Sampling Vineyard Ecology and Biodiversity

in Bordeaux

2012 FIELD REPORT

Background Information

Lead PI: Maarten van Helden

Report completed by: Josepha Guenser

Period Covered by this report: May-July 2012

Date report completed: 2012-09-03 13:58:23

To all participants of the 2012 ‘Sampling Ecology in Bordeaux Vineyard” Earthwatch

expedition.

Gradignan France, Monday, 03 September 2012

Dear all,

We hope you all have good memories of the two weeks we spend together at Château les

Vergnes. We as organisers thoroughly enjoyed your very positive and precious contribution

to our project. The student that participated during the expeditions (Quentin Sanz) has

continued his analyses that we started with you and we are very pleased with the 2012

results. These expeditions are the last ones of a five-year experiment and collaboration with

Château les Vergnes around the topic of biodiversity. A big conference will take place at

Château les Vergnes in 2013 to celebrate the successful and fulfilling collaboration between

all the partners of the project and to disseminate the results obtained on farm biodiversity.

But the story does not stop there! These pilot experiments we have performed with the

Earthwatch volunteers help were very valuable for us both scientifically (data provided) and

educationally (we improved our way of communicating with professionals on the topic of

biodiversity).

Also:

• It allowed us to move towards bigger-scale projects, like the BioDiVine for which you

recorded the cartography data around Saint Emilion (7 sites in 3 European countries)

• At Château les Vergnes the staff will continue to apply the farmscaping practices we

proposed during our collaboration and spread the message beyond. We now start to

welcome groups of external stakeholders for training on biodiversity and landscape

management.

We hope this expedition gave you a meaningful appreciation of the complexity farmers and

scientists have to face to better understand how to preserve biodiversity while still remaining

economically sustainable. And we also hope you brought back with you some simple, easy-

to-tell stories in order to talk about your experience and raise environmental (and wine!)

interest around you! Thank you!

SECTION ONE: Scientific research achievements

Top highlight from the past season

The work on biodiversity measurements and conservation was initiated at Château les

Vergnes in 2007 thanks to partnerships between Earthwatch, Syngenta, Château les

Vergnes and Enita de Bordeaux scientific team. During five years this partnership allowed

working on a very new, specific and unusual topic for agricultural stakeholders. Involving

volunteers in this experience was a keystone for several aspects; the main ones are

certainly education and production of user-friendly methods to “measure” biodiversity. This

last season, with two teams, confirmed the good success regarding these two objectives.

The protocols tested at château les Vergnes with Earthwatch volunteers are now used for

bigger projects. Volunteers were interesting and interested, and we thank all the participants

for the great job done!

Reporting against research objectives

1. To quantify and validate the effect of farmscaping measures on general and functional

biodiversity at a landscape scale (functional biodiversity in this case refers to effects on pest

population levels)

2. To propose validated farmscaping practices appropriate to viticulture

3. To help farmers to communicate with the general public, politicians etc. in order to show

the multi-functionality of viticulture

4. To develop a simplified monitoring schedule for general biodiversity and its development

over time that can be applied by students, technicians and farmers

5. To validate the monitoring schedule both scientifically and educationally

6. To increase awareness of technical staff (on farm, extension workers) and other

stakeholders on biodiversity and the possibilities of farmscaping in viticulture

7. To develop methods to communicate on all multifunctional aspects of farmscaping and

biodiversity with scientists, students, practitioners, technicians, farmers, and other non-

specialist stakeholders

8. To propose and apply farmscaping practices for Château les Vergnes

Summary: The plant species richness of vineyard inter-row ground cover can maintain

biodiversity of plants and arthropods, and could improve conservation bio control by

reducing pest insect pressure. Five different seed mixtures were tested during three years

on a vineyard plot at the Bordeaux vineyard of Château of Les Vergnes. The initial plant

composition and its evolution over time were studied in 2009 (installation year) and 2010

through exhaustive botanical monitoring during the growing season (April-July). Large

differences were observed in plant species composition and abundance, even though the

existing seed bank seems of major influence on the species richness.

