Water Stress Index (CWSI) in the “present” climatic period, in order to evaluate the effects of the re-distribution of the cultivars over the study area on vineyards’ water balance.

This approach is being applied to other crops and other production systems towards quantitative, realistic studies on the adaptation of agriculture to climate evolution.

KEYWORDS Grapevine adaptative capacity - Amerine & Winkler index – SWAP - Climate changes - quality

viticulture

2

INTRODUCTION Understanding the effects of climate change in the areas traditionally vocated to vineyard

production allows to assess the future viability of current cultivation. There are many pedo-climatic factors that affect vineyard production, as described by Carey,

2001 by means of the “terroir” concept, i.e. a complex of natural environmental factors, which cannot easily be modified by the producer.

The large influence of climate on vine growth and in determining the character of the wine was discussed in detail by Saayman, 1977 and Carey, 2001, among others.

The effect of soil water availability on vine functioning and on grape and wine quality is of established importance (e.g. Matthews and Anderson, 1988; Medrano et al., 2003).

The assessment of the influence of climate change on vineyard production can rely upon the availability and the combination of research tools like raster GIS (Bonfante et al., 2005; Scaglione et al., 2008; Acevedo-Opazo et al., 2008), DEM derived analysis (MacMillan et al., 2000), hydrological simulation models (Ben-Asher et al., 2006), global and regional climate simulation models (Jones et al., 2005).

The main aim of the present work is to evaluate, in a viticultural area of Southern Italy, the effects of a climate variation on the thermal regime and on the distribution of the cultivation area of different grape cultivars.

MATERIALS AND METHODS The study area: environmental dataset The work has been performed in “Valle Telesina”, a 20.000 ha complex landscape located in

the Campania region, Southern Italy. The area has a high soil and climate spatial variability; it is a traditional setting for vineyards producing high quality wines, including three DOC wines (Guardiolo, Solopaca and Sannio).

The landscape has a complex geomorphology and it is characterized by an E-W elongated graben where the Calore river flows. The area shows five different geomorphic environments (Fig.1), namely (i) limestone mountains having volcanic ash deposits at the surface; (ii) hills constituted by marl arenaceous flysch; (iii) pediment plain constituted by colluvial material of the slope fan of the limestone relieves; (iv) ancient alluvial terraces and (v) actual alluvial plain.

Soil information were derived from a soil map at 1:50,000 scale (Terribile et al., 1996) consisting of 47 soil mapping units and about 60 soil typological units. In this work 32 of 47 soil units were used, covering almost completely the area.

The digital elevation model (DEM) was obtained from the digitalization of topographic maps, produced by the Istituto Geografico Militare Italiano at 1:25,000 scale, producing a DEM having a 20×20 m resolution.

The climatic data were obtained from Campania Region agrometeorological office and refer to two climatic periods: Fig.1 - The main geomorphic environments of “Valle Telesina”.

3 - 96



1

THE ADAPTATIVE CAPACITY OF A VITICULTURAL AREA (VALLE TELESINA, SOUTHERN ITALY) TO CLIMATE CHANGES

A. Bonfante(1), A. Basile (1), F. De Lorenzi (1), G. Langella (1), F. Terribile (2), M. Menenti (3)

(1) Institute for Mediterranean Agricultural and Forest Systems (ISAFOM-CNR), Ercolano (NA), Italy ([email protected])

(2)University of Naples Federico II, Portici (NA). Italy (3) Delft University of Technology, Delft, The Netherlands

ABSTRACT The viticulture aiming at the production of high quality wine is very important for the landscape

conservation, because it allows to combine high farmer income with soil conservation. The quality of grape and wine is variety-specific and it depends significantly on the pedoclimatic conditions. The evolution of climate may thus endanger not only yield (IPCC, 2007) but, more significantly, the sustainability of current varieties. Adaptation of current production systems may be feasible, but requires a timely evaluation of whether adaptation to climate evolution might be limited to improving crop and soil management or should involve replacement of cvs or species altogether.

This study addressed this question by evaluating the adaptive capacity of a 20000 ha viticultural area in the “Valle Telesina” (Campania Region, Southern Italy). This area has a long tradition in the production of high quality wines (DOC and DOCG) and it is characterized by a complex geomorphology with a large soil and climate variability.

Two climate periods were considered: “past” (1984-1996) and “present” (2000-2009), which show a pattern of climate variability. The periods were taken as an example of different scenarios generated by climate changes.

The Amerine & Winkler index was calculated in each climate period and compared with the thermal requirements of a set of grapevine cvs, including the ones currently cultivated in the area.

Due to the observed trend of temperature increase from the “past” to the “present” period, differences were detected in the A&W index’s values and spatial distribution. When compared with the A&W indexes of the grape varieties the temperature increase resulted in a considerable increase of the area eligible to some varieties (Guarnaccia and Forastera) and a strong reduction of the area suitable for some of the most important current varieties (Aglianico and Falanghina).

Moreover, the hydrological model SWAP was applied to estimate the Crop Water Stress Index (CWSI) in the “present” climatic period, in order to evaluate the effects of the re-distribution of the cultivars over the study area on vineyards’ water balance.

This approach is being applied to other crops and other production systems towards quantitative, realistic studies on the adaptation of agriculture to climate evolution.

KEYWORDS Grapevine adaptative capacity - Amerine & Winkler index – SWAP - Climate changes - quality

viticulture

2

INTRODUCTION Understanding the effects of climate change in the areas traditionally vocated to vineyard

production allows to assess the future viability of current cultivation. There are many pedo-climatic factors that affect vineyard production, as described by Carey,

2001 by means of the “terroir” concept, i.e. a complex of natural environmental factors, which cannot easily be modified by the producer.

The large influence of climate on vine growth and in determining the character of the wine was discussed in detail by Saayman, 1977 and Carey, 2001, among others.

The effect of soil water availability on vine functioning and on grape and wine quality is of established importance (e.g. Matthews and Anderson, 1988; Medrano et al., 2003).

The assessment of the influence of climate change on vineyard production can rely upon the availability and the combination of research tools like raster GIS (Bonfante et al., 2005; Scaglione et al., 2008; Acevedo-Opazo et al., 2008), DEM derived analysis (MacMillan et al., 2000), hydrological simulation models (Ben-Asher et al., 2006), global and regional climate simulation models (Jones et al., 2005).

The main aim of the present work is to evaluate, in a viticultural area of Southern Italy, the effects of a climate variation on the thermal regime and on the distribution of the cultivation area of different grape cultivars.

MATERIALS AND METHODS The study area: environmental dataset The work has been performed in “Valle Telesina”, a 20.000 ha complex landscape located in

the Campania region, Southern Italy. The area has a high soil and climate spatial variability; it is a traditional setting for vineyards producing high quality wines, including three DOC wines (Guardiolo, Solopaca and Sannio).

The landscape has a complex geomorphology and it is characterized by an E-W elongated graben where the Calore river flows. The area shows five different geomorphic environments (Fig.1), namely (i) limestone mountains having volcanic ash deposits at the surface; (ii) hills constituted by marl arenaceous flysch; (iii) pediment plain constituted by colluvial material of the slope fan of the limestone relieves; (iv) ancient alluvial terraces and (v) actual alluvial plain.

Soil information were derived from a soil map at 1:50,000 scale (Terribile et al., 1996) consisting of 47 soil mapping units and about 60 soil typological units. In this work 32 of 47 soil units were used, covering almost completely the area.

The digital elevation model (DEM) was obtained from the digitalization of topographic maps, produced by the Istituto Geografico Militare Italiano at 1:25,000 scale, producing a DEM having a 20×20 m resolution.

The climatic data were obtained from Campania Region agrometeorological office and refer to two climatic periods: Fig.1 - The main geomorphic environments of “Valle Telesina”.

3 - 97



3

1984-1996 and 2000-2009 (“past” and “present” period, respectively). The spatial distribution of mean air temperature was derived as reported by Bonfante at al.

(2005). Data of 12 weather stations located at different altitudes, inside and in the near surroundings of the study area, were used. The spatial distribution of mean daily air temperature was obtained by means of a simple regression function among elevation and temperature.

Daily reference evapotranspiration (ETo) was evaluated according to the equation of Hargreaves (Hargreaves, Samani, 1985). Thermal index of Amerine & Winkler (A&W)

The thermal index of Amerine & Winkler, 1944, is the sum of average daily temperature, accounting for the thermal level at which no growth occurs, calculated in the period between 1 April and 31 October:

∑ −=10/31

04/1

)10()( mTDDI [1]

where I is the A&W thermal index value expressed in degree day (DD), Tm is the average daily temperature and 10 is a constant representing the zero vegetative of grapevine.

Through the daily temperature data spatially distributed, the A&W thermal index was calculated for the whole study area. In order to calculate the A&W index in each point of the “Valle Telesina” 214 matrices of daily temperature (one matrix per day, from 1 April to 31 October) were built for each climatic period. The differences between the index’s values in 2000-2009 versus 1984-1996 period were computed.

The thermal needs of grape varieties of the study area were taken from Scaglione et al., 2008.

Hydrological modelling Soil water balance simulation was performed using the Soil–Water–Atmosphere–Plant (SWAP)

model (Kroes, van Dam, 2003). This application enables to estimate the components of the water balance at daily time dynamics. Assuming 1-D vertical flow processes, the model calculates the water flow in the soil matrix through the Richards’ equation , taking into account root water extraction by an additive term.

Soil water retention and hydraulic conductivity curves were described by the van Genuchten-Mualem expression (van Genuchten, 1980) and their parameters were derived applying the pedo transfer function HYPRES (Wösten et al., 1998) whose reliability was tested and validated on three soils of the same area. Upper boundary condition data were derived from weather dataset and unit gradient in hydraulic head was set as lower boundary condition.

SWAP model runs were applied for to calculate crop actual and potential transpiration in the 32 soil units.

Crop Water Stress Index (CWSI) The CWSI is often defined as:

( )[ ] 1001 ⋅−= potact TTCWSI [2]

where Tact/Tpot is the ratio between actual and potential transpiration.

4

In the present work Tact and Tpot were calculated daily by the SWAP model applying crop-specific input data (Allen et al., 1998; Kroes, Van Dam, 2003).

The vineyard CWSI was estimated in the climatic period 2000-2009; the cumulated Crop Water Stress Index (CWSIcum) was calculated summing CWSI (eq. 2) in the period from shooting (1 April) to harvest (15 September). The results were averaged over the ten year period, a map of the CWSIcum was thus derived. RESULTS AND DISCUSSION Thermal index of Amerine & Winkler The difference in mean daily temperature between the “present” and the “past” climatic period was 1.4 °C. The values of the A&W indexes of the periods are shown in Tab. 1.

The maximum and minimum differences of the index’s values in the two periods resulted 270 and 110 DD, respectively. Fig. 2 shows the spatially

distributed differences of A&W index between the two climatic periods. The differences in the index values were higher than 220 DD in a preponderant part of the study area; in the limestone mountain zone the differences were smaller.

The A&W index of the two climatic periods, spatially distributed in the study area, was compared with the heat requirements of local grapevine varieties (Scaglione et al., 2008) and results are reported in Fig. 3. In the “present” period the thermal regime allowed the increase of the land surface suitable for Forastera, a variety with rather high thermal requirements (1950 DD) and Guarnaccia, a variety whose thermal needs were not met in the “past”. The latter, in the warmer “present” period, would have the largest suitable cultivation area (21 % of the entire ”Valle Telesina”). Conversely, the difference in thermal regime between “present” and “past” determined a strong reduction of the area eligible to Aglianico and Falanghina, two of the most important current black (b) and white (w) varieties; moreover it can be supposed that Aglianico and Falanghina would move towards areas at higher altitude.

Tab. 1 Amerine and Winkler indexes (DD) climatic period max min avg

1984-1996 (“past”) 1870 737 1657 2000-2009 (“present”) 2140 849 1866

Fig. 2 – Map of the Amerine and Winkler index differences

between the “present” (2000-2009) and the “past” (1984-1996) climatic periods.

Fig. 3 – Local grape cvs suitable surface (%) in the “past” and “present” climatic periods.

3 - 98



3

1984-1996 and 2000-2009 (“past” and “present” period, respectively). The spatial distribution of mean air temperature was derived as reported by Bonfante at al.

(2005). Data of 12 weather stations located at different altitudes, inside and in the near surroundings of the study area, were used. The spatial distribution of mean daily air temperature was obtained by means of a simple regression function among elevation and temperature.

Daily reference evapotranspiration (ETo) was evaluated according to the equation of Hargreaves (Hargreaves, Samani, 1985). Thermal index of Amerine & Winkler (A&W)

The thermal index of Amerine & Winkler, 1944, is the sum of average daily temperature, accounting for the thermal level at which no growth occurs, calculated in the period between 1 April and 31 October:

∑ −=10/31

04/1

)10()( mTDDI [1]

where I is the A&W thermal index value expressed in degree day (DD), Tm is the average daily temperature and 10 is a constant representing the zero vegetative of grapevine.

Through the daily temperature data spatially distributed, the A&W thermal index was calculated for the whole study area. In order to calculate the A&W index in each point of the “Valle Telesina” 214 matrices of daily temperature (one matrix per day, from 1 April to 31 October) were built for each climatic period. The differences between the index’s values in 2000-2009 versus 1984-1996 period were computed.

The thermal needs of grape varieties of the study area were taken from Scaglione et al., 2008.

Hydrological modelling Soil water balance simulation was performed using the Soil–Water–Atmosphere–Plant (SWAP)

model (Kroes, van Dam, 2003). This application enables to estimate the components of the water balance at daily time dynamics. Assuming 1-D vertical flow processes, the model calculates the water flow in the soil matrix through the Richards’ equation , taking into account root water extraction by an additive term.

Soil water retention and hydraulic conductivity curves were described by the van Genuchten-Mualem expression (van Genuchten, 1980) and their parameters were derived applying the pedo transfer function HYPRES (Wösten et al., 1998) whose reliability was tested and validated on three soils of the same area. Upper boundary condition data were derived from weather dataset and unit gradient in hydraulic head was set as lower boundary condition.

SWAP model runs were applied for to calculate crop actual and potential transpiration in the 32 soil units.

Crop Water Stress Index (CWSI) The CWSI is often defined as:

( )[ ] 1001 ⋅−= potact TTCWSI [2]

where Tact/Tpot is the ratio between actual and potential transpiration.

4

In the present work Tact and Tpot were calculated daily by the SWAP model applying crop-specific input data (Allen et al., 1998; Kroes, Van Dam, 2003).

The vineyard CWSI was estimated in the climatic period 2000-2009; the cumulated Crop Water Stress Index (CWSIcum) was calculated summing CWSI (eq. 2) in the period from shooting (1 April) to harvest (15 September). The results were averaged over the ten year period, a map of the CWSIcum was thus derived. RESULTS AND DISCUSSION Thermal index of Amerine & Winkler The difference in mean daily temperature between the “present” and the “past” climatic period was 1.4 °C. The values of the A&W indexes of the periods are shown in Tab. 1.

The maximum and minimum differences of the index’s values in the two periods resulted 270 and 110 DD, respectively. Fig. 2 shows the spatially

distributed differences of A&W index between the two climatic periods. The differences in the index values were higher than 220 DD in a preponderant part of the study area; in the limestone mountain zone the differences were smaller.

The A&W index of the two climatic periods, spatially distributed in the study area, was compared with the heat requirements of local grapevine varieties (Scaglione et al., 2008) and results are reported in Fig. 3. In the “present” period the thermal regime allowed the increase of the land surface suitable for Forastera, a variety with rather high thermal requirements (1950 DD) and Guarnaccia, a variety whose thermal needs were not met in the “past”. The latter, in the warmer “present” period, would have the largest suitable cultivation area (21 % of the entire ”Valle Telesina”). Conversely, the difference in thermal regime between “present” and “past” determined a strong reduction of the area eligible to Aglianico and Falanghina, two of the most important current black (b) and white (w) varieties; moreover it can be supposed that Aglianico and Falanghina would move towards areas at higher altitude.

Tab. 1 Amerine and Winkler indexes (DD) climatic period max min avg

1984-1996 (“past”) 1870 737 1657 2000-2009 (“present”) 2140 849 1866

Fig. 2 – Map of the Amerine and Winkler index differences

between the “present” (2000-2009) and the “past” (1984-1996) climatic periods.

Fig. 3 – Local grape cvs suitable surface (%) in the “past” and “present” climatic periods.

3 - 99



5

If the temperature trend from the “past” to the “present” period will be confirmed as a long-term trend in the area, the adaptability of grapevine production will have to rely on a larger use of

different, and possibly new, grape cvs. Crop Water Stress Index (CWSI) Further analysis was conducted by

means of the SWAP hydrological simulation model to estimate functional properties relevant to vineyard production. Specifically, the CWSIcum values were calculated and spatially distributed over the study area (Fig. 4). The minimum CWSIcum value (10.5 %) was found in the limestone mountains (see Fig. 1) and the maximum value (27.5 %) occurred in the pediment plain. The maximum CWSIcum corresponds to a Tact/Tpot ratio of 0.73. Trambouze and Woltz, 2001 measured a similar value of the ratio when 2/3 of soil water storage

was depleted. The maximum CWSIcum value therefore indicated a significant plant water shortage. Moreover, the results show a quite high variability of CWSIcum in the whole study area.

CONCLUSIONS The effects produced by climate change on potential vineyard distribution over an inland area

of Southern Italy (“Valle Telesina”) were analyzed. The different climatic pattern between two periods (“past” 1984-1996 and “present” 2000-2009) was assumed to be an example of future trends generated by climate changes.

The different thermal regime between the two periods was translated into differences in A&W indexes’ values; moreover the distribution of the Crop Water Stress Index was assessed for vineyards in the “present” climate.

The match between thermal regime and requirements of local cultivars allowed to assess a redistribution: traditional and widely distributed Falanghina and Aglianico would have a smaller area with favourable environment; whereas the extension of the area of cvs Forastera and Guarnaccia, the latter not cultivated in the “past” climate, would increase.

An interplay between the re-distribution of cultivars and the pattern of CWSIcum over the study area would cause different consequences on vines’ physiological processes, and therefore, on quantity and quality of production. The analysis could be further detailed to examine the effects of thermal and water regimes occurring at different phenological phases.

ACKNOWLEDGEMENTS Part of the present work was carried out within the project “Scenari di adattamento

dell’agricoltura italiana ai cambiamenti climatici – AGROSCENARI”.

Fig. 4 – Map of the CWSIcum values, cumulated from shooting to

harvesting, and averaged over the “present” climatic period.

6

REFERENCES Acevedo-Opazo C., Tisseyre B., Guillaume S., Ojeda, H., 2008. The potential of high spatial

resolution information to define within-vineyard zones related to vine water status. Precision Agric., 9:285-302.

Allen R., Pereira L.S., Raes D., Smith M. 1998. Crop evapotranspiration – Guidelines for computing crop water requirements. FAO Irrigation and Drainage Paper, vol. 56, Rome, Italy.

Amerine M.A., Winkler A.J., 1944. Composition and quality of must and wines of California grapes. Hilgardia, 15: 493-675.

Bonfante A., Basile A., Buonanno M., Manna P., Terribile F., 2005. Un esempio metodologico di utilizzo della modellistica idrologica e delle procedure GIS nella zonazione viticola. Bollettino della Società Italiana della Scienza del Suolo (SISS), 54: 1-2.

Ben-Asher J., van Dam J., Feddes R.A., Jhorar R.K., 2006. Irrigation of grapevines with saline water II. Mathematical simulation of vine growth and yield. Agric. Wat. Manag., 83: 22-29.

Carey V., 2001. Spatial characterization of natural terroir units for viticolture in the Bottelaryberg-Simonsberg-Helderberg wine growing area. M.Sc. Thesis, Stellenbosch University, Private Bag X1, 7602 Matieland (Stellenbosch), South Africa.

Kroes J.G., Van Dam J.C., 2003. Reference manual SWAP version 3.0.3. Alterra, Green World Research, Wageningen (The Netherlands).

Hargreaves G.L., Samani Z.A., 1985. Reference crop evapotranspiration from temperature. Applied Engineer. In Agric., 1, 2: 96-99.

IPCC, 2007. Climate Change 2007: Climate Change Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. In: Parry M.L., Canziani O.F., Palutikof J.P., van der Linden P.J., Hanson C.E. (Eds.). Cambridge University Press, Cambridge, UK.

Jones G.V., White M.A., Cooper O.R., Storchmann K., 2005. Climate change and global wine quality. Climatic Change, 73: 319–343.

MacMillan R.A., Pettapiece W.W., Nolan S. C., Goddard T.W., 2000. Generic procedure for automatically segmenting landforms into landform elements using DEMS, heuristic rules and fuzzy logic. Fuzzy Sets and Systems, 113:81-109.

Matthews M., Anderson M., 1988. Fruit ripening in Vitis vinifera L.: responses to seasonal water deficits. American Journal of Enology and Viticulture, 39: 313–320.

Medrano H., Escalona J.M., Cifre J., Bota J., Flexas J., 2003. A ten-year study on the physiology of two Spanish grapevine cultivars under field conditions: effect of water availability from leaf photosynthesis to grape yield and quality. Functional Plant Biology, 30:607-619.

Saayman D., 1977. The Effect of Soil and Climate on Wine Quality. In: Proc. International Symposium on the Quality of the Vintage, Cape Town, South Africa. 197–208.

Scaglione G., Pasquarella C., Federico R., Bonfante A., Terribile F., 2008. A multidisciplinary approach to grapevine zoning using GIS technology: An example of thermal data elaboration. Vitis, 47 (2):131-132.

Terribile F., Di Gennaro A., De Mascellis R., 1996. Carta dei suoli della Valle Telesina. Progetto U.O.T. Relazione finale convenzione CNR-ISPAIM-Regione Campania Assessorato alla Agricoltura.

Trambouze W., Voltz M. 2001. Measurement and modelling of the transpiration of a Mediterranean vineyard. Agric. For. Metorol, 107:153-166.

3 - 100



5

If the temperature trend from the “past” to the “present” period will be confirmed as a long-term trend in the area, the adaptability of grapevine production will have to rely on a larger use of

different, and possibly new, grape cvs. Crop Water Stress Index (CWSI) Further analysis was conducted by

means of the SWAP hydrological simulation model to estimate functional properties relevant to vineyard production. Specifically, the CWSIcum values were calculated and spatially distributed over the study area (Fig. 4). The minimum CWSIcum value (10.5 %) was found in the limestone mountains (see Fig. 1) and the maximum value (27.5 %) occurred in the pediment plain. The maximum CWSIcum corresponds to a Tact/Tpot ratio of 0.73. Trambouze and Woltz, 2001 measured a similar value of the ratio when 2/3 of soil water storage

was depleted. The maximum CWSIcum value therefore indicated a significant plant water shortage. Moreover, the results show a quite high variability of CWSIcum in the whole study area.

CONCLUSIONS The effects produced by climate change on potential vineyard distribution over an inland area

of Southern Italy (“Valle Telesina”) were analyzed. The different climatic pattern between two periods (“past” 1984-1996 and “present” 2000-2009) was assumed to be an example of future trends generated by climate changes.

The different thermal regime between the two periods was translated into differences in A&W indexes’ values; moreover the distribution of the Crop Water Stress Index was assessed for vineyards in the “present” climate.