To follow-up with these results provided by the exhaustive (and time-consuming!) 2009-2010

surveys, an Earthwatch project related to this experiment was performed in 2011. Simple

and fast protocols were tested with groups of volunteers participating, in order to test the

results obtained with these simplified protocols. For the last year of survey 2012, the protocol

tested in 2011 by volunteers was replicated, as the 2011 results were satisfying. An

exhaustive survey was performed at the same time in order to compare the fast and

simplified protocol (volunteers) and the exhaustive “expert” survey (1.5 month internship,

Quentin Sanz Romero).

Material and method (Objectives 1, 4, 5)

Experimental plot

The plot is located on the Château les Vergnes vineyard (Fig 1)

Fig 1: Château les Vergnes map

It is composed by 22 vine rows of “cabernet franc”, planted in 2006. Its surface is 7200

square meters (1.78 acres). Experiment started in autumn 2008 (first sowings). Five

modalities, so five different treatments (groundcovers) were set, replicated 11 times each:

• Modality n°1 (GRAM): Sowing , Red fescue/white clover 5g/m²

• Modality n°2 (HORT): Sowing, flowers mixture 1, (Company : Nova-Flore), 3g/m²

• Modality n 3 (BANK): No sowing, natural soil-bank of seeds

• Modality n°4 (HAY): Seeds from Hay (harvested in the natural meadow of the Château)

were sown.

• Modality n°5 (WILD): Sowing, flowers mixture 2 “wild flowers” from Nova Flore, 3g/m²

Each modality covers 3 inter-rows for a length of 15m (see Figure 1):

Figure 1

Composition of sown mixtures (HORT and WILD) is detailed in Table 1:

Table 1: Composition of sown mixtures (HORT and WILD)

Protocol 2012: Biodiversity assessment 2012 was performed by volunteers only for

arthropods. Concerning the flora aspects, volunteers applied the simplified protocol which

was tested in 2011 (but no comparison with the exhaustive survey was done then), and an

exhaustive survey was performed by an intern student during the same period (so same

botanic composition for both surveys, which allows us to compare).

Botany

We performed the botanical survey in several steps:

1. One day of training. Participants (duos or trios) were told to sample one specimen of each

single plant they had in one inter-row of each modality. Then we went back to the lab in

order to identify the plants with help from books and the PI’s advice. All specimens were

stuck on big paper sheets in order to create one herbarium per duo.

2. The survey was performed by volunteers the following days. Participants had to recognize

plants located in a one-square-meter area, chosen to be representative of the overall

modality. If they found some plants they were not able to recognize, they sampled it in order

to solve the identification problem at the end of the day with the help of the lead scientists

and the rest of the group. The data collected consists in presence/absence of species on

each square meter (example of datasheet in appendix 2).

Appendix 2: Example of datasheet survey

3. The exhaustive survey was performed on two inter-rows over three per modality. The data

collected consists in presence/absence and percentage cover of each species.

Arthropods

For arthropods we set a pitfall trap in the centre of each modality. The traps were activated

with water (5% salt and a few drops of detergent), on May, 15th for team 1, and June, 5th for

team 2. During each team, arthropods were collected twice, each time after three days of

trapping. Samples were analysed in the laboratory; training to recognise arthropods orders

was required (half a day of training). The planning of experiment is given in table 2.

Table 2: Arthropod experiment planning

Meso-fauna

For meso-fauna, a new Berlese extraction system was built to allow extraction on smaller

quantities (1L) of soil compared to 2011, (the 5L tested in 2011 provided a too high quantity

of organisms to count). The soil was sampled at the same time pitfall traps were set. In order

to start with very extreme situations to compare, we decided to sample soil in modality 3,

both in the middle of the row (where vegetation grows) and under the row (where weed-killer

is sprayed regularly). Extraction was performed during three days and samples analyses

were done in the laboratory at the end of the team.

Analyses

Non-parametric correlation (Spearman) and comparison (Wilcoxon and Kruskall-Wallis) tests

were performed on the dataset, at a confidence level of 0.05.

Results:

General results

Data recorded: quantity and ease of obtaining (objectives 4 and 5). As for 2011, the

volunteers felt confident in recording data once the training session were done. The

pedagogic supplies were used again and helpful for volunteers. Datasheets were prepared

in advance to make data recording easier. The quantity of data recorded during each

expedition is given in table 3.

Table 3: Quantity of data recorded during Earthwatch expeditions

Link between groundcover data recorded by volunteers and by “expert” (objectives 4 and 5)

The results provided by volunteers during the two sessions of survey and the exhaustive

survey (student intern) are shown in figures 2 and 3.

Figure 2: Groundcover data recorded by Earthwatch volunteers

Figure 3: Groundcover data recorded by student intern

A linear regression was performed in order to determine if volunteers’ data could be

sufficient to explain the plot floristic diversity (richness). The results of this test shows that

the total richness found by the expert is better represented that by the volunteers’

data(r²=0.183) (figure 4).

Figure 4

Comparison of modalities considering botany composition/richness

A non-parametric comparison test (Kruskall Wallis) was performed on volunteers’ data (

richness recorded) and “expert” data(Table 4):

Table 4

Results of Earthwatcher Team 1 and Team 2 were analysed combined and separately.

These tests show a difference between modalities. A complementary Dunn test allows

determining which modalities differ from the others. Results are presented in table 5:

Table 5

We can see that the results differ according to the data considered. The expert data shows a

very clear distinction between modality 2 as the poorest one and the four other modalities

which are in the same category. No modality is classified as “intermediate” according to the

expert dataset, whereas Team 1 and Team 2 do not show the same distinction between

groups, and show some intermediate modalities, which vary according to the team

considered. Volunteers’ data nowadays always classify modality 2 in the “poorest” group

which is consistent with expert’s data.

Potential impact on arthropods

Arthropods total abundance has been calculated for each replicate in each modality. The

distribution of data is shown in figure 5:

Figure 5

A Kruskall-Wallis test has been performed on this dataset. No difference can be shown

between modalities in total arthropods abundance (Kobs=2.36 ; Kcrit=9.49, p-value=0.67). In

order to test if some specific orders could be more appreciated as indicators for biodiversity,

the comparison test was performed for two abundant orders: Spider and Coleoptera.

Abundance of spiders was not significantly different from one modality to another

(Kobs=3.96 ; Kcrit=9.49, p-value=0.41), but abundance of coleoptera did differ significantly

from one modality to another (Kobs=16.11 ; Kcrit=9.49, p-value=0.003). The two modalities

which differ are modality 4 (Hay) and modality 5 (wild). Modality 4 seemed to be the most

favorable to important quantities of coleoptera (average 9.5 individuals per sample), while

much less coleoptera were found in modality 5 (average 3.8 individuals per sample).

In order to determine if the floristic richness recorded could be linked to arthropods

abundance a correlation test was performed. No significant correlation was found (p-

value=0.73). Meso-fauna experiment:the two treatments “presence of ground cover” and

“weed-killer spraying” were compared thanks to a Wilcoxon test. No significant difference

were found in the quantity of meso-organisms according to the treatment (p-value = 0.508).

Conclusions and discussions:

These first analyses on the 2012 dataset allow drawing the following conclusions:

1. In determining whether volunteer’s data was comparable to experts data to find some

differences between the five treatments, the 2012 comparison between volunteers’ and

expert’s results suggest we need to take some precautions. The relationship between

richness found by volunteers and expert is very lowand the comparison tests do not show

exactly the same differences between modalities: while the expert’s data shows a drastic

difference between modality 2 (poorest modality) and the others, volunteers’ data is less

distinct. In fact modality 3 is classified as intermediate. However, while the precision seems

to be lower, the conclusion remains the same in both comparisons: modality 2 is the poorest

one.