The match between thermal regime and requirements of local cultivars allowed to assess a redistribution: traditional and widely distributed Falanghina and Aglianico would have a smaller area with favourable environment; whereas the extension of the area of cvs Forastera and Guarnaccia, the latter not cultivated in the “past” climate, would increase.

An interplay between the re-distribution of cultivars and the pattern of CWSIcum over the study area would cause different consequences on vines’ physiological processes, and therefore, on quantity and quality of production. The analysis could be further detailed to examine the effects of thermal and water regimes occurring at different phenological phases.

ACKNOWLEDGEMENTS Part of the present work was carried out within the project “Scenari di adattamento

dell’agricoltura italiana ai cambiamenti climatici – AGROSCENARI”.

Fig. 4 – Map of the CWSIcum values, cumulated from shooting to

harvesting, and averaged over the “present” climatic period.

6

REFERENCES Acevedo-Opazo C., Tisseyre B., Guillaume S., Ojeda, H., 2008. The potential of high spatial

resolution information to define within-vineyard zones related to vine water status. Precision Agric., 9:285-302.

Allen R., Pereira L.S., Raes D., Smith M. 1998. Crop evapotranspiration – Guidelines for computing crop water requirements. FAO Irrigation and Drainage Paper, vol. 56, Rome, Italy.

Amerine M.A., Winkler A.J., 1944. Composition and quality of must and wines of California grapes. Hilgardia, 15: 493-675.

Bonfante A., Basile A., Buonanno M., Manna P., Terribile F., 2005. Un esempio metodologico di utilizzo della modellistica idrologica e delle procedure GIS nella zonazione viticola. Bollettino della Società Italiana della Scienza del Suolo (SISS), 54: 1-2.

Ben-Asher J., van Dam J., Feddes R.A., Jhorar R.K., 2006. Irrigation of grapevines with saline water II. Mathematical simulation of vine growth and yield. Agric. Wat. Manag., 83: 22-29.

Carey V., 2001. Spatial characterization of natural terroir units for viticolture in the Bottelaryberg-Simonsberg-Helderberg wine growing area. M.Sc. Thesis, Stellenbosch University, Private Bag X1, 7602 Matieland (Stellenbosch), South Africa.

Kroes J.G., Van Dam J.C., 2003. Reference manual SWAP version 3.0.3. Alterra, Green World Research, Wageningen (The Netherlands).

Hargreaves G.L., Samani Z.A., 1985. Reference crop evapotranspiration from temperature. Applied Engineer. In Agric., 1, 2: 96-99.

IPCC, 2007. Climate Change 2007: Climate Change Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. In: Parry M.L., Canziani O.F., Palutikof J.P., van der Linden P.J., Hanson C.E. (Eds.). Cambridge University Press, Cambridge, UK.

Jones G.V., White M.A., Cooper O.R., Storchmann K., 2005. Climate change and global wine quality. Climatic Change, 73: 319–343.

MacMillan R.A., Pettapiece W.W., Nolan S. C., Goddard T.W., 2000. Generic procedure for automatically segmenting landforms into landform elements using DEMS, heuristic rules and fuzzy logic. Fuzzy Sets and Systems, 113:81-109.

Matthews M., Anderson M., 1988. Fruit ripening in Vitis vinifera L.: responses to seasonal water deficits. American Journal of Enology and Viticulture, 39: 313–320.

Medrano H., Escalona J.M., Cifre J., Bota J., Flexas J., 2003. A ten-year study on the physiology of two Spanish grapevine cultivars under field conditions: effect of water availability from leaf photosynthesis to grape yield and quality. Functional Plant Biology, 30:607-619.

Saayman D., 1977. The Effect of Soil and Climate on Wine Quality. In: Proc. International Symposium on the Quality of the Vintage, Cape Town, South Africa. 197–208.

Scaglione G., Pasquarella C., Federico R., Bonfante A., Terribile F., 2008. A multidisciplinary approach to grapevine zoning using GIS technology: An example of thermal data elaboration. Vitis, 47 (2):131-132.

Terribile F., Di Gennaro A., De Mascellis R., 1996. Carta dei suoli della Valle Telesina. Progetto U.O.T. Relazione finale convenzione CNR-ISPAIM-Regione Campania Assessorato alla Agricoltura.

Trambouze W., Voltz M. 2001. Measurement and modelling of the transpiration of a Mediterranean vineyard. Agric. For. Metorol, 107:153-166.

3 - 101



7

van Genuchten M. Th., 1980. A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J., 44:892-898.

Wösten J.H.M., Lilly A., Nemes A., Le Bas, C., 1998. Using existing soil data to derive hydraulic properties for simulation models in environmental studies and in land use planning. Report 156, Winand Staring Centre, The Netherlands.

CLIMATIC INFLUENCES ON MENCÍA GRAPEVINE PHENOLOGY AND GRAPE COMPOSITION

FOR AMANDI (RIBEIRA SACRA, SPAIN)

I. Rodríguez (1), J. Queijeiro (1), A. Masa(2), and M. Vilanova(2)

(1) Sciences Faculty of Ourense, Edificio Politécnico, As Lagos s/n 32004. Ourense (Spain).

(2) Misión Biológica de Galicia-CSIC. PO BOX 28. Pontevedra (Spain).

Email: [email protected]

ABSTRACT During the year 2009 we have studied the phenology and grape composition of Mencía

cultivar in seven different situations (orientation and altitude) for Amandi subzone (D.O. Ribeira

Sacra, Spain). The results showed the influence of terroir on the Mencía growth stages (budburst,

floraison, veraison, and harvest). All phenological data indicate that there is a delay in budburst

for V-2 of 15 days respect to V-5 and V-6. A delay for floraison also was found for V-2 and V-3

(8 days respect to the others vineyards). In the veraison the delay was for V-1 and V-2 (3 days)

respect to other vineyards studied. Significant differences were found in grape composition: total

acidity, pH, malic acid, color intensity and anthocyanins. The volatiles also were influenced by

the terroir, showed higher concentration of free compounds for V-2 (416 and SW) than the others

vineyards and the total bound composition shower the highest values for V-4.

KEYWORDS Mencía, Phenology, Amandi, Spain

INTRODUCTION Phenology is the study of the timing of natural phenomena that occur periodically in plants

and animals. For grapevines, phenology refers to the timing of grown stages and the influence of

climate and weather on them (Pearce, Coombe, 2004). Grapevines are grown in distinct climate

regimes worldwide that provide ideal situations to produce high quality grapes.

Amandi is a subzone of Denomination of Origen Ribeira Sacra, NW Spain. This area has some

orographic and weather characteristics that make it particularly suitable for growing grapes and

wines of high quality. With a south-southwest direction, the vineyards are protected from cold

winds from the north and the sun bathes the terraces throughout the day. The stone warmed by

the sun during the day blunted the lower night temperatures avoiding frost. These are the

characteristics that differentiate the Mencía grape in Amandi from other denominations and sub-

areas of the D.O. Ribeira Sacra.

The aim of this study was to know the effect of orientation and altitude in the phenological stages

of Mencía grapes from Amandi (D.O. Ribeira Sacra, Spain) and perform a multivariate analysis

to evaluate this possible differentiation according to the terroir.

3 - 102



7

van Genuchten M. Th., 1980. A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J., 44:892-898.

Wösten J.H.M., Lilly A., Nemes A., Le Bas, C., 1998. Using existing soil data to derive hydraulic properties for simulation models in environmental studies and in land use planning. Report 156, Winand Staring Centre, The Netherlands.

CLIMATIC INFLUENCES ON MENCÍA GRAPEVINE PHENOLOGY AND GRAPE COMPOSITION

FOR AMANDI (RIBEIRA SACRA, SPAIN)

I. Rodríguez (1), J. Queijeiro (1), A. Masa(2), and M. Vilanova(2)

(1) Sciences Faculty of Ourense, Edificio Politécnico, As Lagos s/n 32004. Ourense (Spain).

(2) Misión Biológica de Galicia-CSIC. PO BOX 28. Pontevedra (Spain).

Email: [email protected]

ABSTRACT During the year 2009 we have studied the phenology and grape composition of Mencía

cultivar in seven different situations (orientation and altitude) for Amandi subzone (D.O. Ribeira

Sacra, Spain). The results showed the influence of terroir on the Mencía growth stages (budburst,

floraison, veraison, and harvest). All phenological data indicate that there is a delay in budburst

for V-2 of 15 days respect to V-5 and V-6. A delay for floraison also was found for V-2 and V-3

(8 days respect to the others vineyards). In the veraison the delay was for V-1 and V-2 (3 days)

respect to other vineyards studied. Significant differences were found in grape composition: total

acidity, pH, malic acid, color intensity and anthocyanins. The volatiles also were influenced by

the terroir, showed higher concentration of free compounds for V-2 (416 and SW) than the others

vineyards and the total bound composition shower the highest values for V-4.

KEYWORDS Mencía, Phenology, Amandi, Spain

INTRODUCTION Phenology is the study of the timing of natural phenomena that occur periodically in plants

and animals. For grapevines, phenology refers to the timing of grown stages and the influence of

climate and weather on them (Pearce, Coombe, 2004). Grapevines are grown in distinct climate

regimes worldwide that provide ideal situations to produce high quality grapes.

Amandi is a subzone of Denomination of Origen Ribeira Sacra, NW Spain. This area has some

orographic and weather characteristics that make it particularly suitable for growing grapes and

wines of high quality. With a south-southwest direction, the vineyards are protected from cold

winds from the north and the sun bathes the terraces throughout the day. The stone warmed by

the sun during the day blunted the lower night temperatures avoiding frost. These are the

characteristics that differentiate the Mencía grape in Amandi from other denominations and sub-

areas of the D.O. Ribeira Sacra.

The aim of this study was to know the effect of orientation and altitude in the phenological stages

of Mencía grapes from Amandi (D.O. Ribeira Sacra, Spain) and perform a multivariate analysis

to evaluate this possible differentiation according to the terroir.

3 - 103




Viticulture Vitis vinifera Mencía grape grown in Amandi subzone from Denomination of Origin Ribeira

Sacra (Spain), during 2009 vintage, was considered in this study. Six vineyards (V-1 to V-6) with

different situation (altitude and orientation) were analyzed. The characteristics of the six

vineyards studied in Amandi are shown in Fig. 1 and Tab. 1.

Figure 1. Situation of vineyards in Amandi (D.O. Ribeira Sacra)

Table 1. Characteristics of vineyards studied in Amandi subzone from Ribeira Sacra.

Mencía Phenology The phenological data from these referenced vineyards are for the average dates of budburst,

floraison, veraison and harvest for 2009 vintage. The budburst, floraison and veraison are

considered to occur when, for a given varietal, 50% of the plants are exhibiting the physiological

response. Harvest data is recorder as the point as which, due to the optimun sugar levels, the

harvest commences.

Vineyard Altitude Latitude Length Coordinates UTM Orientation

V-1 466m 42º 24´ 38.58" N 7º 26´51.76" O 29T 0627801 4696742 153ºSSE

V-2 416m 42º 24´36.68" N 7º26´53.20" O 29T 0627796 4696703 213ºSW

V-3 352m 42º 24´34.20" N 7º 26´51.41" O 29T 0627759 4696620 162ºSSE

V-4 351m 42º 24´33.32" N 7º 26´56.81" O 29T 0627712 4696591 198ºS

V-5 355m 42º 24´32,85" N 7º 26´52.00" O 29T 0627798 4696590 144ºSE

V-6 240m 42º 24´27.97" N 7º 26´55.57" O 29T 0627741 4696426 136ºSE

Grape composition At harvest chemical analyses of Mencía grape must from the six different vineyards were

carried out. In each vineyard a sample of 300 berries from different points were collected. All

chemical analyses were carried out in triplicate by Foss analyzer. Volatiles (free and

glicosidically) were analized by GC-MS.

Climate The climatic conditions from D.O. Ribeira Sacra were analyzed in 2009 vintage. Micro

stations climatic-HOBOS were situated in the six points referenced (Tab. 1). The data consist of

daily observations of maximum, minimum and average of temperature. The data of mean,

maximum and minimum temperature vs altitude is showed in Fig. 2.

Statistical analysis An analysis of variance was performed using the XLSTAT statistical package (Addinsof,

2009). The effect of terroir (orientation, altitude and climate parameters) was evaluated using a

priori contrasts (p<0.05).

Figure 2. Temperature vs altitude in different vineyards of Amandi (D.O. Ribeira Sacra)

RESULTS AND DISCUSSION In Amandi subzone (D.O. Ribeira Sacra), phenological observations have been followed in six

Mencía vineyards with different altitude and orientation from the 2009 vintage. The phenological

data from these reference vineyards are showed in Tab. 2.

In the most viticulture regions, on average, budburst starts to occur when the mean daily

temperature exceeds 10ºC for five consecutive days (Amerine et al. 1980; Mullins et al. 1992).

Therefore, for 2009 vintage, the mean daily temperature was compiled and analyzed.

3 - 104




Viticulture Vitis vinifera Mencía grape grown in Amandi subzone from Denomination of Origin Ribeira

Sacra (Spain), during 2009 vintage, was considered in this study. Six vineyards (V-1 to V-6) with

different situation (altitude and orientation) were analyzed. The characteristics of the six

vineyards studied in Amandi are shown in Fig. 1 and Tab. 1.

Figure 1. Situation of vineyards in Amandi (D.O. Ribeira Sacra)

Table 1. Characteristics of vineyards studied in Amandi subzone from Ribeira Sacra.

Mencía Phenology The phenological data from these referenced vineyards are for the average dates of budburst,

floraison, veraison and harvest for 2009 vintage. The budburst, floraison and veraison are

considered to occur when, for a given varietal, 50% of the plants are exhibiting the physiological

response. Harvest data is recorder as the point as which, due to the optimun sugar levels, the

harvest commences.

Vineyard Altitude Latitude Length Coordinates UTM Orientation

V-1 466m 42º 24´ 38.58" N 7º 26´51.76" O 29T 0627801 4696742 153ºSSE

V-2 416m 42º 24´36.68" N 7º26´53.20" O 29T 0627796 4696703 213ºSW

V-3 352m 42º 24´34.20" N 7º 26´51.41" O 29T 0627759 4696620 162ºSSE

V-4 351m 42º 24´33.32" N 7º 26´56.81" O 29T 0627712 4696591 198ºS

V-5 355m 42º 24´32,85" N 7º 26´52.00" O 29T 0627798 4696590 144ºSE

V-6 240m 42º 24´27.97" N 7º 26´55.57" O 29T 0627741 4696426 136ºSE

Grape composition At harvest chemical analyses of Mencía grape must from the six different vineyards were

carried out. In each vineyard a sample of 300 berries from different points were collected. All

chemical analyses were carried out in triplicate by Foss analyzer. Volatiles (free and

glicosidically) were analized by GC-MS.

Climate The climatic conditions from D.O. Ribeira Sacra were analyzed in 2009 vintage. Micro

stations climatic-HOBOS were situated in the six points referenced (Tab. 1). The data consist of

daily observations of maximum, minimum and average of temperature. The data of mean,

maximum and minimum temperature vs altitude is showed in Fig. 2.

Statistical analysis An analysis of variance was performed using the XLSTAT statistical package (Addinsof,

2009). The effect of terroir (orientation, altitude and climate parameters) was evaluated using a

priori contrasts (p<0.05).

Figure 2. Temperature vs altitude in different vineyards of Amandi (D.O. Ribeira Sacra)

RESULTS AND DISCUSSION In Amandi subzone (D.O. Ribeira Sacra), phenological observations have been followed in six

Mencía vineyards with different altitude and orientation from the 2009 vintage. The phenological

data from these reference vineyards are showed in Tab. 2.

In the most viticulture regions, on average, budburst starts to occur when the mean daily

temperature exceeds 10ºC for five consecutive days (Amerine et al. 1980; Mullins et al. 1992).

Therefore, for 2009 vintage, the mean daily temperature was compiled and analyzed.

3 - 105



The mean data of budburst was 23 March and ranged from 24 March (eleven consecutive days

when the temperature exceeds 10ºC) for V-5 and V-6 to 8 April for V-2. The floraison occurred

as early 2 June for V-1, V-4, V-5 and V-6 and later as 10 June for V-2 and V-3. Budburst and

floraison are later in vineyards with South and West orientation (V-2 and V-3) and earlier in

vineyards oriented to east (V-5 and V-6). The average veraison data for these vineyards was 1 to

4 August. The veraison was produced for V-1 (466m) and V-2 (416m) later than the others

vineyards (with minor altitude). The harvest commenced 10 September for V-2, V-3, V-4 and V-

5 and the 11 September harvest data for V-1 and V-6 is the latest. In V-6 the grape size also is

higher.

Often more important than the date of phonological stage is the interval between stages, which

gives an indication of the overall climate during those periods (Jones and Davis, 2000). Short

intervals are associated with optimum conditions that facilitate rapid physiological growth and

differentiation (Coombe 1988). Long intervals among stages indicate less than ideal climate

conditions and a delay in growth and maturation (Calo and Tomasi, 1996; Gladstones, 1992).

One of the more important intervals is the length of the growing season (budburst to harvest) and

it was ranged from 172 days for V-5 and V-6 and 158 days for V-2. The interval between

floraison and veraison was 63 days (V-1) and 55 days (V-2). The maximum period of time from

flowering until harvest was 154 days.

All phenological data indicate that there is a delay in budburst for V-2 of 15 days respect to V-5

and V-6. A delay for floraison also was found for V-2 and V-3 (8 days respect to the others

vineyards). In the veraison the delay was for V-1 and V-2 (3 days) respect to the other vineyards

studied.

Table 2. Dates for major stages for Mencía grape variety grown in six vineyards situated to

different altitude in Amandi subzone from D.O. Ribeira Sacra (Spain).

In addition to phenology, grape composition has also been tabulated from the evaluation of the

reference vineyards. Fig. 3 shows the results for the general chemical analysis (sugar content,

potential ethanol, pH and total acidity) of musts obtained from Mencía cultivar grown in Amandi

during ripening. V-5 showed the highest values for reducing sugar and therefore potential ethanol

during ripening. Total acidity was highest for V-1.

Near harvest time, the key vintage quality characteristics are the chemical composition of grapes

(Jones, Davis, 2000). Tab. 3 shows the musts composition of Mencía cultivar grown in Amandi

at harvest. Significant differences were found for five parameters among samples: Total acidity,

pH, malic acid, color intensity and anthocyanins. The highest value of glucose+fructose, Brix

was for V-5, vineyard sited to 355m and with orientation SE. The total acidity and tartaric acid of

the musts was higher for V-1 (466m, SSE). V-2 (416, SW) showed higher malic acid than the

other vineyards studied and V-3 showed the highest color intensity.

Major stages V-1 V-2 V-3 V-4 V-5 V-6

Budburst 27-Mar 8-Apr 31-Mar 27-Mar 24-Mar 24-Mar

Floraison 2-Jun 10-Jun 10-Jun 2-Jun 2-Jun 2-Jun

Veraison 4-Aug 4-Aug 1-Aug 1-Aug 1-Aug 1-Aug

Harvest 11-Sep 10-Sep 10-Sep 10-Sep 10-Sep 11-Sep

Weigth/100 berries (g) 196 259 217 235 226 319

Figure 3. Chemical composition of Mencía grapevine during ripening from Amandi.

Parameters V-1 V-2 V-3 V-4 V-5 V-6 Sig Glucose-Fructose (g/L) 219.50 182.00 210.50 217.00 223.00 207.00 ns

ºBrix 21.90 18.90 21.30 21.75 22.30 21.00 ns

Total acidity (g/L) 3.96 2.59 2.48 2.37 2.14 1.97 ***

pH 3.25 3.54 3.56 3.49 3.60 3.52 **

Tartaric acid (g/L) 4.55 3.80 4.20 4.20 4.45 4.00 ns

Malic acid (g/L) 1.30 1.70 1.35 1.00 0.65 0.70 **

Folin index 231.20 202.30 233.25 187.65 231.75 177.80 ns

Color intensity 4.30 4.20 5.10 4.65 4.85 4.10 *

Anthocyanins (mg/L) 50.50 117.00 99.50 80.00 110.50 81.00 **

Table 3. Chemical composition, at harvest, for vineyards referenced in Amandi.

Fig. 4 shows the total concentration of free (A) and bound (B) volatile compounds identified the

references vineyards in Amandi at harvest.

The total concentration was obtained as the sum of individual concentrations of all compounds

detected under the experimental conditions used, including C6-compounds, alcohols,

monotepenes, C13-norisoprenoids, volatile acids, volatile phenols and carbonyl compounds. The

total free volatile composition was higher for V-2 (416 and SW) than the others vineyards and the

total bound composition shower the highest values for V-4, the only vineyard oriented to SE in

our study.

3 - 106



The mean data of budburst was 23 March and ranged from 24 March (eleven consecutive days

when the temperature exceeds 10ºC) for V-5 and V-6 to 8 April for V-2. The floraison occurred

as early 2 June for V-1, V-4, V-5 and V-6 and later as 10 June for V-2 and V-3. Budburst and

floraison are later in vineyards with South and West orientation (V-2 and V-3) and earlier in

vineyards oriented to east (V-5 and V-6). The average veraison data for these vineyards was 1 to

4 August. The veraison was produced for V-1 (466m) and V-2 (416m) later than the others

vineyards (with minor altitude). The harvest commenced 10 September for V-2, V-3, V-4 and V-

5 and the 11 September harvest data for V-1 and V-6 is the latest. In V-6 the grape size also is

higher.

Often more important than the date of phonological stage is the interval between stages, which

gives an indication of the overall climate during those periods (Jones and Davis, 2000). Short

intervals are associated with optimum conditions that facilitate rapid physiological growth and

differentiation (Coombe 1988). Long intervals among stages indicate less than ideal climate

conditions and a delay in growth and maturation (Calo and Tomasi, 1996; Gladstones, 1992).

One of the more important intervals is the length of the growing season (budburst to harvest) and

it was ranged from 172 days for V-5 and V-6 and 158 days for V-2. The interval between

floraison and veraison was 63 days (V-1) and 55 days (V-2). The maximum period of time from

flowering until harvest was 154 days.

All phenological data indicate that there is a delay in budburst for V-2 of 15 days respect to V-5

and V-6. A delay for floraison also was found for V-2 and V-3 (8 days respect to the others

vineyards). In the veraison the delay was for V-1 and V-2 (3 days) respect to the other vineyards

studied.

Table 2. Dates for major stages for Mencía grape variety grown in six vineyards situated to

different altitude in Amandi subzone from D.O. Ribeira Sacra (Spain).

In addition to phenology, grape composition has also been tabulated from the evaluation of the

reference vineyards. Fig. 3 shows the results for the general chemical analysis (sugar content,

potential ethanol, pH and total acidity) of musts obtained from Mencía cultivar grown in Amandi

during ripening. V-5 showed the highest values for reducing sugar and therefore potential ethanol

during ripening. Total acidity was highest for V-1.

Near harvest time, the key vintage quality characteristics are the chemical composition of grapes

(Jones, Davis, 2000). Tab. 3 shows the musts composition of Mencía cultivar grown in Amandi

at harvest. Significant differences were found for five parameters among samples: Total acidity,

pH, malic acid, color intensity and anthocyanins. The highest value of glucose+fructose, Brix

was for V-5, vineyard sited to 355m and with orientation SE. The total acidity and tartaric acid of

the musts was higher for V-1 (466m, SSE). V-2 (416, SW) showed higher malic acid than the

other vineyards studied and V-3 showed the highest color intensity.

Major stages V-1 V-2 V-3 V-4 V-5 V-6

Budburst 27-Mar 8-Apr 31-Mar 27-Mar 24-Mar 24-Mar

Floraison 2-Jun 10-Jun 10-Jun 2-Jun 2-Jun 2-Jun

Veraison 4-Aug 4-Aug 1-Aug 1-Aug 1-Aug 1-Aug

Harvest 11-Sep 10-Sep 10-Sep 10-Sep 10-Sep 11-Sep

Weigth/100 berries (g) 196 259 217 235 226 319

Figure 3. Chemical composition of Mencía grapevine during ripening from Amandi.

Parameters V-1 V-2 V-3 V-4 V-5 V-6 Sig Glucose-Fructose (g/L) 219.50 182.00 210.50 217.00 223.00 207.00 ns

ºBrix 21.90 18.90 21.30 21.75 22.30 21.00 ns

Total acidity (g/L) 3.96 2.59 2.48 2.37 2.14 1.97 ***

pH 3.25 3.54 3.56 3.49 3.60 3.52 **

Tartaric acid (g/L) 4.55 3.80 4.20 4.20 4.45 4.00 ns

Malic acid (g/L) 1.30 1.70 1.35 1.00 0.65 0.70 **

Folin index 231.20 202.30 233.25 187.65 231.75 177.80 ns

Color intensity 4.30 4.20 5.10 4.65 4.85 4.10 *

Anthocyanins (mg/L) 50.50 117.00 99.50 80.00 110.50 81.00 **

Table 3. Chemical composition, at harvest, for vineyards referenced in Amandi.

Fig. 4 shows the total concentration of free (A) and bound (B) volatile compounds identified the

references vineyards in Amandi at harvest.

The total concentration was obtained as the sum of individual concentrations of all compounds

detected under the experimental conditions used, including C6-compounds, alcohols,

monotepenes, C13-norisoprenoids, volatile acids, volatile phenols and carbonyl compounds. The

total free volatile composition was higher for V-2 (416 and SW) than the others vineyards and the

total bound composition shower the highest values for V-4, the only vineyard oriented to SE in

our study.

3 - 107



Figure 4. Free and bound compounds concentration (µg/L) of Mencía grapevine

CONCLUSIONS This work showed the effect terroir (orientation and altitude) on phenology and chemical and

volatile composition of Mencía grapevine from Amandi.

Budburst and floraison are later in vineyards with South and West orientation (V-2 and V-3) and

with high values of altitude and earlier in vineyards oriented to East (V-5 and V-6) and with low

values of altitude. In wine composition the highest total acidities and malic acid were found for

V-1, V-2 and V-3 according to phonological data.

ACKNOWLEDGMENTS We would like to thank the Consellería de Innovación e Industria from Xunta de Galicia

(Spain) for the financial support of this research project (08MRU029403PR) and Isidro Parga

Pondal” program. We would like to acknowledge to the D. O. Ribeira Sacra for their assistance. BIBLIOGRAPHY

Amerine M.A. and Berg H.W., 1980. The technology of wine making (4 ed). 795 pp. AVI

Publishers Company, Inc., Westport, CT.

Calo A. and Tomasi D., 1996. Relatuionship between environmental factors and the dynamics of

growth and composition of the grapevine. Proceedings of the workshop strategies to optimize

wine grape quality. S Poni, E. Peterlunger et al (Eds) pp 217-231. Acta hhorticulturae.

Coombe B.G., 1988. Grapevine phenology. In: Viticulture, Vol. 1, pp.139-153. Australian

Industrial Publishers, Adelaide.

Gladstones J., 1992. Viticulture and Environment. 310 pp. Winetitles, Adelaide.

Jones G.V. and Davis R.E., 2000. Climate influences on grapevine phenology, grape


Enology and Viticulture, 51: 249-261.

Mullins M.G., Bouquet A. and Williams L.E.,1992. Biology of grapevine 239 pp. Cambridge

University Press, Great Britain.

Pearce I. and Coombe B.G., 2004. Grapevine phenology. In: Viticulture, Vol. 1, pp.150-166.

Australian Industrial Publishers, Adelaide.

ANALYSE CLIMATIQUE A L’ECHELLE DES COTEAUX DU LAYON

C. Bonnefoy1, H. Quénol1, G. Barbeau2, M. Madelin31 Laboratoire COSTEL, UMR6554 LETG du CNRS, Université Rennes 2 - Haute Bretagne, Rennes.

[email protected] [email protected] 2 INRA UE1117, UMT Vinitera, Beaucouzé, [email protected]

3 PRODIG, UMR CNRS 8586, université Paris Diderot [email protected]

RESUME Les études d’impact du climat sur la vigne nécessite de descendre à des échelles très fines

car les facteurs climatiques sont tributaires de la topographie, la végétation, les expositions … Dans le cadre du programme ANR-JC Terviclim, 22 capteurs ont été installés dans les vignobles des Coteaux du Layon afin de caractériser le climat particulier de ces terroirs. L’analyse des températures montre de fortes disparités entre les data loggers et pourtant situés parfois sur les mêmes parcelles ou sur des parcelles voisines. Les indices bioclimatiques tels les degrés jours sont également contrastés suivant la situation des capteurs sur les coteaux.

MOTSCLEEtudes d’impact – vigne – échelles fines – Coteaux du Layon – terroirs – indices

bioclimatiques

ABSTRACT Climate impact studies on vine require downscaling because climatic factors depend on

topography, vegetation, orientation …In the framework of the ANR-JC Terviclim, 22 data loggers were settled in the “Coteaux du Layon” vineyards to characterize the particular climate of these terroirs. Temperatures analysis shows strong disparities between data loggers locate on the same plots or on nearby plots. Bioclimatic index as growing degree days are also contrasting depending on the data loggers situation in the vineyard.

KEYWORDSImpact studies –vine – downscaling – Coteaux du Layon – terroirs – bioclimatic index

INTRODUCTION

Les nombreuses interrogations posées par le changement climatique engendrent une multitude de questions sur le fonctionnement des géosystèmes aux échelles locales. Un changement global du climat aura obligatoirement des répercussions sur le climat local et sur les terroirs viticoles. Dans ce contexte, les impacts attendus d’un éventuel changement climatique posent un certain nombre de questions, ne serait-ce que pour améliorer l’adaptation.

Les changements climatiques en cours engendrent des modifications dans le cycle phénologique des plantes et modifieront très certainement la géographie de certains cépages. La vigne est sujette à ces changements avec une avancée des stades phénologiques déjà observée (Jones et al., 2005 ; Madelin et al., 2008). Afin d’étudier les futurs impacts du

3 - 108



Figure 4. Free and bound compounds concentration (µg/L) of Mencía grapevine

CONCLUSIONS This work showed the effect terroir (orientation and altitude) on phenology and chemical and

volatile composition of Mencía grapevine from Amandi.

Budburst and floraison are later in vineyards with South and West orientation (V-2 and V-3) and

with high values of altitude and earlier in vineyards oriented to East (V-5 and V-6) and with low

values of altitude. In wine composition the highest total acidities and malic acid were found for

V-1, V-2 and V-3 according to phonological data.

ACKNOWLEDGMENTS We would like to thank the Consellería de Innovación e Industria from Xunta de Galicia

(Spain) for the financial support of this research project (08MRU029403PR) and Isidro Parga

Pondal” program. We would like to acknowledge to the D. O. Ribeira Sacra for their assistance. BIBLIOGRAPHY

Amerine M.A. and Berg H.W., 1980. The technology of wine making (4 ed). 795 pp. AVI

Publishers Company, Inc., Westport, CT.

Calo A. and Tomasi D., 1996. Relatuionship between environmental factors and the dynamics of

growth and composition of the grapevine. Proceedings of the workshop strategies to optimize

wine grape quality. S Poni, E. Peterlunger et al (Eds) pp 217-231. Acta hhorticulturae.

Coombe B.G., 1988. Grapevine phenology. In: Viticulture, Vol. 1, pp.139-153. Australian

Industrial Publishers, Adelaide.

Gladstones J., 1992. Viticulture and Environment. 310 pp. Winetitles, Adelaide.

Jones G.V. and Davis R.E., 2000. Climate influences on grapevine phenology, grape


Enology and Viticulture, 51: 249-261.

Mullins M.G., Bouquet A. and Williams L.E.,1992. Biology of grapevine 239 pp. Cambridge

University Press, Great Britain.

Pearce I. and Coombe B.G., 2004. Grapevine phenology. In: Viticulture, Vol. 1, pp.150-166.

Australian Industrial Publishers, Adelaide.

ANALYSE CLIMATIQUE A L’ECHELLE DES COTEAUX DU LAYON

C. Bonnefoy1, H. Quénol1, G. Barbeau2, M. Madelin31 Laboratoire COSTEL, UMR6554 LETG du CNRS, Université Rennes 2 - Haute Bretagne, Rennes.

[email protected] [email protected] 2 INRA UE1117, UMT Vinitera, Beaucouzé, [email protected]

3 PRODIG, UMR CNRS 8586, université Paris Diderot [email protected]

RESUME Les études d’impact du climat sur la vigne nécessite de descendre à des échelles très fines

car les facteurs climatiques sont tributaires de la topographie, la végétation, les expositions … Dans le cadre du programme ANR-JC Terviclim, 22 capteurs ont été installés dans les vignobles des Coteaux du Layon afin de caractériser le climat particulier de ces terroirs. L’analyse des températures montre de fortes disparités entre les data loggers et pourtant situés parfois sur les mêmes parcelles ou sur des parcelles voisines. Les indices bioclimatiques tels les degrés jours sont également contrastés suivant la situation des capteurs sur les coteaux.

MOTSCLEEtudes d’impact – vigne – échelles fines – Coteaux du Layon – terroirs – indices

bioclimatiques

ABSTRACT Climate impact studies on vine require downscaling because climatic factors depend on

topography, vegetation, orientation …In the framework of the ANR-JC Terviclim, 22 data loggers were settled in the “Coteaux du Layon” vineyards to characterize the particular climate of these terroirs. Temperatures analysis shows strong disparities between data loggers locate on the same plots or on nearby plots. Bioclimatic index as growing degree days are also contrasting depending on the data loggers situation in the vineyard.

KEYWORDSImpact studies –vine – downscaling – Coteaux du Layon – terroirs – bioclimatic index

INTRODUCTION

Les nombreuses interrogations posées par le changement climatique engendrent une multitude de questions sur le fonctionnement des géosystèmes aux échelles locales. Un changement global du climat aura obligatoirement des répercussions sur le climat local et sur les terroirs viticoles. Dans ce contexte, les impacts attendus d’un éventuel changement climatique posent un certain nombre de questions, ne serait-ce que pour améliorer l’adaptation.

Les changements climatiques en cours engendrent des modifications dans le cycle phénologique des plantes et modifieront très certainement la géographie de certains cépages. La vigne est sujette à ces changements avec une avancée des stades phénologiques déjà observée (Jones et al., 2005 ; Madelin et al., 2008). Afin d’étudier les futurs impacts du

3 - 109



réchauffement climatique sur les vignobles, le programme ANR-JC07194103 TERVICLIM1

propose de définir le climat actuel de plusieurs vignobles dans le monde dont celui du Val de Loire. L’étude du climat à l’échelle d’un terroir viticole se réalise suivant une démarche d’échelles spatiales imbriquées. Les raisons de cette approche sont liées au fonctionnement de l'atmosphère. Sachant qu'une même masse d'air, générant un même type de temps peut concerner un territoire étendu il est certain que sur une entité spatiale plus petite, des caractères communs sont présents. Plus le niveau d'observation s'affine, plus le nombre d'éléments influant sur les paramètres météorologiques est grand et schématiquement, la compréhension augmente au fur et à mesure que l'espace se réduit.

Une étude diagnostique à l’échelle régionale sur l’importance du réchauffement climatique a déjà été menée dans le Val de Loire à partir des stations du réseau météorologique de Météo-France. Les résultats montrent un réchauffement généralisé avec une rupture climatique en 1987 concernant les températures moyennes (Bonnefoy et al., 2010). Afin de caractériser le climat à une échelle plus fine, des stations complètes Campbell (température de l’air, humidité de l’air, vitesse et direction du vent, insolation, précipitations) ont été installées dans les vignobles de l’Anjou et des capteurs de températures type « Tinytag » dans les Coteaux du Layon, la plupart dans les vignobles de l’appellation « Quart de Chaume ». Ces capteurs ont été installés en fonction des caractéristiques topographiques (altitude, pente, exposition) et de la nature de l’occupation du sol (type de sol, distance à la rivière « Le Layon », …). Ainsi, une bonne connaissance des conditions climatiques multi-échelles des terroirs permettra de se projeter de manière plus confiante dans l’avenir grâce aux simulations futures du modèle climatique régionale RAMS (Regional Atmospheric Modelling System).

Nous présentons dans cet article les résultats préliminaires concernant une partie des terroirs des coteaux du Layon (en particulier l’appellation « Quarts de Chaume »), situés au sud-ouest d’Angers. Le climat particulier de ces terroirs permet la production de vin liquoreux suite à la formation d’un champignon (Botrytis Cinerea) et à des vendanges tardives (Duchêne, 1996), en général au mois d’octobre voire novembre. Nous présentons l’analyse des données de températures des capteurs sur la saison végétative de la vigne de 2009 ainsi que des indices bioclimatiques afin de caractériser les nuances climatiques de ce secteur.

DONNEES ET METHODES

Afin de caractériser le climat des coteaux du Layon, 22 capteurs de températures type « Tinytag » ont été disposé au sein même des vignobles, durant deux campagnes de terrain, une au printemps 2008 et une autre au printemps 2009. Ces capteurs sont directement implantés sur les piquets de vigne et sont situés sous abri à 1 mètre du sol. Pour une bonne représentativité du climat, les data loggers ont été installés en prenant en compte la topographie, les pentes, les expositions, la nature du sol …Ces informations ont été obtenues sous SIG (système d’information géographique) à partir du modèle numérique de terrain et de la carte des unités de terroirs de base de l’Institut National de Recherche Agronomique (INRA). L’ensemble des capteurs sont répartis sur un secteur d’environ 600 ha (Fig. 1), la majorité d’entre eux étant situé dans l’appellation Quart de Chaumes. Cette appellation est la plus prestigieuse des coteaux du Layon puisqu’il a été reconnu par décret que ce secteur, exposé au sud, permet le développement quasi-régulier de la pourriture noble à l’origine des vins liquoreux (Institut National des Appelations d’Origine). La plupart des capteurs se 1 « Observation et modélisation spatiale du climat à l’échelle des terroirs viticoles dans un contexte de changement climatique »

trouvent donc en exposition sud, mais parfois avec des conditions très différentes les uns des autres. Par exemple, le capteur Bellevue 1, situé à quelques dizaines de mètres du capteur Bellevue 2, est disposé sur un sol peu profond et par conséquent très humide alors que l’autre capteur est sur un sol beaucoup plus profond et donc plus sec. Deux stations Campbell viennent compléter ce réseau, celle de Chaumes, dans le fond de vallée et celle de Beaulieu en haut de coteau.

Fig 1. Cadre d’étude à différentes échelles et localisation des capteurs de températures dans les coteaux du Layon

Les « Tinytags » et les stations ont été programmés afin de fournir une donnée de température toutes les 15 minutes. La saison phénologique de la vigne commence le 1er avril mais ne disposant des données que depuis le 18 avril 2009 pour la plupart des capteurs, nous avons calculé les moyennes de températures mensuelles de mai à octobre et les indices bioclimatiques ont été calculés à partir de cette date. Ces indices sont à l’origine utilisés à une échelle régionale et permettent la classification des vignobles dans différentes catégories climatiques. Cependant, aux échelles fines, ces indices peuvent donner une information supplémentaire sur la variabilité climatique entre et au sein des terroirs. Nous nous sommes basés sur l’indice de Winkler (Winkler et al., 1974) pour les degrés jours. Les degrés jours de Winkler correspondent à la somme des températures moyennes journalières (Tmj) à partir de la base de 10°C qui est effectuée du 1er avril au 31 octobre : degrés-jours = ∑ Tmj – 10 (avec Tmj > 10). Cet indice permet une régionalisation climatique en cinq catégories du climat le plus frais au plus chaud.

Nous avons également calculé un autre indice basé sur le calcul d’Huglin (Huglin et Schneider, 1998). La valeur de cet indice correspond à la valeur cumulée du 1er avril au 30

3 - 110



réchauffement climatique sur les vignobles, le programme ANR-JC07194103 TERVICLIM1

propose de définir le climat actuel de plusieurs vignobles dans le monde dont celui du Val de Loire. L’étude du climat à l’échelle d’un terroir viticole se réalise suivant une démarche d’échelles spatiales imbriquées. Les raisons de cette approche sont liées au fonctionnement de l'atmosphère. Sachant qu'une même masse d'air, générant un même type de temps peut concerner un territoire étendu il est certain que sur une entité spatiale plus petite, des caractères communs sont présents. Plus le niveau d'observation s'affine, plus le nombre d'éléments influant sur les paramètres météorologiques est grand et schématiquement, la compréhension augmente au fur et à mesure que l'espace se réduit.

Une étude diagnostique à l’échelle régionale sur l’importance du réchauffement climatique a déjà été menée dans le Val de Loire à partir des stations du réseau météorologique de Météo-France. Les résultats montrent un réchauffement généralisé avec une rupture climatique en 1987 concernant les températures moyennes (Bonnefoy et al., 2010). Afin de caractériser le climat à une échelle plus fine, des stations complètes Campbell (température de l’air, humidité de l’air, vitesse et direction du vent, insolation, précipitations) ont été installées dans les vignobles de l’Anjou et des capteurs de températures type « Tinytag » dans les Coteaux du Layon, la plupart dans les vignobles de l’appellation « Quart de Chaume ». Ces capteurs ont été installés en fonction des caractéristiques topographiques (altitude, pente, exposition) et de la nature de l’occupation du sol (type de sol, distance à la rivière « Le Layon », …). Ainsi, une bonne connaissance des conditions climatiques multi-échelles des terroirs permettra de se projeter de manière plus confiante dans l’avenir grâce aux simulations futures du modèle climatique régionale RAMS (Regional Atmospheric Modelling System).

Nous présentons dans cet article les résultats préliminaires concernant une partie des terroirs des coteaux du Layon (en particulier l’appellation « Quarts de Chaume »), situés au sud-ouest d’Angers. Le climat particulier de ces terroirs permet la production de vin liquoreux suite à la formation d’un champignon (Botrytis Cinerea) et à des vendanges tardives (Duchêne, 1996), en général au mois d’octobre voire novembre. Nous présentons l’analyse des données de températures des capteurs sur la saison végétative de la vigne de 2009 ainsi que des indices bioclimatiques afin de caractériser les nuances climatiques de ce secteur.

DONNEES ET METHODES

Afin de caractériser le climat des coteaux du Layon, 22 capteurs de températures type « Tinytag » ont été disposé au sein même des vignobles, durant deux campagnes de terrain, une au printemps 2008 et une autre au printemps 2009. Ces capteurs sont directement implantés sur les piquets de vigne et sont situés sous abri à 1 mètre du sol. Pour une bonne représentativité du climat, les data loggers ont été installés en prenant en compte la topographie, les pentes, les expositions, la nature du sol …Ces informations ont été obtenues sous SIG (système d’information géographique) à partir du modèle numérique de terrain et de la carte des unités de terroirs de base de l’Institut National de Recherche Agronomique (INRA). L’ensemble des capteurs sont répartis sur un secteur d’environ 600 ha (Fig. 1), la majorité d’entre eux étant situé dans l’appellation Quart de Chaumes. Cette appellation est la plus prestigieuse des coteaux du Layon puisqu’il a été reconnu par décret que ce secteur, exposé au sud, permet le développement quasi-régulier de la pourriture noble à l’origine des vins liquoreux (Institut National des Appelations d’Origine). La plupart des capteurs se 1 « Observation et modélisation spatiale du climat à l’échelle des terroirs viticoles dans un contexte de changement climatique »

trouvent donc en exposition sud, mais parfois avec des conditions très différentes les uns des autres. Par exemple, le capteur Bellevue 1, situé à quelques dizaines de mètres du capteur Bellevue 2, est disposé sur un sol peu profond et par conséquent très humide alors que l’autre capteur est sur un sol beaucoup plus profond et donc plus sec. Deux stations Campbell viennent compléter ce réseau, celle de Chaumes, dans le fond de vallée et celle de Beaulieu en haut de coteau.

Fig 1. Cadre d’étude à différentes échelles et localisation des capteurs de températures dans les coteaux du Layon

Les « Tinytags » et les stations ont été programmés afin de fournir une donnée de température toutes les 15 minutes. La saison phénologique de la vigne commence le 1er avril mais ne disposant des données que depuis le 18 avril 2009 pour la plupart des capteurs, nous avons calculé les moyennes de températures mensuelles de mai à octobre et les indices bioclimatiques ont été calculés à partir de cette date. Ces indices sont à l’origine utilisés à une échelle régionale et permettent la classification des vignobles dans différentes catégories climatiques. Cependant, aux échelles fines, ces indices peuvent donner une information supplémentaire sur la variabilité climatique entre et au sein des terroirs. Nous nous sommes basés sur l’indice de Winkler (Winkler et al., 1974) pour les degrés jours. Les degrés jours de Winkler correspondent à la somme des températures moyennes journalières (Tmj) à partir de la base de 10°C qui est effectuée du 1er avril au 31 octobre : degrés-jours = ∑ Tmj – 10 (avec Tmj > 10). Cet indice permet une régionalisation climatique en cinq catégories du climat le plus frais au plus chaud.

Nous avons également calculé un autre indice basé sur le calcul d’Huglin (Huglin et Schneider, 1998). La valeur de cet indice correspond à la valeur cumulée du 1er avril au 30

3 - 111



septembre. Ainsi, l’indice héliothermique d’Huglin est calculé selon la formule : IH = ∑ [(Tm-10)+(Tx-10)/2]*k, (où Tm = Température moyenne, Tx = Température maximale et k le coefficient de longueur du jour variant de 1,02 à 1,06 entre 40 et 50 degrés de latitude). Cet indice permet de la même façon de classer les climats viticoles en six catégories.

RESULTATS ET DISCUSSION

1. Une forte variabilité spatiale des températures

La Fig.2 montre l’évolution des températures minimales et maximales moyennes mensuelles sur la saison phénologique pour cinq capteurs de température. Ces cinq capteurs permettent de montrer assez clairement les contrastes thermiques observés dans les vignobles du fait de leur situation (Fig.1) très contrastée (haut et bas de coteau, exposition nord ou sud …).