2. In 2011 and 2012, no clear correlation has been found between flora diversity and

arthropods abundance.

3. In 2012, the meso-fauna protocol was designed differently to allow analyses. This part of

the objective was satisfying (the quantity of material allowed us to assess the 20 samples in

one afternoon). No clear difference was shown between the quantity of organisms sampled

in the inter-row and the samples collected where herbicide is regularly sprayed.

SECTION TWO: Impacts

Partnerships

Bordeaux Sciences Agro (Ex-ENITAB): support in scientific protocols, knowledge,

organization UNIVITIS: experimentation site, support in logistic (rooms, coffee breaks,

celebratory dinner, visits, relationship between volunteers and viticulture professionals).

Syngenta: funding support. The new Life+ BioDiVine project (same thematic : biodiversity,

landscape and viticulture) is still on-going (2nd year). Partnership with IFV (French Institute

of Viticulture), two Spanish partners (Diputacio de Barcelona and Consejo Superior de

Investigaciones Científicas), and one Portuguese partner (Associação para o

Desenvolvimento da Viticultura Duriense). Vitinnov is in charge of the scientific coordination

of the project.

Contributions to conventions, agendas, policies, management plans

International

Life+ BioDiVine project is about two main axes:

1. Monitoring and enhancing knowledge and comprehension of biodiversity in viticulture

landscape. Protocols validated and used in this project are directly linked to Earthwatch

results. In fact RBA on Arthropods will be implemented on each demonstration site of the

project. Communication towards professionals is also inspired from Earthwatch experience,

as the project includes “demonstration” and “dissemination” aspects.

2. Conservation actions proposed are also linked to previous results obtained on Château

les Vergnes with Earthwatch. Hedgerows, melliferous meadows are planned to be restored

in all demonstration sites. The species composition selected to implement these actions are

based on results and observations done on the Château les Vergnes experiment.

3. A new project has been submitted and selected for funding by the EU. This project

QUESSA will aim at measuring the ecosystem services linked to semi-natural habitats in

agricultural landscapes. The project (2013- 2017) is a European wide collaboration with

academic partners in which Bordeaux Sciences Agro will be in charge of two wine growing

areas, Saint Emilion and Costières de Nîmes.

National or regional

The sites we are currently working on at a landscape scale are Saint Emilion (Bordeaux) and

Corton hill (Burgundy). Saint Emilion: Diagnosis achieved and first conservation actions set

(8 ha of melliferous meadows sown, 2km of hedgerows planned, 800m planted). Local

funding obtained for 2011 but still pending for 2012. Life + funding validated and available

until 2014. Burgundy: Diagnosis achieved, Life+ funding near to be accepted by European

Commission. Apart from this scientific thematic, our expertise is required to help with the

implementation of a Natura 2000 project dedicated to bats conservation (site “carrières de

Villegouge”, FR7200705) A new national research program ‘BIOCONTROL’ has been

initiated and funding has been acquired. Project partners are the French Institute of

viticulture, INRA and chambers of agriculture of several regions. This project (2013-2015) is

aiming at determining the link between landscape composition and structure and the impact

of natural enemies on the main insect pests of the vineyard.

Local

The Bordeaux Wine Board (CIVB) is now disseminating a methodology in order to

encourage wine companies to work collectively for obtaining the international environmental

standard ISO14001. The project (Called SME) targets the Bordeaux Region and several

workshops are organized on a large number of topics (from waste management to energy,

and including biodiversity). We are regularly contacted to spend half a day on the

biodiversity topic, and our experience is very appreciated both for scientific and practical

aspects. Moreover a training session was performed for technical staff of “chambres

d’agriculture” in March 2012 and the demand was renewed for 2013. The Bordeaux Wine

Board is also editing a leaflet with technical guidelines (‘cahier technique’) on the topic of

biodiversity management in vineyards. We have been supplying part of the technical and

scientific information and illustrations.