Fig 2. Evolution des températures minimales (a) et maximales (b) moyennes de mai à octobre 2009 pour 5 capteurs de température dans les coteaux du Layon

L’analyse des températures minimales moyennes montrent d’importants contrastes thermiques à cette échelle. La température minimale moyenne la plus basse est observée pour l’ensemble des mois sur le capteur « Suronde_fond » avec une moyenne de 9,5°C sur la période mai-octobre. Ce capteur est situé en versant sud-est dans le fond de la vallée du Layon à une altitude de 33 mètres. Les capteurs, situés en milieu et haut de coteau sur le même versant, ont un comportement thermique complètement différent. Le capteur enregistrant la température minimale moyenne la plus élevée est « Papin_haut » qui est situé à 82 mètres en haut de coteau en exposition sud-ouest. La température minimale moyenne sur la même période pour ce capteur est de 11,4°C soit pratiquement 2°C de plus que celui de « Suronde_fond ». Le capteur situé en haut de parcelle et enregistrant la température minimale moyenne la plus basse est « Jolivet_haut ». Ce capteur est situé en versant nord et à une altitude de 32 mètres.

Les températures maximales moyennes les plus élevées sont observées à « Suronde_fond », donc dans le bas de coteau. La température maximale moyenne sur la période est de 26,1°C pour ce capteur de fond de vallée. Les plus basses températures maximales sont alors observées en haut de coteau comme au Breuil (capteur le plus élevé à 90 mètres) ou à « Papin_haut ». Les moyennes des températures maximales sur cette période sont respectivement de 24°C et 23,9°C pour ces deux capteurs de haut de coteau. En conséquence,

les amplitudes thermiques les plus fortes sont observées en bas de coteau alors que les amplitudes sont plus modérées sur les hauts de coteau.

Ces disparités climatiques non négligeables s’expliquent par la topographie variée du secteur des Coteaux du Layon avec notamment la mise en place d’inversions thermiques assez fréquentes lors de nuits radiatives claires (Bonnefoy et al., 2009). Ainsi, l’air froid a tendance à stagner dans les bas fonds à proximité du Layon, ce qui peut poser des problèmes de gels tardifs en période de débourrement de la vigne. Les plateaux et hauts de coteaux sont moins exposés à ce risque de gel mais ils sont défavorisés par les températures maximales qui augmentent moins en journée du fait de l’altitude et de la pente moins marquée.

2. Des indices bioclimatiques contrastés

La forte variabilité spatiale des températures se répercutent sur celle des indices bioclimatiques. Une nouvelle fois, si ces indices sont à la base utilisée à une échelle régionale et peuvent montrer d’importantes disparités au sein d’une même région, comme dans le Val de Loire (Bonnefoy et al ; 2009), ils permettent à très fine échelle de faire ressortir les disparités climatiques liées aux caractéristiques du milieu.

Fig 3. Cumul des degrés jours dans les coteaux du Layon (à gauche)et indice basé sur l’indice de Huglin (à droite)

Les calculs du cumul des degrés jours et de l’indice de Huglin n’ont été effectués que sur 20 capteurs car deux ont des données manquantes. La Fig. 3 montre assez clairement que les cumuls de degrés jours les plus faibles sont observés en versant nord ou en fond de vallée et en haut de coteaux. En revanche, les cumuls les plus élevés se trouvent en général à mi-coteau et en versant sud. Ainsi, le cumul des degrés jours (dj) varient de 1184 dj à 1480 dj. Ces observations laissent suggérer que celles-ci auront des conséquences sur le déroulement du cycle phénologique de la vigne et notamment sur la date de maturité.

L’indice basé sur l’indice de Huglin montre également une assez forte variabilité au sein des vignobles (valeurs variant de 1890 à 2190). Les valeurs les plus faibles sont localisées en haut de coteau alors que les valeurs les plus fortes sont clairement en bas et à mi-coteau. Nous observons des indices relativement élevés en versant nord, contrairement aux degrés jours, du fait que le calcul de l’indice de Huglin donne un poids plus important aux températures maximales. En effet les températures maximales sont assez élevées sur ce versant du fait de son altitude moindre. Enfin, l’importance du type de sol est mise en évidence par les capteurs

3 - 112



septembre. Ainsi, l’indice héliothermique d’Huglin est calculé selon la formule : IH = ∑ [(Tm-10)+(Tx-10)/2]*k, (où Tm = Température moyenne, Tx = Température maximale et k le coefficient de longueur du jour variant de 1,02 à 1,06 entre 40 et 50 degrés de latitude). Cet indice permet de la même façon de classer les climats viticoles en six catégories.

RESULTATS ET DISCUSSION

1. Une forte variabilité spatiale des températures

La Fig.2 montre l’évolution des températures minimales et maximales moyennes mensuelles sur la saison phénologique pour cinq capteurs de température. Ces cinq capteurs permettent de montrer assez clairement les contrastes thermiques observés dans les vignobles du fait de leur situation (Fig.1) très contrastée (haut et bas de coteau, exposition nord ou sud …).

Fig 2. Evolution des températures minimales (a) et maximales (b) moyennes de mai à octobre 2009 pour 5 capteurs de température dans les coteaux du Layon

L’analyse des températures minimales moyennes montrent d’importants contrastes thermiques à cette échelle. La température minimale moyenne la plus basse est observée pour l’ensemble des mois sur le capteur « Suronde_fond » avec une moyenne de 9,5°C sur la période mai-octobre. Ce capteur est situé en versant sud-est dans le fond de la vallée du Layon à une altitude de 33 mètres. Les capteurs, situés en milieu et haut de coteau sur le même versant, ont un comportement thermique complètement différent. Le capteur enregistrant la température minimale moyenne la plus élevée est « Papin_haut » qui est situé à 82 mètres en haut de coteau en exposition sud-ouest. La température minimale moyenne sur la même période pour ce capteur est de 11,4°C soit pratiquement 2°C de plus que celui de « Suronde_fond ». Le capteur situé en haut de parcelle et enregistrant la température minimale moyenne la plus basse est « Jolivet_haut ». Ce capteur est situé en versant nord et à une altitude de 32 mètres.

Les températures maximales moyennes les plus élevées sont observées à « Suronde_fond », donc dans le bas de coteau. La température maximale moyenne sur la période est de 26,1°C pour ce capteur de fond de vallée. Les plus basses températures maximales sont alors observées en haut de coteau comme au Breuil (capteur le plus élevé à 90 mètres) ou à « Papin_haut ». Les moyennes des températures maximales sur cette période sont respectivement de 24°C et 23,9°C pour ces deux capteurs de haut de coteau. En conséquence,

les amplitudes thermiques les plus fortes sont observées en bas de coteau alors que les amplitudes sont plus modérées sur les hauts de coteau.

Ces disparités climatiques non négligeables s’expliquent par la topographie variée du secteur des Coteaux du Layon avec notamment la mise en place d’inversions thermiques assez fréquentes lors de nuits radiatives claires (Bonnefoy et al., 2009). Ainsi, l’air froid a tendance à stagner dans les bas fonds à proximité du Layon, ce qui peut poser des problèmes de gels tardifs en période de débourrement de la vigne. Les plateaux et hauts de coteaux sont moins exposés à ce risque de gel mais ils sont défavorisés par les températures maximales qui augmentent moins en journée du fait de l’altitude et de la pente moins marquée.

2. Des indices bioclimatiques contrastés

La forte variabilité spatiale des températures se répercutent sur celle des indices bioclimatiques. Une nouvelle fois, si ces indices sont à la base utilisée à une échelle régionale et peuvent montrer d’importantes disparités au sein d’une même région, comme dans le Val de Loire (Bonnefoy et al ; 2009), ils permettent à très fine échelle de faire ressortir les disparités climatiques liées aux caractéristiques du milieu.

Fig 3. Cumul des degrés jours dans les coteaux du Layon (à gauche)et indice basé sur l’indice de Huglin (à droite)

Les calculs du cumul des degrés jours et de l’indice de Huglin n’ont été effectués que sur 20 capteurs car deux ont des données manquantes. La Fig. 3 montre assez clairement que les cumuls de degrés jours les plus faibles sont observés en versant nord ou en fond de vallée et en haut de coteaux. En revanche, les cumuls les plus élevés se trouvent en général à mi-coteau et en versant sud. Ainsi, le cumul des degrés jours (dj) varient de 1184 dj à 1480 dj. Ces observations laissent suggérer que celles-ci auront des conséquences sur le déroulement du cycle phénologique de la vigne et notamment sur la date de maturité.

L’indice basé sur l’indice de Huglin montre également une assez forte variabilité au sein des vignobles (valeurs variant de 1890 à 2190). Les valeurs les plus faibles sont localisées en haut de coteau alors que les valeurs les plus fortes sont clairement en bas et à mi-coteau. Nous observons des indices relativement élevés en versant nord, contrairement aux degrés jours, du fait que le calcul de l’indice de Huglin donne un poids plus important aux températures maximales. En effet les températures maximales sont assez élevées sur ce versant du fait de son altitude moindre. Enfin, l’importance du type de sol est mise en évidence par les capteurs

3 - 113



Bellevue 1 (sol humide) et 2 (sol sec) qui ont respectivement un indice de 1912 et 1946 malgré leur espacement de quelques dizaines de mètres.

CONCLUSIONS

Cette étude montre très clairement la variabilité spatiale des températures à l’échelle des Coteaux du Layon. Les zones plus « chaudes » ont été mises en évidence par les indices bioclimatiques et correspondent en général aux mi-coteaux alors que les températures minimales descendent de façon plus conséquente en bas de coteau et près de la rivière, notamment lors de nuits à inversion thermique. Les plateaux quant à eux connaissent des températures maximales plus faibles donc des indices bioclimatiques moins élevés. Les amplitudes thermiques journalières sont plus élevées dans le bas des parcelles, près du Layon, ce qui est favorable au développement de la pourriture noble à l’origine des vins liquoreux produits dans ce secteur. L’indice de fraîcheur des nuits sera également étudié car celui-ci est très important pour la maturité. Enfin, prochainement des simulations climatiques avec le modèle RAMS seront réalisées et permettront d’évaluer les futurs effets du réchauffement climatique sur le climat actuel des coteaux du Layon.

BIBLIOGRAPHIE

Bonnefoy C., Quénol H., Planchon O., Barbeau G., 2010. Analyse régionale des températures dans le Val de Loire. Echogéo, sous presse

Bonnefoy C., Quénol H., Barbeau G., Madelin M., 2009. Analyse muti scalaire des températures dans le vignoble du Val de Loire. In Geographia Technica. Actes du XXIIème colloque de l’Association Internationale de Climatologie, Cluj (Roumanie), Numéro spécial, 85-90

Duchêne C., 1996. Le vignoble Angevin. Conseil Interprofessionnel des Vins d’Anjou et de Saumur, 11 p.

Huglin P., Schneider C., 1998. Biologie et écologie de la vigne. Paris, Lavoisier, 370 p.

INAO, A la découverte des vignobles des Coteaux du Layon, milieu naturel et facteurs de qualité, 25 p.

Jones G.V., Duchene E., Tomasi D., Yuste J., Braslavska O., Schultz H., Martinez C., Boso S., Langellier F., Perruchot C. et Guimberteau G., 2005. Changes in 53 european winegrape phenology and relationships with climate. GESCO 2005, Allemagne, Vol. 1, 55-62.

Madelin M., Chabin J.-P., Bonnefoy C., 2008. Global warming and its consequences on the Beaune vineyards,. Enometrica, Vol.1, 2, 9-19

Winkler A.J., Cook J.A., Kliewer W.M., Lider L.A., 1974. General viticulture. Berkeley, University of California, 710 p.

INFLUENZA DELL’ESPOSIZIONE DEL VIGNETO SULLAMATURAZIONE DELL’UVA

Influences of vineyard location on grapevine performances.

Guidoni S., Gangemi L., Ferrandino A.Dipartimento di Colture Arboree. Università di Torino. Via L. Da Vinci, 44. 10095 Grugliasco (TO), Italy

[email protected]

RIASSUNTOLo studio è stato condotto in vigneti commerciali di Vitis vinifera cv Nebbiolo localizzati inPiemonte, Italia del Nord-Ovest, intorno alla sommità di una collina. L'obiettivo dello studio èstato di determinare come l’esposizione del vigneto possa influenzare il comportamentovegetativo della vite, il manifestarsi delle fasi fenologiche, e la cinetica di maturazionedell’uva con particolare riguardo all’accumulo di antociani e flavonoli. Le esposizioni piùmeridionali hanno indotto precocità di germogliamento e fioritura ma diminuzione dellafertilità per gemma e, di conseguenza, della resa per pianta influenzando anche il peso deigrappoli, degli acini e delle bucce; hanno promosso una maggiore concentrazione dei solidisolubili nelle ultime fasi di maturazione ma la sintesi degli antociani e dei flavonoli ha subitoun rallentamento durante le fasi tardive di maturazione. L’esposizione occidentale ha favoritoil ritardo delle fasi fenologiche e un aumento della fertilità per gemma, del peso delgrappolo e della resa produttiva, determinando un minore accumulo di solidi solubili nel mostoma una maggiore sintesi di antociani. Si è evidenziata, in oltre, una probabile influenzadella temperatura non solo sulla sintesi degli antociani ma anche dei flavonoli delle bucce.

PAROLE CHIAVENebbiolo – fasi fenologiche – produttività – antociani – flavonoli

ABSTRACTThe study was conducted in Sinio (Piedmont, Northwest Italy) in commercial vineyards of

Vitis vinifera cv. Nebbiolo, situated on the top of a 30 % slope hillside, thus they weredifferently exposed: two of these (A) was exposed to South, another (B) to East-South-East,the fourth (C) to West-North-West. The clone CVT 141 grafted onto 420 A, was cultivated inevery vineyard. Vines were VSP trained and pruned to the Guyot system (10 bud cane plus 2bud spur). Vine theoretical density was 5200 vine/ha. The aim of this study was to determinehow the vineyard exposition influences vine vegetative behaviour, phenological phase timing,grape ripening kinetic and grape properties including colour and flavonols. The results wereused to characterize the vineyards in a sort of farm zoning, helping to choose the besttechnical management.

The 2009 vintage was characterized by a very rainy winter and spring, and a very hotsummer (from mid July until the beginning of September the maximum temperature, asaverage, exceeded 32 °C). Bud burst and flowering resulted delayed in C, respect to A and Bvineyards, whereas bud fertility was higher in C. That fact induced a higher bunch weight(313 g) in vineyard facing West (C), respect to those Southward (A and D) where bunchweight was similar (224 g) also thanks to a higher berry mass (1.87 g in A and D, 2.09 g in B,and 2.07 g in C). Furthermore, vineyard exposition influenced the vine vigour and yield thatin C and D were twice that in A and B vineyards. Soluble solid content at harvest appeared

3 - 114



Bellevue 1 (sol humide) et 2 (sol sec) qui ont respectivement un indice de 1912 et 1946 malgré leur espacement de quelques dizaines de mètres.

CONCLUSIONS

Cette étude montre très clairement la variabilité spatiale des températures à l’échelle des Coteaux du Layon. Les zones plus « chaudes » ont été mises en évidence par les indices bioclimatiques et correspondent en général aux mi-coteaux alors que les températures minimales descendent de façon plus conséquente en bas de coteau et près de la rivière, notamment lors de nuits à inversion thermique. Les plateaux quant à eux connaissent des températures maximales plus faibles donc des indices bioclimatiques moins élevés. Les amplitudes thermiques journalières sont plus élevées dans le bas des parcelles, près du Layon, ce qui est favorable au développement de la pourriture noble à l’origine des vins liquoreux produits dans ce secteur. L’indice de fraîcheur des nuits sera également étudié car celui-ci est très important pour la maturité. Enfin, prochainement des simulations climatiques avec le modèle RAMS seront réalisées et permettront d’évaluer les futurs effets du réchauffement climatique sur le climat actuel des coteaux du Layon.

BIBLIOGRAPHIE

Bonnefoy C., Quénol H., Planchon O., Barbeau G., 2010. Analyse régionale des températures dans le Val de Loire. Echogéo, sous presse

Bonnefoy C., Quénol H., Barbeau G., Madelin M., 2009. Analyse muti scalaire des températures dans le vignoble du Val de Loire. In Geographia Technica. Actes du XXIIème colloque de l’Association Internationale de Climatologie, Cluj (Roumanie), Numéro spécial, 85-90

Duchêne C., 1996. Le vignoble Angevin. Conseil Interprofessionnel des Vins d’Anjou et de Saumur, 11 p.

Huglin P., Schneider C., 1998. Biologie et écologie de la vigne. Paris, Lavoisier, 370 p.

INAO, A la découverte des vignobles des Coteaux du Layon, milieu naturel et facteurs de qualité, 25 p.

Jones G.V., Duchene E., Tomasi D., Yuste J., Braslavska O., Schultz H., Martinez C., Boso S., Langellier F., Perruchot C. et Guimberteau G., 2005. Changes in 53 european winegrape phenology and relationships with climate. GESCO 2005, Allemagne, Vol. 1, 55-62.

Madelin M., Chabin J.-P., Bonnefoy C., 2008. Global warming and its consequences on the Beaune vineyards,. Enometrica, Vol.1, 2, 9-19

Winkler A.J., Cook J.A., Kliewer W.M., Lider L.A., 1974. General viticulture. Berkeley, University of California, 710 p.

INFLUENZA DELL’ESPOSIZIONE DEL VIGNETO SULLAMATURAZIONE DELL’UVA

Influences of vineyard location on grapevine performances.

Guidoni S., Gangemi L., Ferrandino A.Dipartimento di Colture Arboree. Università di Torino. Via L. Da Vinci, 44. 10095 Grugliasco (TO), Italy

[email protected]

RIASSUNTOLo studio è stato condotto in vigneti commerciali di Vitis vinifera cv Nebbiolo localizzati inPiemonte, Italia del Nord-Ovest, intorno alla sommità di una collina. L'obiettivo dello studio èstato di determinare come l’esposizione del vigneto possa influenzare il comportamentovegetativo della vite, il manifestarsi delle fasi fenologiche, e la cinetica di maturazionedell’uva con particolare riguardo all’accumulo di antociani e flavonoli. Le esposizioni piùmeridionali hanno indotto precocità di germogliamento e fioritura ma diminuzione dellafertilità per gemma e, di conseguenza, della resa per pianta influenzando anche il peso deigrappoli, degli acini e delle bucce; hanno promosso una maggiore concentrazione dei solidisolubili nelle ultime fasi di maturazione ma la sintesi degli antociani e dei flavonoli ha subitoun rallentamento durante le fasi tardive di maturazione. L’esposizione occidentale ha favoritoil ritardo delle fasi fenologiche e un aumento della fertilità per gemma, del peso delgrappolo e della resa produttiva, determinando un minore accumulo di solidi solubili nel mostoma una maggiore sintesi di antociani. Si è evidenziata, in oltre, una probabile influenzadella temperatura non solo sulla sintesi degli antociani ma anche dei flavonoli delle bucce.

PAROLE CHIAVENebbiolo – fasi fenologiche – produttività – antociani – flavonoli

ABSTRACTThe study was conducted in Sinio (Piedmont, Northwest Italy) in commercial vineyards of

Vitis vinifera cv. Nebbiolo, situated on the top of a 30 % slope hillside, thus they weredifferently exposed: two of these (A) was exposed to South, another (B) to East-South-East,the fourth (C) to West-North-West. The clone CVT 141 grafted onto 420 A, was cultivated inevery vineyard. Vines were VSP trained and pruned to the Guyot system (10 bud cane plus 2bud spur). Vine theoretical density was 5200 vine/ha. The aim of this study was to determinehow the vineyard exposition influences vine vegetative behaviour, phenological phase timing,grape ripening kinetic and grape properties including colour and flavonols. The results wereused to characterize the vineyards in a sort of farm zoning, helping to choose the besttechnical management.

The 2009 vintage was characterized by a very rainy winter and spring, and a very hotsummer (from mid July until the beginning of September the maximum temperature, asaverage, exceeded 32 °C). Bud burst and flowering resulted delayed in C, respect to A and Bvineyards, whereas bud fertility was higher in C. That fact induced a higher bunch weight(313 g) in vineyard facing West (C), respect to those Southward (A and D) where bunchweight was similar (224 g) also thanks to a higher berry mass (1.87 g in A and D, 2.09 g in B,and 2.07 g in C). Furthermore, vineyard exposition influenced the vine vigour and yield thatin C and D were twice that in A and B vineyards. Soluble solid content at harvest appeared

3 - 115



higher in A, B and D (24.3 Brix as average) than in C vineyard (23.7 Brix). Southernexpositions (A and D) delayed the beginning of veraison and reduced the anthocyaninconcentration at harvest (600 mg/kg) respect to B (670 mg/kg) and C (770 mg/k); furtherdifferences among vineyards were observed both in the pattern of flavonol accumulation andin their concentration at harvest. In synthesis the Southern expositions advanced thephenological phases and decreased bud fertility, yield per vine and weight of bunches, berriesand berry skins. In addition, it promoted a high concentration of soluble solids at harvest butnot of anthocyanins whose concentration slowed down during the late phases of ripening.Western exposition (C) promoted a delay of phenological phases, and an increase of budfertility, bunch weight and yield per vine; it induced a medium accumulation of soluble solidsbut the highest synthesis of anthocyanins. Due to the global warming we can expect a highvariability between vintages from a weather point of view. We think that a sort of farm zoningmatched with data obtained from observations executed in successive vintages could be auseful help to choose the best technical management for a specific year and to foresee inadvance the vintage results.

KEY-WORDSNebbiolo – phenological phases – yield – anthocyanins – flavonols

INTRODUZIONEL’influenza dell’esposizione del vigneto sulla maturazione dell’uva è nota ed è la base sulla

quale si tende a identificare a priori la migliore combinazione esposizione/vitigno perpermettere una ottimale maturazione dell’uva. E’ altresì noto, anche se l’evidenza non èaffatto banale, che anche il diverso orientamento dei filari nel vigneto sottopone i frutti adiversi regimi microclimatici che, con il passare della stagione, possono fortementediversificare gli ambienti di maturazione identificabili con le due pareti del filare. E’ statoevidenziato, per esempio, che la temperatura e la luce che investono i grappoli possonovariare molto in funzione dell’esposizione delle facce del filare o del grado diombreggiamento in cui essi si trovano e che questo influenza in modo rilevante molti deiparametri che caratterizzano le piante e la qualità dell’uva prodotta, dai più macroscopicicome, per esempio, la produttività delle piante, ai più specifici, come il profilo antocianico oflavonolico delle uve (Berqvist et al., 2001; Deloire, Hunter, 2005; Tarara et al., 2008; Pereiraet al., 2009; Chorti et al., 2010). Il microclima della fascia fruttifera può essere condizionatosia dalle scelte pre impianto sia dagli interventi colturali che, quindi, sono potenziali strumentiper influenzare la maturazione delle uve (Keller, 2010). Non molti studi sono stati effettuatiper valutare l’influsso dell’esposizione del vigneto sui parametri microclimatici esull’accumulo dei metaboliti nell’uva ma, alla luce della forte variabilità climatica che hacaratterizzato le ultime stagioni, ciò riveste un interesse, non solo applicativo, perché non ècosì chiaramente prevedibile l’evoluzione dei parametri microclimatici, né quella dellamaturazione in contesti ambientali diversi da questo punto di vista. L'obiettivo dello studio èstato di determinare come l’esposizione del vigneto possa influenzare, il comportamentovegetativo della vite e la cinetica di accumulo di alcuni metaboliti nell’uva.

MATERIALI E METODILo studio è stato condotto in vigneti commerciali di Vitis vinifera cv Nebbiolo, localizzati in

Piemonte, Italia Nord-Ovest, intorno alla sommità di una collina (Fig. 1): il vigneto A e il Dsono esposti a Sud (S), il B a Est-Sud-Est (ESE), il C a Ovest-Nord-Ovest (ONO). I vigneti

A, B e C si collocano ad altitudine compresa fra 428 e 415 m con una pendenza intorno al 19%, il vigneto D è situato ad altitudine inferiore (fra 370 e 340 m) e la pendenza è intorno al25 %. Il vigneto A è stato piantato nel 2004, il B nel 2003, il C e il D nel 2001. In tutti gliappezzamenti è coltivato il clone Nebbiolo CVT 141 innestato su 420A, allevato acontrospalliera con potatura a Guyot (circa 12 gemme per pianta), come tipico per il vitignonella zona di osservazione, con una densità teorica di impianto di 5.200 viti/ha; i filari sonodisposti in traverso.

Durante la stagione vegetative sono state rilevate le fasi fenologiche, è stata stimata lasuperficie fogliare come indice del vigore delle piante ed è stata pesata la produzione allaraccolta. Nelle fasi finali della maturazione è stata analizzata la composizione del mosto (pH,acidità, zuccheri) e la composizione in polifenoli delle bucce (antociani e flavonoli).

Figura 1. Immagine dei vigneti in cui si sono effettuate le osservazioni (da Google Earth).

RISULTATI E DISCUSSIONEL’annata 2009 è stata caratterizzata da un inverno e da una primavera molto piovosi che

hanno consentito la costituzione di una buona riserva idrica nei suoli; nel corso dellamaturazione dell’uva si sono invece succeduti periodi piuttosto prolungati durante i quali letemperature massime hanno superato i 32 °C, circostanza piuttosto anomala per l’ambienteconsiderato.

Le fasi fenologiche di riferimento (germogliamento e invaiatura) si sono presentate piùtardivamente nel vigneto esposto più a ovest (C) e nel vigneto situato a minore altitudine (D)(tab. 1) con effetti positivi sulla fertilità del germoglio; in D si è inoltre rilevata una maggiorefertilità delle gemme basali che è stata, al contrario, molto bassa nel vigneto A, insieme al B ilpiù precoce, e una minore incidenza delle gemme cieche. I vigneti più tardivi hanno quindiraggiunto una maggiore produttività non solo per un maggiore numero di grappoli per piantama, nel vigneto C, anche per maggiori dimensioni dei grappoli. Il vigneto C ha manifestato,inoltre, anche una migliore uniformità di vigore e piante con pareti fogliari di maggioresviluppo (tab. 1). Poiché la produttività dei vigneti è stata piuttosto diversa il valoredell’Indice Vegeto Produttivo (IVP), calcolato rapportando la superficie fogliare allaproduzione per pianta, è risultato piuttosto variabile e, in ogni caso non coincidente con ivalori considerati ottimali (tra 1,5 e 2 m2/kg di uva). In generale, gli apparati fogliari sonorisultati abbastanza radi; in A e D la superficie fogliare è diminuita nel corso della stagione ilche ha certamente incrementato il grado di esposizione dei grappoli alla radiazione luminosa.

Alla raccolta il peso dell’acino è risultato più basso nei vigneti A e D, il contenuto in solidisolubili è risultato simile nei vigneti A, B e D (intorno a 24 Brix), mentre nel vigneto C il loroaccumulo, iniziato più tardi, non ha raggiunto alla raccolta i 24 Brix. L’acidità titolabile delmosto ha subito una brusca diminuzione nell’ultima settimana di maturazione particolarmentein A dove ha raggiunto, alla raccolta del 7 ottobre, valori inferiori a 6 g/L; in C, invece, si è

3 - 116



higher in A, B and D (24.3 Brix as average) than in C vineyard (23.7 Brix). Southernexpositions (A and D) delayed the beginning of veraison and reduced the anthocyaninconcentration at harvest (600 mg/kg) respect to B (670 mg/kg) and C (770 mg/k); furtherdifferences among vineyards were observed both in the pattern of flavonol accumulation andin their concentration at harvest. In synthesis the Southern expositions advanced thephenological phases and decreased bud fertility, yield per vine and weight of bunches, berriesand berry skins. In addition, it promoted a high concentration of soluble solids at harvest butnot of anthocyanins whose concentration slowed down during the late phases of ripening.Western exposition (C) promoted a delay of phenological phases, and an increase of budfertility, bunch weight and yield per vine; it induced a medium accumulation of soluble solidsbut the highest synthesis of anthocyanins. Due to the global warming we can expect a highvariability between vintages from a weather point of view. We think that a sort of farm zoningmatched with data obtained from observations executed in successive vintages could be auseful help to choose the best technical management for a specific year and to foresee inadvance the vintage results.

KEY-WORDSNebbiolo – phenological phases – yield – anthocyanins – flavonols

INTRODUZIONEL’influenza dell’esposizione del vigneto sulla maturazione dell’uva è nota ed è la base sulla

quale si tende a identificare a priori la migliore combinazione esposizione/vitigno perpermettere una ottimale maturazione dell’uva. E’ altresì noto, anche se l’evidenza non èaffatto banale, che anche il diverso orientamento dei filari nel vigneto sottopone i frutti adiversi regimi microclimatici che, con il passare della stagione, possono fortementediversificare gli ambienti di maturazione identificabili con le due pareti del filare. E’ statoevidenziato, per esempio, che la temperatura e la luce che investono i grappoli possonovariare molto in funzione dell’esposizione delle facce del filare o del grado diombreggiamento in cui essi si trovano e che questo influenza in modo rilevante molti deiparametri che caratterizzano le piante e la qualità dell’uva prodotta, dai più macroscopicicome, per esempio, la produttività delle piante, ai più specifici, come il profilo antocianico oflavonolico delle uve (Berqvist et al., 2001; Deloire, Hunter, 2005; Tarara et al., 2008; Pereiraet al., 2009; Chorti et al., 2010). Il microclima della fascia fruttifera può essere condizionatosia dalle scelte pre impianto sia dagli interventi colturali che, quindi, sono potenziali strumentiper influenzare la maturazione delle uve (Keller, 2010). Non molti studi sono stati effettuatiper valutare l’influsso dell’esposizione del vigneto sui parametri microclimatici esull’accumulo dei metaboliti nell’uva ma, alla luce della forte variabilità climatica che hacaratterizzato le ultime stagioni, ciò riveste un interesse, non solo applicativo, perché non ècosì chiaramente prevedibile l’evoluzione dei parametri microclimatici, né quella dellamaturazione in contesti ambientali diversi da questo punto di vista. L'obiettivo dello studio èstato di determinare come l’esposizione del vigneto possa influenzare, il comportamentovegetativo della vite e la cinetica di accumulo di alcuni metaboliti nell’uva.

MATERIALI E METODILo studio è stato condotto in vigneti commerciali di Vitis vinifera cv Nebbiolo, localizzati in

Piemonte, Italia Nord-Ovest, intorno alla sommità di una collina (Fig. 1): il vigneto A e il Dsono esposti a Sud (S), il B a Est-Sud-Est (ESE), il C a Ovest-Nord-Ovest (ONO). I vigneti

A, B e C si collocano ad altitudine compresa fra 428 e 415 m con una pendenza intorno al 19%, il vigneto D è situato ad altitudine inferiore (fra 370 e 340 m) e la pendenza è intorno al25 %. Il vigneto A è stato piantato nel 2004, il B nel 2003, il C e il D nel 2001. In tutti gliappezzamenti è coltivato il clone Nebbiolo CVT 141 innestato su 420A, allevato acontrospalliera con potatura a Guyot (circa 12 gemme per pianta), come tipico per il vitignonella zona di osservazione, con una densità teorica di impianto di 5.200 viti/ha; i filari sonodisposti in traverso.

Durante la stagione vegetative sono state rilevate le fasi fenologiche, è stata stimata lasuperficie fogliare come indice del vigore delle piante ed è stata pesata la produzione allaraccolta. Nelle fasi finali della maturazione è stata analizzata la composizione del mosto (pH,acidità, zuccheri) e la composizione in polifenoli delle bucce (antociani e flavonoli).

Figura 1. Immagine dei vigneti in cui si sono effettuate le osservazioni (da Google Earth).

RISULTATI E DISCUSSIONEL’annata 2009 è stata caratterizzata da un inverno e da una primavera molto piovosi che

hanno consentito la costituzione di una buona riserva idrica nei suoli; nel corso dellamaturazione dell’uva si sono invece succeduti periodi piuttosto prolungati durante i quali letemperature massime hanno superato i 32 °C, circostanza piuttosto anomala per l’ambienteconsiderato.

Le fasi fenologiche di riferimento (germogliamento e invaiatura) si sono presentate piùtardivamente nel vigneto esposto più a ovest (C) e nel vigneto situato a minore altitudine (D)(tab. 1) con effetti positivi sulla fertilità del germoglio; in D si è inoltre rilevata una maggiorefertilità delle gemme basali che è stata, al contrario, molto bassa nel vigneto A, insieme al B ilpiù precoce, e una minore incidenza delle gemme cieche. I vigneti più tardivi hanno quindiraggiunto una maggiore produttività non solo per un maggiore numero di grappoli per piantama, nel vigneto C, anche per maggiori dimensioni dei grappoli. Il vigneto C ha manifestato,inoltre, anche una migliore uniformità di vigore e piante con pareti fogliari di maggioresviluppo (tab. 1). Poiché la produttività dei vigneti è stata piuttosto diversa il valoredell’Indice Vegeto Produttivo (IVP), calcolato rapportando la superficie fogliare allaproduzione per pianta, è risultato piuttosto variabile e, in ogni caso non coincidente con ivalori considerati ottimali (tra 1,5 e 2 m2/kg di uva). In generale, gli apparati fogliari sonorisultati abbastanza radi; in A e D la superficie fogliare è diminuita nel corso della stagione ilche ha certamente incrementato il grado di esposizione dei grappoli alla radiazione luminosa.

Alla raccolta il peso dell’acino è risultato più basso nei vigneti A e D, il contenuto in solidisolubili è risultato simile nei vigneti A, B e D (intorno a 24 Brix), mentre nel vigneto C il loroaccumulo, iniziato più tardi, non ha raggiunto alla raccolta i 24 Brix. L’acidità titolabile delmosto ha subito una brusca diminuzione nell’ultima settimana di maturazione particolarmentein A dove ha raggiunto, alla raccolta del 7 ottobre, valori inferiori a 6 g/L; in C, invece, si è

3 - 117



ERRATA CORRIGE attestata su un livello un po’ superiore a quello degli altri vigneti. L’accumulo degli zuccheri si è

arrestato, in tutti i vigneti, alcuni giorni prima della raccolta (fig. 2). Tabella 1. Parametri fenologici e produttivi nei vigneti di Sinio (vendemmia 2009).

2009 S1 (A) ESE (B) ONO (C) S (D) stadio germogliamento al 14.04.09* 4.7 4.6 3.6 3.9 fertilità totale (grappoli/germoglio) 0.6 0.8 0.8 1.0 50 % invaiatura (data) 5 ago 8 ago 12 ago 12 ago superficie fogliare (m2/pianta) 2.65 b 3.66 a 3.14 a 2.70 b produzione stimata q/ha 66 c 84 bc 168 a 105 b grappoli/pianta 5.7 b 5.8 b 10.4 a 9.2 a indice di fittezza (luglio) 1.10 b 1.27 a 1.24 a 1.02 b peso grappolo alla raccolta (g) 225 c 274 b 313 a 224 c Indice Vegeto Produttivo (m2/kg) 2.2 a 2.1 a 1.1 b 1.3 b 1 = la sigla corrisponde all’esposizione dei vigneti (vedi capitolo Materiali e metodi) * = valori medi di una scala arbitraria: 3=09 BBCH (gemma schiusa); 4=11 BBCH (prima foglia distesa); 5=12

BBCH (due foglie distese).

A

1.5

2.0

2.5

10Sep 20Sep 30Sep

gpeso acino

5.06.07.08.09.0

10.0

10Sep 20Sep 30Sep

g/L

202122232425

Brix

acidità titolabile g/Lsolidi solubili Brix

500

600

700

800

900

10Sep 20Sep 30Sep

anto

cian

i mg/

kg

100

150

200

250

300

flavo

noli

mg/

kgantociani flavonoli

B

1.5

2.0

2.5

10Sep 20Sep 30Sep

g

peso acino

5.06.07.08.09.0

10.0

10Sep 20Sep 30Sep

g/L

202122232425

Brix


500

600

700

800

900

10Sep 20Sep 30Sep

anto

cian

i mg/

kg

100

150

200

250

300fla

vono

li m

g/kgantociani

flavonoli

C

1.5

2.0

2.5

10Sep 20Sep 30Sep

g

peso acino

5.06.07.08.09.0

10.0

10Sep 20Sep 30Sep

g/L

202122232425

Brix


500

600

700

800

900

10Sep 20Sep 30Sep

anto

cian

i mg/

kg

100

150

200

250

300

flavo

nolli

mg/

kgantocianiflavonoli

D

1.5

2.0

2.5

10Sep 20Sep 30Sep

g peso acino

5.06.07.08.09.0

10.0

10Sep 20Sep 30Sep

g/L

202122232425

Brix


450

550

650

750

850

10Sep 20Sep 30Sep

anto

cian

i mg/

kg

100

150

200

250

300

flavo

noli

mg/

kgantocianiflavonoli

Figura 2. Evoluzione dei parametri di maturazione nei quattro vigneti di Nebbiolo CVT 141/420A, a confronto nell’anno 2009.

Nei contesti meno vigorosi (A e D) sono stati raggiunti, alla raccolta, minori quantitativi di

antociani rispetto a B e C; il picco massimo di concentrazione è stato raggiunto intorno alla


3 - 118

attestata su un livello un po’ superiore a quello degli altri vigneti. L’accumulo degli zuccherisi è arrestato, in tutti i vigneti, alcuni giorni prima della raccolta (fig. 2).

Tabella 1. Parametri fenologici eproduttivi nei vigneti di Sinio(vendemmia 2009). 2009

stadio germogliamento al 14.04.09*fertilità totale (grappoli/germoglio) 50 %invaiatura (data)superficie fogliare (m2/pianta) produzionestimata q/ha grappoli/piantaindice di fittezza (luglio)peso grappolo alla raccolta (g) IndiceVegeto Produttivo (m2/kg)

1 = la sigla corrisponde all’esposizione dei vigneti (vedi capitolo Materiali e metodi)* = valori medi di una scala arbitraria: 3=09 BBCH (gemma schiusa); 4=11 BBCH (prima foglia distesa);

5=12 BBCH (due foglie distese).

Figura 2. Evoluzione dei parametri di maturazione nei quattro vigneti di Nebbiolo CVT141/420A, a confronto nell’anno 2009.

Nei contesti meno vigorosi (A e D) sono stati raggiunti, alla raccolta, minori quantitativi diantociani rispetto a B e C; il picco massimo di concentrazione è stato raggiunto intorno allametà di settembre senza che si siano registrati ulteriori incrementi nel periodo successivo.Negli altri due vigneti, B e C, accomunati da una superficie fogliare più elevata rispetto adA e D (tab. 1), la sintesi degli antociani è proseguita più a lungo consentendo ilraggiungimento di concentrazioni ben più elevate di quelle ottenute in questi ultimi (+ 30 %circa) (fig. 2). E’ interessante notare come l’andamento dell’accumulo dei flavonoli sia statopiuttosto simile a quello degli antociani, composti con i quali condividono in parte il percorsobiosintetico. E’ noto che la sintesi degli antociani risente in modo rilevante del regimetermico (Yamane et al., 2006) e quella dei flavonoli di quello luminoso (Pereira et al., 2006;Tarara et al., 2008), ma non è ancora stato chiarito se le alte temperature possano avere effettinegativi anche sulla sintesi dei flavonoli. L’esposizione più meridionale e il minor vigore delvigneto A, possono avere causato un incremento della temperatura dei grappoli che ha limitatola sintesi di tutti i composti fenolici analizzati compresi i flavonoli che potrebbero esserediminuiti nel corso

S1 (A) ESE (B) ONO (C) S (D)4.7 4.6 3.6 3.90.6 0.8 0.8 1.0

5 ago 8 ago 12 ago 12 ago2.65 b 3.66 a 3.14 a 2.70 b66 c 84 bc 168 a 105 b

5.7 b 5.8 b 10.4 a 9.2 a1.10 b 1.27 a 1.24 a 1.02 b225 c 274 b 313 a 224 c2.2 a 2.1 a 1.1 b 1.3 b

g/L

anto

ciani

mg/

kg

g

700

600

500

400

10.0

800

2.5

2.0

1.5

9.08.07.06.05.0

10Sep 20Sep 30Sep

peso acino (g)


antocianiflavonoli

252423222120

250

200

50

15

100

A

Stia

flavo

nolim

g/kg

g

g/L

anto

ciani

mg/

kg

10.01

700

400

800

600

500

2.5

2.0

9.0

7.06.0

1.5

8.0

5.0

10Sep 20Sep 30Sep


antocianiflavonoli

peso acino (g)

252423222120

250

200

50

150

100

Stia

B

flavo

nolim

g/kg

della maturazione proprio per sensibilità agli eccessi di temperatura. Nei vigneti B e C igrappoli hanno verosimilmente goduto di un microclima più fresco, non solo a causadell’esposizione del vigneto, ma anche, e probabilmente in modo prioritario, grazie ad unamaggiore fogliosità delle piante mantenutasi anche nelle ultime fasi di maturazione.

Nel caso specifico del Nebbiolo, vitigno il cui profilo antocianico è a preponderanza dipeonidina-3-glucoside e per il quale il maggiore limite è rappresentato dall’intensità delcolore non sempre sufficiente, è emerso che nel vigneto più produttivo e vigoroso, il C, si èraggiunto un livello zuccherino inferiore a quello degli altri vigneti (fig. 2) ma un livelloquantitativo delle sostanze coloranti decisamente più soddisfacente e con un rapportopeonidina-3-glucoside/malvidina-3-glucoside migliore rispetto a quello registrato per ilvigneto B, nel quale il livello complessivo degli antociani era risultato simile (fig. 3).

IAT pn mv acet pcum totali

Figura 3. Concentrazione in antociani totali (IAT), peonidina- (pn) e malvidina- (mv) 3-glucosidi e presenza proporzionale delle forme acilate (pcum = p-cumaril derivati; ac = acetil

derivati; totali = somma delle forma acilate) nelle bucce di Nebbiolo all’1 (in alto) e al 7ottobre 2009 (in basso) in funzione dei vigneti in osservazione (media ± errore standard, n=3).

Da notare, inoltre, che in B e in C si è riscontrata anche una minore presenza delle formeacilate degli antociani (fig. 4) che sono, al contrario, risultate maggiori in A e D; questorisultato contribuisce a confermare l’ipotesi che l’acilazione delle forme libere sia una dellerisposte della pianta alle alte temperature (Downey et al., 2004) e che nei vigneti A e D, per iquali non sono attualmente disponibili dati specifici di temperatura, siano effettivamente stateraggiunte temperature dei grappoli superiori a quelle raggiunte in B e C.

CONCLUSIONILa variabilità emersa fra i vigneti consente di confermare l’importanza dell’esposizione nel

condizionare, in particolare, la temperatura dell’ambiente vigneto. Poiché essa ha un notevoleeffetto sui processi metabolici che sottendono allo sviluppo delle piante e al metabolismo dei

900

800

mg/

kg 700

600

500

400

300

mg/

kg 200

100

0

20

15

% 10

5

0IAT pn mv acet pcum totali

A B C D

mg/

kg

900

700

600

500

800

mg/

kg

400

300

200

100

0

%

20

15

10

5

0

A B C

3 - 119



frutti risulta di interesse notevole attuare accorgimenti, quali per esempio, il controllo dellavigoria del vigneto, che possano limitarne oltre che l’entità anche la variabilità. Tale aspetto,non secondario, è da tenere in speciale considerazione soprattutto nelle stagioni e sui versanticon le esposizioni più calde, anche alla luce della notevole variabilità climatica osservatanegli ultimi anni che può parzialmente modificare il giudizio, formatosi su dati storici rilevatiin periodi di maggiore omogeneità climatica, relativo alla vocazionalità dei vigneti.

RINGRAZIAMENTIGli autori desiderano ringraziare l’Azienda Agricola G. D. Vajra di Barolo, nei cui vigneti

si è svolta la prova, per la disponibilità ed il supporto finanziario alla ricerca.

BIBLIOGRAFIABergqvist J., Dokoozlian N., Ebisuda N. 2001. Sunlight exposure and temperature effects on

berry growth and composition of Cabernet Sauvignon and Grenache in the Central SanJoaquim Valley of California. Am. J. Enol. Vitic., 52: 1-7.

Chorti E., Guidoni S., Ferrandino A., Novello V. 2010 Effects of different cluster sunlightexposure levels on ripening and anthocyanin accumulation in Nebbiolo grapes. Am J.Enol. Vitic., 61: 23-30.

Deloire A. and Hunter J.J. 2005. Microclimat des grappes et maturation du raisin. Le ProgrèsAgricole et Viticole, 122: 151-157.

Downey M.O., Harvey J.S., Robinson S.P. 2004. The effect of bunch shading on berrydevelopment and flavonoid accumulation in Shiraz grapes. Aust. J. of Grapes and WineResearcher 10, 55-73.

Keller M. 2010. Managing grapevines to optimise fruit development in a challengingenvironment: a climate change primer for viticulturist. Austr. J. Grape and Wine Research,16: 56-69.

Pereira G. E., Gaudillere J.-P., Pieri P., Hilbert G., Maucourt M., Deborde C., Moing A.,Rolin D. 2009. Microclimate Influence on Mineral and Metabolic Profiles of GrapeBerries. J. Agric. Food Chem., 6765-6775.

Tarara J.M., Lee J., Spayd S.E., Scagel C.F. 2008. Berry temperature and solar radiation alteracylation, proportion, and concentration of anthocyaninin Merlot grapes. Am. J. Enol.Vitic., 59: 235-247.

Yamane T., Jeong S.T., Yamamoto N.G., Koshita Y., Kobayashi S. 2006. Effects oftemperature on anthocyanin biosynthesis in grape berry skins. Am. J. Enol. Vitic., 57,1,54-59.

IL MONITORAGGIO METEOROLOGICO COME STRUMENTO PER LA GESTIONE DELLA VARIABILITÀ CLIMATICA IN

FRANCIACORTA Paolo Carnevali (1), Luigi Mariani (1), Osvaldo Failla (1), Lucio Brancadoro

(1), Monica Faccincani (2)

(1) Di.Pro.Ve., Università degli Studi di Milano

Via Celoria 2, Milano, Italia

[email protected] (2) Consorzio per la Tutela del Franciacorta

Via G. Verdi 53, Erbusco (BS), Italia

[email protected]

RIASSUNTO Nel 2007 è stata avviata una ricerca nell’areale di produzione del Franciacorta DOCG che ha

riguardato un ampio numero di vigneti di Chardonnay con riferimento ai quali sono stati

acquisite le serie storiche dal 2001 relative a (i) decorso delle epoche fenologiche, (ii) curve di

maturazione e (iii) dati prodotti dalla rete meteorologica consortile. Tali dati hanno permesso

di produrre un modello empirico agrofenologico relativo allo Chardonnay nell’areale

considerato e di calibrare e validare un modello meccanicistico di simulazione della

produttività primaria, chiamato SIM_PP.