Developing Environmental Leaders

Training session for wine growers : 03/05/2012 (20 participants), at Château Grand Corbin

d’Espagne (training on Biodiversity for SME) 10/07/2012 (9 participants), at Saint Germain

du Puch (training on Biodiversity for SME) 06/12/2011 (13 participants), at Vauvert (training

on biodiversity for wine growers, included in a global training on agro-forestry) Training

sessions for extension services 21/03/2012 (12 participants), at Château les Vergnes (two

days on biodiversity) 05/07/2012 (50 participants), at ISVV 06/04/2012 (30 participants), at

Bordeaux Sciences Agro (general presentation to Bordeaux Sciences Agro staff on

Vitinnov’s activities)

Conservation of Habitats

Main habitats restored in the different projects are diversified hedgerows and melliferous

meadows. The type of enhancement is considered at the landscape scale as hedges and

meadow are introduced (planted / sown), using local varieties. Positive impact in terms of

connectivity and resilience amelioration is expected and now measured (see results of

habitats comparison 2009-2010 at château les Vergnes, 2011 in the different demonstration

sites of Biodivine project).

Ecosystem Services

Yes through reintroducing diversified hedges and meadows and promoting respectful

farming practices (mowing frequency etc.).

Dissemination of research results

Scientific peer-reviewed publications

IOBC meeting on “Landscape management for functional biodiversity” in Lleida, Spain

(2012, May, 7-11th).

• Vine ground cover experiment of ADERA-VITINNOV presented a scientific poster (see

below).

• Presentation of EW project and Biodivine project by Maarten van Helden (see below).

0

5

10

15

20

25

30

35

40

45

GRAM HORT FOIN SAUV

% o

f d

iffe

ren

tce b

etw

een

tr

ea

tmen

t an

d C

ON

T

Treatment (X)

2009

2010

Contacts Acknowledgements

Evaluation of different ground cover mixtures

to maintain botanical biodiversity in viticulture

Maarten van Helden, Josépha Guenser, Emma Fulchin

Vineyard inter-row ground cover could maintain botanical biodiversity.

Five different seed mixtures were tested during two years on a vineyard plot of Château Les Vergnes

(Appellation Bordeaux, Gironde, France).

• Five different treatments and 11 replicates

• Plot set up in autumn 2008

• 3 botanical surveys in 2009 and 2010

1.RFWC : Red Fescue/ White Clover 5g/m²

(2 species)

2.HORT : Horticultural flower mixture, (13 species

Nova-Flore), 3g/m²

3.CONT : Control expressing seed bank

4.HAY : Seeds applied through hay collected in

adjacent meadow

5.WILD : Mixture of 22 Species of wild native

Plants, 3g/m²

Experimental plot and observations

Figure 1. Map of the plot with each sub plot

(modalities 1 to 5, reps A to K)

Inter-row Vine row

Results

Figure 2. Richness of each modality during the three sessions of botanical survey (2009 and 2010)

Figure 3. Hamming distance (% difference)

with CONT in 2009 and 2010.

The hamming distance (% of species which differ) between CONTrol and sown ► treatments is ‘small’ or ‘intermediate’ for all modalities.

• Differences of RFWC, HORT compared to CONT stable over years,

• Differences WILD compared to CONT, increase between years due to perennials

• HAY approaches CONT

◄ Richness does not differ

among treatments (2009, 2010)

• Number of species recorded much

higher than sown (expected);

• Seed bank richness (control)

comparable to other treatments

Sowing seeds not always useful ?

Sowing locally harvested seeds (HAY) seems to be the richest modality, but quickly evolves towards CONT.

This shows that cultural practices select in the local species pool.

• It is the cheapest and the most adapted « local » solution, avoiding potential genetic pollution.