PAROLE CHIAVE Chardonnay – Franciacorta – variabilità climatica – modelli di simulazione – accumulo

zuccherino

ABSTRACT In 2007 a research was started on an high number of vineyards in the Franciacorta AOC

area. From 2001 to 2009, phonological stages records and ripening kinetics data were

collected. Starting from phenological data, an empiric agrophenological model was build, in

order to estimate principal stages by using daily cumulated temperature. Furthermore,

ripening kinetics were compared to mechanicistic model simulations (SIM_PP, Mariani and

Maugeri, 2002). Starting from air daily temperatures, SIM_PP simulates the Net Primary

Production, allocation dynamics in sink organs and the sugars storage in berries, using a

mechanism based on transpiration and mass transport flux.

The comparison between real in-field situation and gathered simulations allowed to evaluate

mechanicistic and empirical models performance.

KEYWORDS Chardonnay – Franciacorta – climatic variability – models – sugar storage

INTRODUZIONE Molti studi sono stati incentrati sulle relazioni intercorrenti tra clima, fenologia della vite,

composizione delle bacche, quantità e qualità delle produzioni (Jones, Davis, 2000). Le sottili

differenze climatiche, che si hanno tra località anche molto prossime fra loro, determinano

degli effetti sulla fenologia e sulla composizione chimica delle uve, oltre a determinare ritardi

o anticipi di maturazione e differenze sostanziali nei parametri qualitativi alla vendemmia, che

si riflettono sulla qualità dei vini prodotti (Conradie et al., 2002).

Le fasi fenologiche avvengono come effetto diretto del clima (Jones, Davis, 2000) ed

influiscono sui processi legati alla crescita dei differenti organi delle piante, all’entità delle

3 - 120



frutti risulta di interesse notevole attuare accorgimenti, quali per esempio, il controllo dellavigoria del vigneto, che possano limitarne oltre che l’entità anche la variabilità. Tale aspetto,non secondario, è da tenere in speciale considerazione soprattutto nelle stagioni e sui versanticon le esposizioni più calde, anche alla luce della notevole variabilità climatica osservatanegli ultimi anni che può parzialmente modificare il giudizio, formatosi su dati storici rilevatiin periodi di maggiore omogeneità climatica, relativo alla vocazionalità dei vigneti.

RINGRAZIAMENTIGli autori desiderano ringraziare l’Azienda Agricola G. D. Vajra di Barolo, nei cui vigneti

si è svolta la prova, per la disponibilità ed il supporto finanziario alla ricerca.

BIBLIOGRAFIABergqvist J., Dokoozlian N., Ebisuda N. 2001. Sunlight exposure and temperature effects on

berry growth and composition of Cabernet Sauvignon and Grenache in the Central SanJoaquim Valley of California. Am. J. Enol. Vitic., 52: 1-7.

Chorti E., Guidoni S., Ferrandino A., Novello V. 2010 Effects of different cluster sunlightexposure levels on ripening and anthocyanin accumulation in Nebbiolo grapes. Am J.Enol. Vitic., 61: 23-30.

Deloire A. and Hunter J.J. 2005. Microclimat des grappes et maturation du raisin. Le ProgrèsAgricole et Viticole, 122: 151-157.

Downey M.O., Harvey J.S., Robinson S.P. 2004. The effect of bunch shading on berrydevelopment and flavonoid accumulation in Shiraz grapes. Aust. J. of Grapes and WineResearcher 10, 55-73.

Keller M. 2010. Managing grapevines to optimise fruit development in a challengingenvironment: a climate change primer for viticulturist. Austr. J. Grape and Wine Research,16: 56-69.

Pereira G. E., Gaudillere J.-P., Pieri P., Hilbert G., Maucourt M., Deborde C., Moing A.,Rolin D. 2009. Microclimate Influence on Mineral and Metabolic Profiles of GrapeBerries. J. Agric. Food Chem., 6765-6775.

Tarara J.M., Lee J., Spayd S.E., Scagel C.F. 2008. Berry temperature and solar radiation alteracylation, proportion, and concentration of anthocyaninin Merlot grapes. Am. J. Enol.Vitic., 59: 235-247.

Yamane T., Jeong S.T., Yamamoto N.G., Koshita Y., Kobayashi S. 2006. Effects oftemperature on anthocyanin biosynthesis in grape berry skins. Am. J. Enol. Vitic., 57,1,54-59.

IL MONITORAGGIO METEOROLOGICO COME STRUMENTO PER LA GESTIONE DELLA VARIABILITÀ CLIMATICA IN

FRANCIACORTA Paolo Carnevali (1), Luigi Mariani (1), Osvaldo Failla (1), Lucio Brancadoro

(1), Monica Faccincani (2)

(1) Di.Pro.Ve., Università degli Studi di Milano

Via Celoria 2, Milano, Italia

[email protected] (2) Consorzio per la Tutela del Franciacorta

Via G. Verdi 53, Erbusco (BS), Italia

[email protected]

RIASSUNTO Nel 2007 è stata avviata una ricerca nell’areale di produzione del Franciacorta DOCG che ha

riguardato un ampio numero di vigneti di Chardonnay con riferimento ai quali sono stati

acquisite le serie storiche dal 2001 relative a (i) decorso delle epoche fenologiche, (ii) curve di

maturazione e (iii) dati prodotti dalla rete meteorologica consortile. Tali dati hanno permesso

di produrre un modello empirico agrofenologico relativo allo Chardonnay nell’areale

considerato e di calibrare e validare un modello meccanicistico di simulazione della

produttività primaria, chiamato SIM_PP.

PAROLE CHIAVE Chardonnay – Franciacorta – variabilità climatica – modelli di simulazione – accumulo

zuccherino

ABSTRACT In 2007 a research was started on an high number of vineyards in the Franciacorta AOC

area. From 2001 to 2009, phonological stages records and ripening kinetics data were

collected. Starting from phenological data, an empiric agrophenological model was build, in

order to estimate principal stages by using daily cumulated temperature. Furthermore,

ripening kinetics were compared to mechanicistic model simulations (SIM_PP, Mariani and

Maugeri, 2002). Starting from air daily temperatures, SIM_PP simulates the Net Primary

Production, allocation dynamics in sink organs and the sugars storage in berries, using a

mechanism based on transpiration and mass transport flux.

The comparison between real in-field situation and gathered simulations allowed to evaluate

mechanicistic and empirical models performance.

KEYWORDS Chardonnay – Franciacorta – climatic variability – models – sugar storage

INTRODUZIONE Molti studi sono stati incentrati sulle relazioni intercorrenti tra clima, fenologia della vite,

composizione delle bacche, quantità e qualità delle produzioni (Jones, Davis, 2000). Le sottili

differenze climatiche, che si hanno tra località anche molto prossime fra loro, determinano

degli effetti sulla fenologia e sulla composizione chimica delle uve, oltre a determinare ritardi

o anticipi di maturazione e differenze sostanziali nei parametri qualitativi alla vendemmia, che

si riflettono sulla qualità dei vini prodotti (Conradie et al., 2002).

Le fasi fenologiche avvengono come effetto diretto del clima (Jones, Davis, 2000) ed

influiscono sui processi legati alla crescita dei differenti organi delle piante, all’entità delle

3 - 121



produzioni ed alle performance qualitative della coltura (Nendel, 2010): è facilmente

comprensibile come l’applicazione di modelli di crescita in Vitis vinifera L. sia una

condizione necessaria affinché possano essere applicati degli studi di simulazione dinamica

riguardanti la nutrizione idrica o l’influenza del clima sulle produzioni.

La composizione e la concentrazione dei composti chimici nelle bacche variano durante lo

sviluppo e sono influenzate da molti fattori, sia endogeni che esogeni (Coombe, 1992): tra

questi ultimi figurano i fattori ambientali e le pratiche agronomiche. Nonostante l’effetto della

disponibilità di assimilati e di acqua sullo sviluppo dei frutti non sia ancora stato

perfettamente chiarito, soprattutto per quanto concerne il bilancio idrico e carbonioso delle

bacche, il monitoraggio delle variabili meteorologiche risulta essere fondamentale per la

determinazione delle risposte produttive (Cola et al., 2009):

Nell’ambito dei modelli di produzione le variabili fisiche atmosferiche ricoprono un ruolo

chiave nel determinare il comportamento dei sistemi esaminati (Mariani e Failla, 2008): si ha

una riprova di ciò considerando, ad esempio, l’influenza della radiazione solare sulle risposte

fisiologiche della vite e quella del cosiddetto tempo termico (Scott, 2003) sul decorso

fenologico.

L’adozione di modelli empirici e meccanicistici su ampi comprensori vitati permette di

considerare tutte le componenti climatiche (macro, meso e microclima) ed offre la possibilità

di studiare alle diverse scale di interesse l’influenza delle variabili meteorologiche sul decorso

fenologico e maturativo della vite: a tal fine, nell’ambito del progetto SmaCH promosso dalla

Regione Lombardia e dal Consorzio per la Tutela del Franciacorta, è stato valutato

l’andamento delle fasi fenologiche e del decorso maturativo nell’areale franciacortino, sono

stati sviluppati modelli agrofenologici in grado di descrivere l’andamento del ciclo annuale

della vite in relazione al decorso meteorologico e, infine, sono stati utilizzati tali modelli

come ausilio nella determinazione dell’opportuno momento di vendemmia.

MATERIALI E METODI In prima istanza, al fine di predisporre una rete di rilievo agro-fenologico, è stato creato un

sistema capillare di stazioni meteorologiche consortili che, unitamente alla già esistente rete

provinciale, permette di ottenere informazioni relative alle principali variabili meteorologiche

su tutto il territorio della DOCG; nello specifico, sono state installate 4 nuove stazioni nei

comuni di Adro, Erbusco, Passirano e Paderno Franciacorta e, nelle preesistenti capannine di

Capriolo e Cortefranca, sono stati aggiunti ulteriori sensori di radiazione solare e di velocità e

direzione del vento.

Nel corso del primo anno di sperimentazione sono stati selezionati 20 vigneti di Chardonnay

tra quelli utilizzati dal Consorzio per la Tutela del Franciacorta nella determinazione delle

cinetiche di maturazione: i criteri con cui è stata effettuata la scelta dei vigneti sono stati la

rappresentatività all’interno delle 6 Unità Vocazionali in cui è suddivisa la Franciacorta e

l’uniforme distribuzione della parcelle vitate all’interno della DOCG. A cadenza settimanale

sono stati eseguiti rilievi fenologici, dal germogliamento all’invaiatura, e, a partire da

quest’ultima, sono stati raccolti campioni di uve destinati ad analisi dei principali parametri

tecnologici (zuccheri, pH, acidità titolabile, contenuti in acido malico e tartarico).

I rilievi fenologici sono stati eseguiti su 3 germogli di 10 piante rappresentative di ciascun

vigneto, determinando la fase raggiunta in base alla scala BBCH, mentre le misure delle

cinetiche di maturazione sono state effettuate mediante campionamenti randomizzati

all’interno delle parcelle individuate nei vigneti guida.

I dati fenologici sono stati utilizzati per la creazione di un modello empirico di simulazione

della fase raggiunta a partire dalle Ore Normali di Caldo (NHH): per fare ciò è stata utilizzata

una funzione matematica per la conversione della temperatura media oraria nell

corrispondente ora normale di caldo, come riportato in Fig.1.

Fig. 1: Grafico della funzione matematica utilizzata per convertire la temperatura media oraria in ore normali di

Inoltre, al fine di rendere lineare la scala BBCH, è stata utilizzata una scala di comodo, con

un range di valori da 0 (dormienza) a

I dati meteorologici disponibili dal 2006 sono stati inoltre applicati ad un modello

meccanicistico chiamato SIM_PP

Maugeri, 2002): è un modello di simulazione dinamica che sti

delle colture ad una scala di campo sulla base dei valori giornalieri di temperatura dell’aria

(Tx e Tn) e dai dati di acqua disponibile nel terreno.

Tale modello è suddiviso in diversi moduli: un preprocessore meteorologico, c

utilizzando come variabili input le precipitazioni giornaliere, la temperatura massima e la

minima genera i valori giornalieri di radiazione solare, umidità relativa minima e massima,

velocità del vento e copertura nuvolosa; un modulo fenologico;

che simula la produzione giornaliera in vite; il modulo “VACUOLO” che simula l’accumulo

zuccherino nelle bacche.

Fig. 2: Schema di funzionamento del nucleo principale del modello SIM_PP.

Partendo dalla radiazione globale stimata (Rglob) si ottiene una PAR (Radiazione

Fotosinteticamente Attiva) e da questa una Assimilazione Lorda (GASS); l’assimilazione

netta potenziale (PNA) è ottenuta considerando la GASS al netto delle perdite imputabili ai

processi respiratori, traslocativi e di conversione degli assimilati. La produzione primaria

netta (NPP) è calcolata applicando delle restrizioni dovute agli stress idrici e termici.


ora normale di caldo, come riportato in Fig.1.

Grafico della funzione matematica utilizzata per convertire la temperatura media oraria in ore normali di

caldo.


di valori da 0 (dormienza) a 49 (maturità tecnologica).


meccanicistico chiamato SIM_PP – SIMulazione della Produttività Primaria (Mariani,

Maugeri, 2002): è un modello di simulazione dinamica che stima le produzioni giornaliere






copertura nuvolosa; un modulo fenologico; il nucleo del modello (Fig.2),


Schema di funzionamento del nucleo principale del modello SIM_PP.

a radiazione globale stimata (Rglob) si ottiene una PAR (Radiazione



respiratori, traslocativi e di conversione degli assimilati. La produzione primaria


una funzione matematica per la conversione della temperatura media oraria nella




SIMulazione della Produttività Primaria (Mariani,

ma le produzioni giornaliere


Tale modello è suddiviso in diversi moduli: un preprocessore meteorologico, che



il nucleo del modello (Fig.2),







3 - 122



produzioni ed alle performance qualitative della coltura (Nendel, 2010): è facilmente

comprensibile come l’applicazione di modelli di crescita in Vitis vinifera L. sia una

condizione necessaria affinché possano essere applicati degli studi di simulazione dinamica

riguardanti la nutrizione idrica o l’influenza del clima sulle produzioni.

La composizione e la concentrazione dei composti chimici nelle bacche variano durante lo

sviluppo e sono influenzate da molti fattori, sia endogeni che esogeni (Coombe, 1992): tra

questi ultimi figurano i fattori ambientali e le pratiche agronomiche. Nonostante l’effetto della

disponibilità di assimilati e di acqua sullo sviluppo dei frutti non sia ancora stato

perfettamente chiarito, soprattutto per quanto concerne il bilancio idrico e carbonioso delle

bacche, il monitoraggio delle variabili meteorologiche risulta essere fondamentale per la

determinazione delle risposte produttive (Cola et al., 2009):

Nell’ambito dei modelli di produzione le variabili fisiche atmosferiche ricoprono un ruolo

chiave nel determinare il comportamento dei sistemi esaminati (Mariani e Failla, 2008): si ha

una riprova di ciò considerando, ad esempio, l’influenza della radiazione solare sulle risposte

fisiologiche della vite e quella del cosiddetto tempo termico (Scott, 2003) sul decorso

fenologico.

L’adozione di modelli empirici e meccanicistici su ampi comprensori vitati permette di

considerare tutte le componenti climatiche (macro, meso e microclima) ed offre la possibilità

di studiare alle diverse scale di interesse l’influenza delle variabili meteorologiche sul decorso

fenologico e maturativo della vite: a tal fine, nell’ambito del progetto SmaCH promosso dalla

Regione Lombardia e dal Consorzio per la Tutela del Franciacorta, è stato valutato

l’andamento delle fasi fenologiche e del decorso maturativo nell’areale franciacortino, sono

stati sviluppati modelli agrofenologici in grado di descrivere l’andamento del ciclo annuale

della vite in relazione al decorso meteorologico e, infine, sono stati utilizzati tali modelli

come ausilio nella determinazione dell’opportuno momento di vendemmia.

MATERIALI E METODI In prima istanza, al fine di predisporre una rete di rilievo agro-fenologico, è stato creato un

sistema capillare di stazioni meteorologiche consortili che, unitamente alla già esistente rete

provinciale, permette di ottenere informazioni relative alle principali variabili meteorologiche

su tutto il territorio della DOCG; nello specifico, sono state installate 4 nuove stazioni nei

comuni di Adro, Erbusco, Passirano e Paderno Franciacorta e, nelle preesistenti capannine di

Capriolo e Cortefranca, sono stati aggiunti ulteriori sensori di radiazione solare e di velocità e

direzione del vento.

Nel corso del primo anno di sperimentazione sono stati selezionati 20 vigneti di Chardonnay

tra quelli utilizzati dal Consorzio per la Tutela del Franciacorta nella determinazione delle

cinetiche di maturazione: i criteri con cui è stata effettuata la scelta dei vigneti sono stati la

rappresentatività all’interno delle 6 Unità Vocazionali in cui è suddivisa la Franciacorta e

l’uniforme distribuzione della parcelle vitate all’interno della DOCG. A cadenza settimanale

sono stati eseguiti rilievi fenologici, dal germogliamento all’invaiatura, e, a partire da

quest’ultima, sono stati raccolti campioni di uve destinati ad analisi dei principali parametri

tecnologici (zuccheri, pH, acidità titolabile, contenuti in acido malico e tartarico).

I rilievi fenologici sono stati eseguiti su 3 germogli di 10 piante rappresentative di ciascun

vigneto, determinando la fase raggiunta in base alla scala BBCH, mentre le misure delle

cinetiche di maturazione sono state effettuate mediante campionamenti randomizzati

all’interno delle parcelle individuate nei vigneti guida.

I dati fenologici sono stati utilizzati per la creazione di un modello empirico di simulazione

della fase raggiunta a partire dalle Ore Normali di Caldo (NHH): per fare ciò è stata utilizzata


corrispondente ora normale di caldo, come riportato in Fig.1.

Fig. 1: Grafico della funzione matematica utilizzata per convertire la temperatura media oraria in ore normali di


un range di valori da 0 (dormienza) a


meccanicistico chiamato SIM_PP

Maugeri, 2002): è un modello di simulazione dinamica che sti






velocità del vento e copertura nuvolosa; un modulo fenologico;


zuccherino nelle bacche.

Fig. 2: Schema di funzionamento del nucleo principale del modello SIM_PP.

Partendo dalla radiazione globale stimata (Rglob) si ottiene una PAR (Radiazione



processi respiratori, traslocativi e di conversione degli assimilati. La produzione primaria



ora normale di caldo, come riportato in Fig.1.


caldo.


di valori da 0 (dormienza) a 49 (maturità tecnologica).


meccanicistico chiamato SIM_PP – SIMulazione della Produttività Primaria (Mariani,

Maugeri, 2002): è un modello di simulazione dinamica che stima le produzioni giornaliere






copertura nuvolosa; un modulo fenologico; il nucleo del modello (Fig.2),


Schema di funzionamento del nucleo principale del modello SIM_PP.






una funzione matematica per la conversione della temperatura media oraria nella




SIMulazione della Produttività Primaria (Mariani,

ma le produzioni giornaliere


Tale modello è suddiviso in diversi moduli: un preprocessore meteorologico, che



il nucleo del modello (Fig.2),







3 - 123



Tale NPP viene allocata ai differenti organi della pianta, in funzione della fase fenologica in

cui ci si trova: a partire dall’invaiatura, una quota consistente è destinata ai grappoli, a livello

dei quali il modulo VACUOLO simula l’accumulo, basandosi su alcuni presupposti, quali: il

trasporto degli zuccheri nelle bacche avviene per via floematica (Coombe e McCarthy, 2000);

dall’invaiatura la bacca va incontro ad una prima fase di rammollimento dei tessuti ed una

seconda di espansione; il saccarosio entra nelle bacche per flusso di massa ad una

concentrazione di circa il 20% e viene compartimentato nei vacuoli; il movimento di acqua

nelle bacche è legato a meccanismi di traspirazione a livello della buccia (Coombe e

McCarthy, 2000).

Il modello è stato validato ricorrendo ad elaborati grafici e ad opportuni indici di fitting.

RISULTATI E DISCUSSIONE Il modello agrofenologico empirico è stato ricavato attraverso una regressione logaritmica

tra la fenofase individuata in ogni vigneto e le relative ore normali di caldo cumulate, ricavate

dalle temperature medie orarie delle stazioni corrispondenti (Fig. 3).

Fig. 3: Regressione logaritmica tra fenofase e NHH cumulate.

È stata così ricavata una funzione che è stata utilizzata nella stima della piena fioritura,

dell’invaiatura e della maturità tecnologica per il 2008, il 2009 e per la serie storica 1951-

2009. Ciò ha permesso di ricavare una serie di mappe fenologiche (in Fig. 4 si riportano le

mappe relative ai giorni di raggiungimento della maturazione tecnologica simulata per le due

annate e per la serie storica).

Si può notare come le due annate considerate abbiano registrato un anticipo di maturazione

di 10-15 giorni in linea con quanto riscontrato negli ultimi anni, avendo assistito al graduale

spostamento della data di vendemmia dalla prima metà di settembre alla seconda metà di

agosto.

y = 15,946Ln(x) - 80,119

R2 = 0,972

0

5

10

15

20

25

30

35

40

45

50

0 500 1000 1500 2000 2500 3000

NHH cumulate

BB

CH

in

sca

la d

i c

om

od

o

Fig. 4: Mappa dei giorni di raggiungimento della maturità tecnologica simulata (BBCH 85).

Dall’applicazione del modello SIM_PP ai dati meteorologici della Franciacorta è stato

ottenuto un confronto grafico fra le curve di accumulo zuccherino simulato e le cinetiche di

maturazione misurate nei corrispondenti vigneti (Fig. 5).

Fig. 5: Confronto fra zuccheri simulati dal modello SIM_PP e misurati in vigneto

Da tale preliminare analisi si può osservare come il modello descriva con sufficiente

accuratezza gli accumuli zuccherini, ad eccezione del 2008, anno p

riscontrata un’anomalia nelle analisi di laboratorio svolte sui campioni di mosto raccolti: per

tale motivo, nella successiva analisi, l’anno in questione non è stato considerato. Inoltre, si

può notare come nel 2007 SIM_PP tenda a sott

una sottostima iniziale ed una leggera sovrastima in conclusione di maturazione.

Infine sono state valutate le performance

comunemente utilizzati in fase di

squared error (RMSE), model efficiency

of determination (CD) ed R2.

In Tab. 1 sono riportati gli intervalli di ciascun indice utilizzato, i

relativi al modello SIM_PP per le annate 2006, 2007 e 2009Tab. 1: Risultati degli indici di fitting

Parametro MAE RMSE

Min 0.00 0.00

Max +∞ +∞

Best 0.00 0.00

Calcolato 1.68 13.55

: Mappa dei giorni di raggiungimento della maturità tecnologica simulata (BBCH 85).

all’applicazione del modello SIM_PP ai dati meteorologici della Franciacorta è stato



: Confronto fra zuccheri simulati dal modello SIM_PP e misurati in vigneto dal 2006 al 2009


accuratezza gli accumuli zuccherini, ad eccezione del 2008, anno per il quale è stata



può notare come nel 2007 SIM_PP tenda a sottostimare gli accumuli mentre nel 2009 vi sia


performance del modello in base ad una serie di indici di

in fase di validazione: mean absolute error (MAE), relative

model efficiency (EF), coefficient of residual mass (CRM)

In Tab. 1 sono riportati gli intervalli di ciascun indice utilizzato, i valori ottimali e quelli

relativi al modello SIM_PP per le annate 2006, 2007 e 2009 considerate complessivamentefitting per la validazione del modello SIM_PP. (sig. ***, per p

RMSE EF CRM CD R2 Sig. Media

osservata

Media

stimata

-∞ -∞ 0.00 0.00

1.00 +∞ +∞ 1.00

1.00 0.00 1.00 1.00

0.64 -0.04 1.40 0.67 *** 16.01 16.58




dal 2006 al 2009.


er il quale è stata



ostimare gli accumuli mentre nel 2009 vi sia

del modello in base ad una serie di indici di fitting

relative mean

(CRM), coefficient

valori ottimali e quelli

considerate complessivamente. ***, per p≤0.001).

Media

stimata

16.58

3 - 124



Tale NPP viene allocata ai differenti organi della pianta, in funzione della fase fenologica in

cui ci si trova: a partire dall’invaiatura, una quota consistente è destinata ai grappoli, a livello

dei quali il modulo VACUOLO simula l’accumulo, basandosi su alcuni presupposti, quali: il

trasporto degli zuccheri nelle bacche avviene per via floematica (Coombe e McCarthy, 2000);

dall’invaiatura la bacca va incontro ad una prima fase di rammollimento dei tessuti ed una

seconda di espansione; il saccarosio entra nelle bacche per flusso di massa ad una

concentrazione di circa il 20% e viene compartimentato nei vacuoli; il movimento di acqua

nelle bacche è legato a meccanismi di traspirazione a livello della buccia (Coombe e

McCarthy, 2000).

Il modello è stato validato ricorrendo ad elaborati grafici e ad opportuni indici di fitting.

RISULTATI E DISCUSSIONE Il modello agrofenologico empirico è stato ricavato attraverso una regressione logaritmica

tra la fenofase individuata in ogni vigneto e le relative ore normali di caldo cumulate, ricavate

dalle temperature medie orarie delle stazioni corrispondenti (Fig. 3).

Fig. 3: Regressione logaritmica tra fenofase e NHH cumulate.

È stata così ricavata una funzione che è stata utilizzata nella stima della piena fioritura,

dell’invaiatura e della maturità tecnologica per il 2008, il 2009 e per la serie storica 1951-

2009. Ciò ha permesso di ricavare una serie di mappe fenologiche (in Fig. 4 si riportano le

mappe relative ai giorni di raggiungimento della maturazione tecnologica simulata per le due

annate e per la serie storica).

Si può notare come le due annate considerate abbiano registrato un anticipo di maturazione

di 10-15 giorni in linea con quanto riscontrato negli ultimi anni, avendo assistito al graduale

spostamento della data di vendemmia dalla prima metà di settembre alla seconda metà di

agosto.

y = 15,946Ln(x) - 80,119

R2 = 0,972

0

5

10

15

20

25

30

35

40

45

50

0 500 1000 1500 2000 2500 3000

NHH cumulate

BB

CH

in

sca

la d

i c

om

od

o

Fig. 4: Mappa dei giorni di raggiungimento della maturità tecnologica simulata (BBCH 85).

Dall’applicazione del modello SIM_PP ai dati meteorologici della Franciacorta è stato



Fig. 5: Confronto fra zuccheri simulati dal modello SIM_PP e misurati in vigneto


accuratezza gli accumuli zuccherini, ad eccezione del 2008, anno p



può notare come nel 2007 SIM_PP tenda a sott


Infine sono state valutate le performance

comunemente utilizzati in fase di

squared error (RMSE), model efficiency

of determination (CD) ed R2.

In Tab. 1 sono riportati gli intervalli di ciascun indice utilizzato, i

relativi al modello SIM_PP per le annate 2006, 2007 e 2009Tab. 1: Risultati degli indici di fitting

Parametro MAE RMSE

Min 0.00 0.00

Max +∞ +∞

Best 0.00 0.00

Calcolato 1.68 13.55





: Confronto fra zuccheri simulati dal modello SIM_PP e misurati in vigneto dal 2006 al 2009


accuratezza gli accumuli zuccherini, ad eccezione del 2008, anno per il quale è stata



può notare come nel 2007 SIM_PP tenda a sottostimare gli accumuli mentre nel 2009 vi sia


performance del modello in base ad una serie di indici di

in fase di validazione: mean absolute error (MAE), relative

model efficiency (EF), coefficient of residual mass (CRM)

In Tab. 1 sono riportati gli intervalli di ciascun indice utilizzato, i valori ottimali e quelli

relativi al modello SIM_PP per le annate 2006, 2007 e 2009 considerate complessivamentefitting per la validazione del modello SIM_PP. (sig. ***, per p

RMSE EF CRM CD R2 Sig. Media

osservata Media

stimata

-∞ -∞ 0.00 0.00

1.00 +∞ +∞ 1.00

1.00 0.00 1.00 1.00

0.64 -0.04 1.40 0.67 *** 16.01 16.58




dal 2006 al 2009.


er il quale è stata



ostimare gli accumuli mentre nel 2009 vi sia

del modello in base ad una serie di indici di fitting

relative mean

(CRM), coefficient

valori ottimali e quelli

considerate complessivamente. ***, per p≤0.001).

Media

stimata

16.58

3 - 125



Tale analisi indica come le stime del modello siano dei descrittori leggermente migliori

rispetto alla media osservata (EF>0 e quasi prossimo al valore ottimale), pur notando una

lieve sovrastima (CRM leggermente negativo); ciononostante, si può indicare un generale

buon comportamento delle simulazioni (CD prossimo a 1 e buon valore di R2).

CONCLUSIONI Il monitoraggio agrometeorologico è un’attività fondamentale per la moderna viticoltura:

esso permette di valutare l’insorgenza di eventuali patologie, determinare i decorsi della

maturazione e pianificare le operazioni colturali.

Partendo da tali presupposti, una rete di rilievo agrometeorologico capillare sul territorio

risulta essere di primaria importanza sia per il monitoraggio meso e micrometeologico, sia per

i prodotti che da essa possono derivare: le tecniche modellistiche utilizzate nel corso di questo

progetto hanno infatti permesso di formulare dei modelli che, a partire dalle variabili

meteorologiche, forniscono delle simulazioni precoci del ciclo vegetativo e riproduttivo e del

decorso maturativo dello Chardonnay in Franciacorta.

I dati raccolti e simulati dai differenti modelli permettono di pianificare, in relazione alla

variabilità climatica della singola azienda e allo specifico decorso meteorologico stagionale,

la difesa antiparassitaria e la gestione delle operazioni colturali, come gli interventi al suolo,

le irrigazioni e le differenti pratiche di gestione della chioma.

Sempre in funzione dell’andamento meteorologico e della zona di appartenenza di ogni

singolo vigneto, è possibile prevedere con un certo anticipo e con una discreta precisione le

date di vendemmia ed il potenziale qualitativo delle uve.

BIBLIOGRAFIA Cola G. et al., 2009. BerryTone – A simulation model for the daily course of grape berry

temperature. Agricultural and Forest Meteorology, 149: 1215-1228.

Conradie W.J. et al, 2002. Effect of different environmental factors on the performance of

Sauvignon blanc grapevines in the Stellenbosch/Durbanville districts of South Africa. I.

Geology, soil, climate, phenology and grape composition. South African Journal of

Enology and Viticulture., 23 (2): 78-91.

Coombe B.G., 1992. Research on development and ripening of the grape berry. American

Journal of Enology and Viticulture, 43: 101-110.

Coombe B.G., McCarthy M.G., 2000. Dynamics of grape berry growth and physiology of

ripening. Australian Journal of Grape and Wine Research, 6:131-135.

Jones G.V., Davis R.E., 2000. Climate influences on grapevine phenology, grape


Enology and Viticulture, 51 (3): 249-261.

Mariani L., Failla O., 2008. Le grandezze meteo climatiche come variabili guida per gli

ecosistemi agricoli e forestali. Italian Journal of Agronony, 1: 9-16.

Mariani L., Maugeri M., 2002. Alcune considerazioni di tipo agro-climatico su serie storiche

della Sicilia orientale. In: Atti di AIAM 2002. Acireale. 84-95.

Nendel C., 2010. Grapevine bud brake prediction for cool climates. International Journal of

Biometeorology, 54: 231-241.

Scott P.R., 2003. Phase change and regulation of developmental timing in plants. Science,

301: 334-336.

Effect of vine nitrogen status on grape and wine quality:Terroir study in the Vaud vineyard (Switzerland)

J-S Reynard(1), V. Zufferey(1), F. Murisier(1)

(1)Agroscope Changins-Wädenswil ACW, CH-1260 NYON (Switzerland)Corresponding author: [email protected]

Abstract

This study was conducted on soil-climate-plant relations (terroir) and their impact on grapecomposition and wine quality in the canton of Vaud by Agroscope Changins-Wädenswil ACW. Anassessment of the vine nitrogen status on different terroirs was made by means of chlorophyll index,leaf nitrogen content and yeast assimilable nitrogen. Vine nitrogen status was observed to be highlyrelated to soil type. Vines on the soil type "bottom moraines" showed lower vigour, smaller berriesand a lower nitrogen status. Sensory analysis discriminated wines from different soil types. Vinenitrogen status through yeast assimilable nitrogen turned out to be strongly correlated with winepositive sensory descriptors and negatively correlated to wine astringency. In our study, the mainenvironmental factors influencing vine development and wine quality was the soil type via its effecton vine nitrogen level. Our results confirm the role on nitrogen supply in grape and wine quality andunderline nitrogen as a key factor in understanding the terroir effect.

Key-words: Soil component of terroir – vine nitrogen status – ecophysiology – grape and wine quality

Introduction

Nitrogen is one of the mineral element that grape requires in greatest amount. Therefore, it is oneof the most influent mineral nutrients on the physiology of the grapevine. Vine development, yield,fruit composition and wine quality are highly related to the vine nitrogen status.N fertilisation increases vigour, resulting in modification of canopy microclimate (Smart et al., 1991)and a higher sensibility to Botrytis cinerea. Spayd et al. (1994) reported that grapes with an elevatedlevel of N showed delayed fruit ripening and produce wine with a higher pH. Vine nitrogen status hasbeen showed to affect tannin and anthocyans content from grapes (Tregoa et al., 2002). Nitrogennutrition does impact the amount of aromatic compounds or precursors in grapes (Peyrot des Gachonset al., 2005).Most studies reported in the literature on nitrogen have been conducted as fertilisation trials, wheredifferent N supplies (often large amounts) where given to vines. Indeed, the present work aims tostudy the impact of environmental factors itself (mainly climate and soil) on vine. In viticulture thosenatural factors are known under the concept of "terroir". We are investigating the most pertinentparameters of terroir which are responsible for grape and wine quality. A better understanding of theterroir effect will help vine-growers in their vineyard management (irrigation, fertilisation, rootstocksand grape cultivars). Among those parameters, this article will focus on vine nitrogen status affectedby soil variation. The present results are part of a wider study on the various "terroirs" in the Cantonof Vaud and how they affect grape and wine quality. This wider project is conducted on 130 locationsdistributed on the entire Vaud vineyard (3800 ha) and endeavours to evaluate the adaptation of 10grape cultivars to the different region's terroirs.

3 - 126



Tale analisi indica come le stime del modello siano dei descrittori leggermente migliori

rispetto alla media osservata (EF>0 e quasi prossimo al valore ottimale), pur notando una

lieve sovrastima (CRM leggermente negativo); ciononostante, si può indicare un generale

buon comportamento delle simulazioni (CD prossimo a 1 e buon valore di R2).

CONCLUSIONI Il monitoraggio agrometeorologico è un’attività fondamentale per la moderna viticoltura:

esso permette di valutare l’insorgenza di eventuali patologie, determinare i decorsi della

maturazione e pianificare le operazioni colturali.

Partendo da tali presupposti, una rete di rilievo agrometeorologico capillare sul territorio

risulta essere di primaria importanza sia per il monitoraggio meso e micrometeologico, sia per

i prodotti che da essa possono derivare: le tecniche modellistiche utilizzate nel corso di questo

progetto hanno infatti permesso di formulare dei modelli che, a partire dalle variabili

meteorologiche, forniscono delle simulazioni precoci del ciclo vegetativo e riproduttivo e del

decorso maturativo dello Chardonnay in Franciacorta.

I dati raccolti e simulati dai differenti modelli permettono di pianificare, in relazione alla

variabilità climatica della singola azienda e allo specifico decorso meteorologico stagionale,

la difesa antiparassitaria e la gestione delle operazioni colturali, come gli interventi al suolo,

le irrigazioni e le differenti pratiche di gestione della chioma.

Sempre in funzione dell’andamento meteorologico e della zona di appartenenza di ogni

singolo vigneto, è possibile prevedere con un certo anticipo e con una discreta precisione le

date di vendemmia ed il potenziale qualitativo delle uve.

BIBLIOGRAFIA Cola G. et al., 2009. BerryTone – A simulation model for the daily course of grape berry

temperature. Agricultural and Forest Meteorology, 149: 1215-1228.

Conradie W.J. et al, 2002. Effect of different environmental factors on the performance of

Sauvignon blanc grapevines in the Stellenbosch/Durbanville districts of South Africa. I.

Geology, soil, climate, phenology and grape composition. South African Journal of

Enology and Viticulture., 23 (2): 78-91.

Coombe B.G., 1992. Research on development and ripening of the grape berry. American

Journal of Enology and Viticulture, 43: 101-110.

Coombe B.G., McCarthy M.G., 2000. Dynamics of grape berry growth and physiology of

ripening. Australian Journal of Grape and Wine Research, 6:131-135.

Jones G.V., Davis R.E., 2000. Climate influences on grapevine phenology, grape


Enology and Viticulture, 51 (3): 249-261.

Mariani L., Failla O., 2008. Le grandezze meteo climatiche come variabili guida per gli

ecosistemi agricoli e forestali. Italian Journal of Agronony, 1: 9-16.

Mariani L., Maugeri M., 2002. Alcune considerazioni di tipo agro-climatico su serie storiche

della Sicilia orientale. In: Atti di AIAM 2002. Acireale. 84-95.

Nendel C., 2010. Grapevine bud brake prediction for cool climates. International Journal of

Biometeorology, 54: 231-241.

Scott P.R., 2003. Phase change and regulation of developmental timing in plants. Science,

301: 334-336.

Effect of vine nitrogen status on grape and wine quality:Terroir study in the Vaud vineyard (Switzerland)

J-S Reynard(1), V. Zufferey(1), F. Murisier(1)

(1)Agroscope Changins-Wädenswil ACW, CH-1260 NYON (Switzerland)Corresponding author: [email protected]

Abstract

This study was conducted on soil-climate-plant relations (terroir) and their impact on grapecomposition and wine quality in the canton of Vaud by Agroscope Changins-Wädenswil ACW. Anassessment of the vine nitrogen status on different terroirs was made by means of chlorophyll index,leaf nitrogen content and yeast assimilable nitrogen. Vine nitrogen status was observed to be highlyrelated to soil type. Vines on the soil type "bottom moraines" showed lower vigour, smaller berriesand a lower nitrogen status. Sensory analysis discriminated wines from different soil types. Vinenitrogen status through yeast assimilable nitrogen turned out to be strongly correlated with winepositive sensory descriptors and negatively correlated to wine astringency. In our study, the mainenvironmental factors influencing vine development and wine quality was the soil type via its effecton vine nitrogen level. Our results confirm the role on nitrogen supply in grape and wine quality andunderline nitrogen as a key factor in understanding the terroir effect.

Key-words: Soil component of terroir – vine nitrogen status – ecophysiology – grape and wine quality

Introduction

Nitrogen is one of the mineral element that grape requires in greatest amount. Therefore, it is oneof the most influent mineral nutrients on the physiology of the grapevine. Vine development, yield,fruit composition and wine quality are highly related to the vine nitrogen status.N fertilisation increases vigour, resulting in modification of canopy microclimate (Smart et al., 1991)and a higher sensibility to Botrytis cinerea. Spayd et al. (1994) reported that grapes with an elevatedlevel of N showed delayed fruit ripening and produce wine with a higher pH. Vine nitrogen status hasbeen showed to affect tannin and anthocyans content from grapes (Tregoa et al., 2002). Nitrogennutrition does impact the amount of aromatic compounds or precursors in grapes (Peyrot des Gachonset al., 2005).Most studies reported in the literature on nitrogen have been conducted as fertilisation trials, wheredifferent N supplies (often large amounts) where given to vines. Indeed, the present work aims tostudy the impact of environmental factors itself (mainly climate and soil) on vine. In viticulture thosenatural factors are known under the concept of "terroir". We are investigating the most pertinentparameters of terroir which are responsible for grape and wine quality. A better understanding of theterroir effect will help vine-growers in their vineyard management (irrigation, fertilisation, rootstocksand grape cultivars). Among those parameters, this article will focus on vine nitrogen status affectedby soil variation. The present results are part of a wider study on the various "terroirs" in the Cantonof Vaud and how they affect grape and wine quality. This wider project is conducted on 130 locationsdistributed on the entire Vaud vineyard (3800 ha) and endeavours to evaluate the adaptation of 10grape cultivars to the different region's terroirs.

3 - 127



Materials and Methods

1. Experimental sitesA network consisting of 12 locations was set up in a viticultural region of Vaud named "La Côte".Locations have been chosen in order to represent the pedo-climatic conditions from the region. Thestudy sites were planted with Vitis Vinifera L. cv. Doral (Chasselas X Chardonnay) in 2003, thereforestudied vines have the same age all over the plots. Each site had a surface of circa 200 m 2 and waslocated in the middle of a wider commercial vineyard.All vines were grafted onto 3309C rootstock and trained in espalier (single Guyot with vertical shootpositioned foliage). The average plantation density was 6'400 vines/ha.The nonirrigated plots were cultivated by the winegrowers. Yield was limited by clusters thinning atberries pea-sized. Average yield was 10 t/ha. The leaf area/fruit weight ratio was over 1 m 2/kg.Agronomical practices (fertilization and pest control) were similar between locations. Floormanagement was permanent sod each two rows with herbicided strips under the vines.

2. Soil and climate characteristicsa) Soil types

An earlier soil study of the viticultural areas of Vaud has enabled to classify the different soilsresulting in a map of pedological units. The studied locations are situated on representative geologicalunits of the region. The majority of soils in the study area were made up of alpine moraines, aheterogeneous mixture of unevenly sized debris transported by the Rhone glacier during the lastglaciation (100'000-10'000 years BC). Moraines can be grouped into three types (Letessier et al.,2004): lateral moraines, bottom moraines and retreating moraines. Lateral moraines were formed atthe glacier's lateral edges. Bottom moraines originate from materials which were located underneaththe glacier and underlay great pressure. Soils which belong to that category show a high compactness.Retreating moraines were formed during the melting phase of the glacier. The main characteristics ofthe different moraines are summed up in Table 1. Some colluvial deposits found at the foot of slopesoriginate from progressive erosion of the dominating slopes. They contain few coarse elements (0-20%). We regrouped the different soil types in two categories, namely bottom moraines (7 locations)and other soil types (5 locations). This decision was motivated by the fact that the category "other soiltypes" did present a consistent physiological behaviour (see Results).

Soil typeCoarse

elementsCompactness

Claycontent

activelimestone

Lateral moraine 30-60 % porous 10-18% 4-7 %Bottom moraine 10-30 % very high 12-25 % 7-12%

Retreating moraine 60-90 % porous 5-10 % 2-7 %

Table 1 Basic properties of moraine soils (Letessier et al., 2004)

b) Climatic conditions of years 2007 and 2008Climatic parameters were supposed to be homogenous on our 12 locations, since they were situatedno more than 60 km away from each other. The year 2007 was characterised by very heavy rainfallbetween Mai to August. Summer months were cooler in average. April of the same year had very fewprecipitation events and average temperature was high. The 2008 vintage corresponded to an average

vintage in the region. Frequent and abundant rainfalls were recorded in the late growing season(September to October).

3. Experimental measurementsa) Vine nitrogen status

The vine nitrogen status was assessed using three different indicators: foliar analysis (at veraison, %of dry matter), chlorophyll index (N-Tester, Yara, France) and yeast assimilable nitrogen (Aerny,1996).b) Berry composition

Fruit maturation was monitored on every site each 2 weeks from veraison to harvest. The mainanalytical parameters (sugars, acids and yeast assimilable nitrogen) were measured on 200 berriessamples using the WineScan® (FOSS NIRSystems, USA).c) Miocrovinification

Microvinification was conducted identically for all lots by the same winemaker. For each terroir 150kg of grape were harvested within one week. 0.3 gr/L of diammonium phosphate (DAP) was added toeach must to aid the completion of the fermentation.

Results

1. Water statusFor both years, the average predawn water potential measured on various sites never went lower than0.3 MPa, which corresponds to the critical value for low water stress. Furthermore, analyses of carbonisotope discrimination (Gaudillère et al., 2002) on musts gave values ranging from -27.7‰ to -25.2‰.These results show that water supply was sufficient for all the sites during the 2 vintages and thereforewater availability didn't interact with fruit quality on those two vintages.

2. Nitrogen statusThe chlorophyll content was monitored through 2 growing seasons ( Figure 1). During the 2008vintage, chlorophyll content was increasing steadily up until 2 months after flowering. On the otherhand, in 2007, it remained nearly constant after flowering, and then began, nearly 3 months afterflowering, to fall down sharply. On both vintages, vines on bottom moraines did present, in average, alower content of chlorophyll compare to other soil types.Leaf diagnostic was carried out at veraison stage to determine possible mineral deficiency. Leafnitrogen content at veraison was significantly lower for vines established on bottom moraines in 2007and 2008 (Figure 1). Leaf nitrogen content was superior during the 2008 vintage whatever the soiltype. No other noticeable differences were observed for the other minerals.Yeast assimilable nitrogen (YAN) in musts was monitored from veraison to harvest ( Figure 2). Vinesgrowing on bottom moraines had low nitrogen content in musts. Difference between the two soilcategories was more noticeable in 2007, where nitrogen status of vines on bottom moraines wassystematically under the other ones. On the other hand, during the 2008 vintage, the difference wasless obvious, 3 measures out of 5 turned to be significantly different. YAN tended to be higher in2008 compare to 2007.