• Local meadows are to be considered as seed sources for ground cover enrichment if the plot seed bank is insufficient.

Perennial seed mixtures (WILD) show optimal installation in the 2nd year, but do not increase overall species richness.

Vines vigor measurements (2010) did not show any differences among treatments.

Anaïs Moison (2009) and Coralie Pineau (2010) for the botanical surveys,

assisted by Earthwatch expedition members and EW scientific board.

Nova-flore for providing seeds, Jean-Baptiste Rivoal and Coralie Laveau,

Jean Jacques Brethon and Serge Labat of the Univitis / Château les

Vergnes team, ‘The Earthwatch Institute’ and Syngenta for supporting this

experiment.

Maarten VAN HELDEN, Bordeaux Sciences Agro, Univ. Bordeaux,

Gradignan, France - maarten.vanhelden@agro-bordeaux.fr

Josépha GUENSER, Emma FULCHIN, ADERA-Vitinnov

Gradignan, France - vitinnov@agro-bordeaux.fr

0

10

20

30

40

50

60

70

80

90

1 2 3 4 5

No

mb

re d

'esp

èce

s

Modalité

2009

Juin

Juillet

Aout

0

10

20

30

40

50

60

70

80

90

1 2 3 4 5

No

mb

re d

'esp

èce

s

Modalité

2010

Avril

Juin

Aout

RFWC HORT CONT HAY WILD GRAM HORT TEM FOIN SAUV RFWC HORT CONT HAY WILD

RFWC HORT HAY WILD

Tota

l R

ichn

ess

Treatment

Tota

l R

ichn

ess

Treatment

0

10

20

30

40

50

60

70

80

90

1 2 3 4 5

No

mb

re d

'esp

èce

s

Modalité

2010

Avril

Juin

Aout

June

July

august

0

10

20

30

40

50

60

70

80

90

1 2 3 4 5

No

mb

re d

'esp

èce

s

Modalité

2010

Avril

Juin

Aout

april

June

august

Evaluation of different ground covers to maintain botanical

biodiversity in viticulture

Maarten van Helden1, Josépha Guenser

2, Emma Fulchin

2.

1Bordeaux Sciences Agro, Univ. de Bordeaux, ISVV, 1 Cours Général de Gaulle 33170 Gradignan, France,

m-vanhelden@enitab.fr; 2Vitinnov, 1 Cours Général de Gaulle 33170 Gradignan, France, vitinnov@enitab.fr .

Abstract: The plant species richness of vineyard inter-row groundcover can maintain biodiversity of

plants and arthropods, and could improve conservation biocontrol by reducing pest insect pressure. Five

different seed mixtures were tested during two years on a Bordeaux vineyard plot Château Les Vergnes

(Gironde, France). The initial plant composition and its evolution over time were studied in 2009 (first

year after 2008 autumn sowing) and 2010 through exhaustive botanical monitoring during the growing

season (April-July). Large differences were observed in plant species composition and abundance, even

though the existing seed bank seems of major influence on the species richness. The farmers’

management of the plot (mowing, machine passing) shows a strong selection pressure on the plant

species present in the seed mixtures.

Key words: ground cover, viticulture, botany, biodiversity

Introduction

Inter-row groundcover is now part of the usual practices of most winegrowers in Western Europe, at

least when it is adapted to the local context (soil, climate …). Its main objective is to regulate the plant

vigour through competition for water and/or nitrogen between vines and groundcover. It also maintains

soil structure, avoids compaction and reduces erosion (ITV, 2002). The presence of groundcover also

contributes to the ecological connectivity between the different habitats of vineyard landscapes (forests,

hedgerows, meadows) and the plot itself.

This experiment aims to test five different strategies: Sowing of three different seed mixtures, seeds

collected on a local meadow, and ‘spontaneous’ vegetation from the seed bank. The main objective is to

measure the impact of these treatments on the groundcover botanical diversity, and the potential impact

on the vines.