3 - 128



Materials and Methods

1. Experimental sitesA network consisting of 12 locations was set up in a viticultural region of Vaud named "La Côte".Locations have been chosen in order to represent the pedo-climatic conditions from the region. Thestudy sites were planted with Vitis Vinifera L. cv. Doral (Chasselas X Chardonnay) in 2003, thereforestudied vines have the same age all over the plots. Each site had a surface of circa 200 m 2 and waslocated in the middle of a wider commercial vineyard.All vines were grafted onto 3309C rootstock and trained in espalier (single Guyot with vertical shootpositioned foliage). The average plantation density was 6'400 vines/ha.The nonirrigated plots were cultivated by the winegrowers. Yield was limited by clusters thinning atberries pea-sized. Average yield was 10 t/ha. The leaf area/fruit weight ratio was over 1 m 2/kg.Agronomical practices (fertilization and pest control) were similar between locations. Floormanagement was permanent sod each two rows with herbicided strips under the vines.

2. Soil and climate characteristicsa) Soil types

An earlier soil study of the viticultural areas of Vaud has enabled to classify the different soilsresulting in a map of pedological units. The studied locations are situated on representative geologicalunits of the region. The majority of soils in the study area were made up of alpine moraines, aheterogeneous mixture of unevenly sized debris transported by the Rhone glacier during the lastglaciation (100'000-10'000 years BC). Moraines can be grouped into three types (Letessier et al.,2004): lateral moraines, bottom moraines and retreating moraines. Lateral moraines were formed atthe glacier's lateral edges. Bottom moraines originate from materials which were located underneaththe glacier and underlay great pressure. Soils which belong to that category show a high compactness.Retreating moraines were formed during the melting phase of the glacier. The main characteristics ofthe different moraines are summed up in Table 1. Some colluvial deposits found at the foot of slopesoriginate from progressive erosion of the dominating slopes. They contain few coarse elements (0-20%). We regrouped the different soil types in two categories, namely bottom moraines (7 locations)and other soil types (5 locations). This decision was motivated by the fact that the category "other soiltypes" did present a consistent physiological behaviour (see Results).

Soil typeCoarse

elementsCompactness

Claycontent

activelimestone

Lateral moraine 30-60 % porous 10-18% 4-7 %Bottom moraine 10-30 % very high 12-25 % 7-12%

Retreating moraine 60-90 % porous 5-10 % 2-7 %

Table 1 Basic properties of moraine soils (Letessier et al., 2004)

b) Climatic conditions of years 2007 and 2008Climatic parameters were supposed to be homogenous on our 12 locations, since they were situatedno more than 60 km away from each other. The year 2007 was characterised by very heavy rainfallbetween Mai to August. Summer months were cooler in average. April of the same year had very fewprecipitation events and average temperature was high. The 2008 vintage corresponded to an average

vintage in the region. Frequent and abundant rainfalls were recorded in the late growing season(September to October).

3. Experimental measurementsa) Vine nitrogen status

The vine nitrogen status was assessed using three different indicators: foliar analysis (at veraison, %of dry matter), chlorophyll index (N-Tester, Yara, France) and yeast assimilable nitrogen (Aerny,1996).b) Berry composition

Fruit maturation was monitored on every site each 2 weeks from veraison to harvest. The mainanalytical parameters (sugars, acids and yeast assimilable nitrogen) were measured on 200 berriessamples using the WineScan® (FOSS NIRSystems, USA).c) Miocrovinification

Microvinification was conducted identically for all lots by the same winemaker. For each terroir 150kg of grape were harvested within one week. 0.3 gr/L of diammonium phosphate (DAP) was added toeach must to aid the completion of the fermentation.

Results

1. Water statusFor both years, the average predawn water potential measured on various sites never went lower than0.3 MPa, which corresponds to the critical value for low water stress. Furthermore, analyses of carbonisotope discrimination (Gaudillère et al., 2002) on musts gave values ranging from -27.7‰ to -25.2‰.These results show that water supply was sufficient for all the sites during the 2 vintages and thereforewater availability didn't interact with fruit quality on those two vintages.

2. Nitrogen statusThe chlorophyll content was monitored through 2 growing seasons ( Figure 1). During the 2008vintage, chlorophyll content was increasing steadily up until 2 months after flowering. On the otherhand, in 2007, it remained nearly constant after flowering, and then began, nearly 3 months afterflowering, to fall down sharply. On both vintages, vines on bottom moraines did present, in average, alower content of chlorophyll compare to other soil types.Leaf diagnostic was carried out at veraison stage to determine possible mineral deficiency. Leafnitrogen content at veraison was significantly lower for vines established on bottom moraines in 2007and 2008 (Figure 1). Leaf nitrogen content was superior during the 2008 vintage whatever the soiltype. No other noticeable differences were observed for the other minerals.Yeast assimilable nitrogen (YAN) in musts was monitored from veraison to harvest ( Figure 2). Vinesgrowing on bottom moraines had low nitrogen content in musts. Difference between the two soilcategories was more noticeable in 2007, where nitrogen status of vines on bottom moraines wassystematically under the other ones. On the other hand, during the 2008 vintage, the difference wasless obvious, 3 measures out of 5 turned to be significantly different. YAN tended to be higher in2008 compare to 2007.

3 - 129



Figure 1 Left: Seasonal evolution of the SPAD index from two vintages 2007 and2008 and two categories of soil types. Vertical bars indicate standard errorof means. B07, B08 are the bloom time for each vintage. Right: Nitrogencontent of leaves at veraison expressed as % of dry matter. Means withdifferent letters are significantly different (Duncan's test, P<0.01).

Figure 2 Seasonal evolution of yeast assimilable nitrogen (YAN) in grapes asaffected by soil types during the 2007 and 2008) vintage. The last datapoints correspond to harvest. Vertical bars indicate standard error ofmeans. For the same date, means foiowed by the same letter are notsignificantly different at P<5% according to Duncan test.

A principal component analysis (PCA) was run on the several physiological variables measured oneach location (Figure 3). Those variables include indicators of vine vigour, vine mineral status,phenology and must composition. Data about phenology are depicted only for the 2008 vintage, sinceno data were recorded in 2007. PCA provides a visualization of the relation between the physiologicalvariables (Figure 3A and 3C) and a two-dimensional map of the locations (Figure 3B and 3D). Thefirst dimension is linked to nitrogen parameter, vine vigour and must acidity. Those variables arecorrelated. Yeast assimilable nitrogen was well correlated with vine vigour indicators (weight ofpruned branches: r = 0.6 significant at α=0.05, berries weight: r = 0.82 significant at α=0.001).

3 - 130



Figu

re3

Rep

rese

ntat

ion

ofgr

ape

phys

iolo

g yva

riabl

es(A

,C

)by

usin

gPC

Aap

proa

chon

two

vint

ages

(200

7:A

,B

;20

08:

C,D

).O

nFi

gure

5Aan

d5B

,as

teris

ksin

dica

teva

riabl

esw

hich

are

sign

ifica

n tlydi

ffer

ent

betw

een

the

2so

ilca

tego

ries

(Dun

can'

ste

stat

5%(*

),1%

(**)

and

1‰(*

**))

.Loc

atio

nsar

ere

pres

ente

d(C

,D)o

nth

efir

st2

dim

ensi

ons

ofPC

Aan

dco

lore

dac

cord

ing

toth

eir

soil

type

s.Lo

catio

nco

dest

ands

besi

deea

chda

tapo

int.

Bar

ycen

tres

ofth

etw

oso

ilca

tego

ries

are

depi

cted

with

thei

rcon

fiden

cein

terv

alat

95%

.M

_var

iabl

esar

em

easu

red

onm

ust

atha

rves

t,W

_br

anch

es=

wei

ght

ofpr

uned

bran

ches

,L_v

aria

bles

repr

esen

tle

afm

iner

als

cont

ent

atve

rais

on,M

_YA

N=y

east

assi

mila

ble

nitro

gen

inm

ust,

SPA

D=s

easo

n-av

erag

ech

loro

phyl

lcon

tent

,Bud

brea

k=da

te(d

ayof

year

)of

bud

brea

k,B

loom

=da

te(d

ayof

year

)of

bloo

m,G

row

thce

ssat

ion

=da

te(d

ayof

year

)ofs

hoot

grow

thce

ssat

ion.

Figure 1 Left: Seasonal evolution of the SPAD index from two vintages 2007 and2008 and two categories of soil types. Vertical bars indicate standard errorof means. B07, B08 are the bloom time for each vintage. Right: Nitrogencontent of leaves at veraison expressed as % of dry matter. Means withdifferent letters are significantly different (Duncan's test, P<0.01).

Figure 2 Seasonal evolution of yeast assimilable nitrogen (YAN) in grapes asaffected by soil types during the 2007 and 2008) vintage. The last datapoints correspond to harvest. Vertical bars indicate standard error ofmeans. For the same date, means foiowed by the same letter are notsignificantly different at P<5% according to Duncan test.

A principal component analysis (PCA) was run on the several physiological variables measured oneach location (Figure 3). Those variables include indicators of vine vigour, vine mineral status,phenology and must composition. Data about phenology are depicted only for the 2008 vintage, sinceno data were recorded in 2007. PCA provides a visualization of the relation between the physiologicalvariables (Figure 3A and 3C) and a two-dimensional map of the locations (Figure 3B and 3D). Thefirst dimension is linked to nitrogen parameter, vine vigour and must acidity. Those variables arecorrelated. Yeast assimilable nitrogen was well correlated with vine vigour indicators (weight ofpruned branches: r = 0.6 significant at α=0.05, berries weight: r = 0.82 significant at α=0.001).

3 - 131



On the individuals map (Figure 3B), first axis distinguishes significantly locations according to theirsoil type. Vines on bottom moraines showed weaker vigour, lower nitrogen content in leaves andberries and their musts were more acid compare to other soil types.In 2008, as shown in Figure 3C, the first dimension of the PCA analysis was positively correlatedwith vine vigour, nitrogen status, must's acidity (pH and malic acid content) and the day of year ofapex growth cessation. The amount of sugars in must at harvest was negatively correlated with vinevigour, vine nitrogen status and apex growth (weight of pruned branches: r=-0.86 significant atα=0.01, leaf nitrogen: r=-0.92 significant at α=0.001, growth cessation: r=-0.73 significant at α=0.01).Bud break and bloom were linked to the second dimension of the PCA. On the individuals (locations)map (Figure 3D), the two soil categories can be discriminated mainly along the F1 axis. During the2008 vintage, vines growing on bottom moraines had a lower nitrogen status and their grapes reachedhigher sugar content at harvest compare to vines growing on other soil types. Furthermore, locationswhich belong to the category "other soil types" showed vine precocity from bud break (average of 5.5days earlier) and bloom (average of 2 days earlier).As for the physiological variables, a principal component analysis was carried out on data from winesensory analysis (Figure 4). YAN in must has been added to PCA analysis as illustrative variable. Forthe vintage 2007, YAN points on the same direction than positive sensory descriptors such asgustatory persistence and structure (Figure 4A). YAN was observed to be strongly negativelycorrelated to the sensory astringency perceived in wine by our panel (r = 0.89 significant at α=0.001).Wines produced from location situated on bottom moraines were judged astringent and tended toshow lower colour intensity (Figure 4B). On the 2008 vintage, no significant correlation was detectedbetween YAN in the musts and wine astringency (Figure 4C). Locations which were not situated onbottom moraines gave wine more balanced (Figure 4D). On the other hand, vine from bottommoraines were assessed to be more acid, astringent and showed greater colour intensity. For bothvintages, wines from the 2 different soil types were discriminated by our panel during sensoryanalysis sessions. On both vintages, wine produced on bottom moraines were characterized asastringent and were assessed less structured and balanced.

Discussion

It is generally accepted that some sort of stress (mainly mineral deficiency or water limitation) isbeneficial for fruit composition and wine quality (Keller et al., 2005; Van Leuween et al., 2004).Water supply to vines was abundant in our study due to regular rainfall. Therefore, no water limitationwas observed for the 2 vintages on the studied vines. But on the other hand, vines did show adifferentiate nitrogen supply according to vintage and soil types. Nitrogen status of vines was higherin 2008 compare to 2007 (Figure 1 and 2). Musts from vines on bottom moraines had a YAN averageslightly under 140 mg/l at harvest, in 2008, the YAN of the same category were over 160 mg/l. 140mg/l is generally considered to be the under limit of nitrogen content in must in order to avoid a stuckfermentation (Butzke, 1998). The difference in nitrogen supply on these 2 vintages could be explainedby the exceptional climatic conditions of the 2007 vintage. The summer 2007 was extremely rainy andwith lower temperature. These adverse conditions might have interfered with absorption andassimilation of nitrogen (root asphyxiation). This hypothesis is supported by the observation of earlysenescence of basal leaves (yellowing and shedding) around veraison detected in 2007 only.On both vintages, must nitrogen was related to the content of malic acid in the berries at harvest. Thisrelation can be explained through vine vigour, having a higher N supply, vine would growth a densercanopy and thus affecting the microclimate (temperature and sun exposition) of grapes and leading toa lower degradation of malic acid in the berries. Must pH increased linearly with increasing nitrogencompounds in berries (YAN), we observed good correlation on the 2 vintages between YAN and mustpH (2007: r = 0.68; 2008: r = 0.63, both significant at α = 0.05). Similar results have been obtained bySpayd et al. (1994).

Figu

re4

PCA

Rep

rese

ntat

ion

ofth

em

ain

sens

ory

varia

bles

eval

uate

don

the

12lo

catio

ns’w

ines

from

the

2007

(A)a

nd20

08vi

ntag

e(C

).Y

east

assi

mila

ble

nitro

gen

(M_Y

AN

)w

asad

ded

asill

ustra

tive

varia

ble.

Ast

eris

ksin

dica

teva

riabl

esw

hich

are

sign

ifica

ntly

diff

eren

tbet

wee

nth

e2

soil

cate

gorie

s(D

unca

n's

test

at5%

(*)a

nd1%

(**)

).Lo

catio

ns’w

ines

are

repr

esen

ted

(200

7:B

;200

8:D

)on

the

first

2di

men

sion

sof

PCA

and

colo

red

acco

rdin

gto

thei

rsoi

ltyp

es.B

aryc

entre

sof

the

two

soil

cate

gorie

sar

ede

pict

edw

ithth

eir

co

nfi

de

nc

ein

terv

al

at

95

%.

Se

ns

ory

va

ria

ble

s:

V_

=v

isu

al,

O_

=o

lfa

cto

ry,

G_

=g

us

tato

ry

3 - 132



On the individuals map (Figure 3B), first axis distinguishes significantly locations according to theirsoil type. Vines on bottom moraines showed weaker vigour, lower nitrogen content in leaves andberries and their musts were more acid compare to other soil types.In 2008, as shown in Figure 3C, the first dimension of the PCA analysis was positively correlatedwith vine vigour, nitrogen status, must's acidity (pH and malic acid content) and the day of year ofapex growth cessation. The amount of sugars in must at harvest was negatively correlated with vinevigour, vine nitrogen status and apex growth (weight of pruned branches: r=-0.86 significant atα=0.01, leaf nitrogen: r=-0.92 significant at α=0.001, growth cessation: r=-0.73 significant at α=0.01).Bud break and bloom were linked to the second dimension of the PCA. On the individuals (locations)map (Figure 3D), the two soil categories can be discriminated mainly along the F1 axis. During the2008 vintage, vines growing on bottom moraines had a lower nitrogen status and their grapes reachedhigher sugar content at harvest compare to vines growing on other soil types. Furthermore, locationswhich belong to the category "other soil types" showed vine precocity from bud break (average of 5.5days earlier) and bloom (average of 2 days earlier).As for the physiological variables, a principal component analysis was carried out on data from winesensory analysis (Figure 4). YAN in must has been added to PCA analysis as illustrative variable. Forthe vintage 2007, YAN points on the same direction than positive sensory descriptors such asgustatory persistence and structure (Figure 4A). YAN was observed to be strongly negativelycorrelated to the sensory astringency perceived in wine by our panel (r = 0.89 significant at α=0.001).Wines produced from location situated on bottom moraines were judged astringent and tended toshow lower colour intensity (Figure 4B). On the 2008 vintage, no significant correlation was detectedbetween YAN in the musts and wine astringency (Figure 4C). Locations which were not situated onbottom moraines gave wine more balanced (Figure 4D). On the other hand, vine from bottommoraines were assessed to be more acid, astringent and showed greater colour intensity. For bothvintages, wines from the 2 different soil types were discriminated by our panel during sensoryanalysis sessions. On both vintages, wine produced on bottom moraines were characterized asastringent and were assessed less structured and balanced.

Discussion

It is generally accepted that some sort of stress (mainly mineral deficiency or water limitation) isbeneficial for fruit composition and wine quality (Keller et al., 2005; Van Leuween et al., 2004).Water supply to vines was abundant in our study due to regular rainfall. Therefore, no water limitationwas observed for the 2 vintages on the studied vines. But on the other hand, vines did show adifferentiate nitrogen supply according to vintage and soil types. Nitrogen status of vines was higherin 2008 compare to 2007 (Figure 1 and 2). Musts from vines on bottom moraines had a YAN averageslightly under 140 mg/l at harvest, in 2008, the YAN of the same category were over 160 mg/l. 140mg/l is generally considered to be the under limit of nitrogen content in must in order to avoid a stuckfermentation (Butzke, 1998). The difference in nitrogen supply on these 2 vintages could be explainedby the exceptional climatic conditions of the 2007 vintage. The summer 2007 was extremely rainy andwith lower temperature. These adverse conditions might have interfered with absorption andassimilation of nitrogen (root asphyxiation). This hypothesis is supported by the observation of earlysenescence of basal leaves (yellowing and shedding) around veraison detected in 2007 only.On both vintages, must nitrogen was related to the content of malic acid in the berries at harvest. Thisrelation can be explained through vine vigour, having a higher N supply, vine would growth a densercanopy and thus affecting the microclimate (temperature and sun exposition) of grapes and leading toa lower degradation of malic acid in the berries. Must pH increased linearly with increasing nitrogencompounds in berries (YAN), we observed good correlation on the 2 vintages between YAN and mustpH (2007: r = 0.68; 2008: r = 0.63, both significant at α = 0.05). Similar results have been obtained bySpayd et al. (1994).

Figu

re4

PCA

Rep

rese

ntat

ion

ofth

em

ain

sens

ory

varia

bles

eval

uate

don

the

12lo

catio

ns’w

ines

from

the

2007

(A)a

nd20

08vi

ntag

e(C

).Y

east

assi

mila

ble

nitro

gen

(M_Y

AN

)w

asad

ded

asill

ustra

tive

varia

ble.

Ast

eris

ksin

dica

teva

riabl

esw

hich

are

sign

ifica

ntly

diff

eren

tbet

wee

nth

e2

soil

cate

gorie

s(D

unca

n's

test

at5%

(*)a

nd1%

(**)

).Lo

catio

ns’w

ines

are

repr

esen

ted

(200

7:B

;200

8:D

)on

the

first

2di

men

sion

sof

PCA

and

colo

red

acco

rdin

gto

thei

rsoi

ltyp

es.B

aryc

entre

sof

the

two

soil

cate

gorie

sar

ede

pict

edw

ithth

eir

co

nfi

de

nc

ein

terv

al

at

95

%.

Se

ns

ory

va

ria

ble

s:

V_

=v

isu

al,

O_

=o

lfa

cto

ry,

G_

=g

us

tato

ry

3 - 133



N status of vine was highly dependant on soil type (Figure 1 and 2). Bottom moraines are verycompact (Letessier et al., 2004), thus it's probably difficult for plants to colonize those soils and toestablish their root system. As a consequence grapes grown on that soil type showed minor vigour,smaller berries and, when nitrogen supply is not too severe (like during 2008), greater amount ofsugars. Vegetative period was shorter for vines on bottom moraine, they showed a delayed bud breakand bloom and an earlier growth slackening. During the 2008 vintage, limited nitrogen supply wasrelated to early vegetative growth cessation and higher sugars accumulation. It was the not the case ofthe previous vintage. Environmental stress enhances grape quality, as far as the stress intensity is nottoo high, which it has probably be the case in 2007. Our results confirm the positive impact of amild nitrogen deficiency on grape quality obtained in others study (Choné et al. (2001); Tregoatet al. (2002)). During a mild nitrogen deficiency, berry weight decreases and a higher concentration offavourable compounds in the berry is obtained. On contrary, as it has been demonstrated by Peyrot desGachons et al. (2005), a severe nitrogen deficiency is detrimental to the aroma potential ofSauvignon blanc grapes. In the 2007 vintage, where the grapes nitrogen status was very low, weobserved a strong link between low YAN in must and marquee astringency in wines (significant at1‰). For the 2008 vintage with generally higher nitrogen level in must, the correlation was notsignificant anymore. In our study, vines located on the terroir "bottom moraines" showed consistentlylow nitrogen content and the quality of their wines tended to be depreciated.

Conclusion

The aim of this study was to evaluate the influence of terroir on vine behaviour and wine quality. Theterroir effect on vine development and wine quality can be explained in large part by the influence ofsoil type (soil compactness and rooting depth) on vine nitrogen level. Thus, our study confirms therole of nitrogen supply in grape and wine quality and underline nitrogen as a key factor inunderstanding the terroir effect.

Ackowledgments

Special acknowledgments go to G. Nicol for winemaking and supplying data about sensory analysis.

ReferencesAERNY J., Composés azotes des moûts et des vins. Revue suisse Vitic., Arboric., Hortic. 28: 161:165 (1996)BUTSKE C., Survey of Yeast Assimilable Nitrogen Status in Musts from California, Oregon, and Washington. Am. J.

Enol. Vitic. 49:220-224 (1998).CHONE X. et al., Terroir influence on water status and nitrogen status of non irrigated Cabernet-Sauvignon (Vitis

vinifera). Vegetative development, must and wine composition (example of a Médoc top estate vineyard, Saint Julienarea, Bordeaux). S. Afr. J. Enol. Vitic. 22: 8-15 (2001).

GAUDILLERE J. P. et al., Carbon isotope composition of sugars in grapevine, an integrated indicator of vineyard waterstatus. J. Exp. Bot. 53:757-763 (2002)

KELLER M. et al., Nitrogen and water management strategies for wine-grape quality. Acta Hort. 640: 61–67 (2005).LETESSIER I. et al., Etude des terroirs viticoles vaudois, Géopédologie. Rapport SIGALES, études de sols et de terroirs,38410 St Martin d'Uriage (France) (2004)

PEYROT DES GACHONS C. et al., Influence of water and nitrogen deficit on fruit ripening and aroma potential of Vitisvinifera L cv Sauvignon blanc in field conditions. J. Sci. Food Agric. 85:73-85 (2005).

TREGOAT O. et al., The assessment of vine water and nitrogen uptake conditions by means of physiological indicators.Influence on vine development and berry potential. J. Int. Sci. Vigne Vin 35:133-142 (2002).

SMART R. E. et al., Canopy microclimate implications for nitrogen effects on yield and quality. Proc. Intl. Symposium onNitrogen in Grapes and Wine, Seattle, 90–101 (1991)

SPAYD S. E et al., Nitrogen fertilization of White Riesling grapes in Washington. Must and wine composition. Am. J.Enol. Vitic. 45:34-42 (1994).

VAN LEEUWEN C. et al., Influence of climate, soil and cultivars on terroir. J. Int. Sci. Vigne Vin 55:207-217 (2004).ZUFFEREY V. et al., Assessment of plant hydraulics in grapevine on various terroirs in the canton of Vaud(Switzerland). J. Int. Sci. Vigne Vin 41:95-102 (2007).

3 - 134



Terroir climate change

Technology

Transcript of Terroir climate change