Material and Methods

The experimental plot The 1 ha plot, planted in 2005 at 3300 plants/ha on shallow clay-lime soil is located at château Les

Vergnes (Les Lèves et Thoumeyragues, Gironde, France). The experiment started in autumn 2008

(sowing) using a randomized block design. Botanical surveys were done three times in May/June, July

and August 2009, and three times in June, July and August 2010.

Five different treatments and 11 replicates were tested:

1. RFWC : Sowing , red fescue/white clover 5g/m²

2. HORT : Sowing, horticultural flower mixture, (13 species Nova-Flore), 3g/m²

3. CONT : No seeds applied, control expressing seed bank (Control)

4. HAY : Seeds applied through hay (harvested in meadow of the Château)

5. WILD : Sowing, custom made mixture of 22 “wild flowers” from Nova Flore, 3g/m²

Details on mixtures 2 (HORT) and 5 (WILD) in table 1.

Figure 1. Map of the plot with each sub plot (modalities 1 to 5, reps A to K)

Table 1: Composition of seed mixtures 2 (HORT) and 5 (WILD).

Species list of HORT (Nova Flore) : Achillea millefolium, Adonis aestivalis, Anthemis tinctoria,

Bellis perennis, Cheiranthus cheiri,, Coriandrum sativum, (Echium plantagineum), Lotus

corniculatus, Matricaria reculata, Medicago lupulina, Myosotis alpestris ; Saponaria

ocymoides, Silene pendula (13 species)

Species list of WILD (Nova Flore) : Achillea millefolium ; (Agrimonia eupatoria) ; Anthemis

tinctoria ; Bellis perennis , Centaurea cyanus ; Centaurea scabiosa ; Crepis biennis, Daucus

carota ; Festuca rubra dawson ;( Festuca ovina ‘spartan)’, Galium verum ; Hypericum

perforatum ; Hypochoeris radicata; Lotus corniculatus ‘bako’ ; (Lychnis flos- cuculi) ; Medicago

lupulina ; Origanum vulgare, Papaver rhoeas ; (Prunella vulgaris) ; Sanguisorba minor,

(Scabiosa columbaria), Silene vulgaris (22 species)

IN BRACKETS (Plant species never observed), in BOLD: Species in strong regression between 2009 and 2010,

UNDERLINED: dominant species in 2010

The botany survey An exhaustive botanical survey was done in 2009 on the whole surface of the plot through three sessions

in May/June, July and August. Three surveys were done in 2010 (one third of the surface of each plot) in

April, June and July 2010. During each session each single species is recorded, and the percentage of

ground cover of each single species and of each major group of plants (‘Monocots’, ‘Fabaceae’ , ‘other

dicots’ and ‘bare soil’) was recorded (% of surface covered, Braun-Blanquet scale). Identification was

based on Blamey & Grey-Wilson (1991); Bonnier & De Layens (1986); Bonnier (1971), Grey-Wilson

(1994) and Hanf (1982).

Plant vigour measurement Vigour of a vine can be measured through the chlorophyll content of leaves, which is strongly correlated

to the nitrogen nutrition of the plant (Decante et al., 2009). This is measured on 30 leaves of each sub-

plot according to the standard protocol by Hydro N Tester (Norsk Hydro ASA, Norway).

Data analysis: In order to compare the composition of the different treatments,

we calculated Hamming distance (H) using the formula H= (1-J) (Cesar &

Daget, 1997) where J is the Jaccard coefficient. (AB) = number of common

species in A and B and (AB) = total number of species found in A and B

Inter-row Row

Results

Richness The 2009 survey shows a very high number of species in all treatments (fig. 2), due to the natural seed

bank of the soil. A drop in the species richness is observed through June, July and August; mainly due to

mowing pressure (annual species disappear). The hamming distance between the different treatments

was decreasing during the season. Modalities 3 (CONT) and 4 (HAY) show the biggest loss in terms of

species number between June and July (20%). Treatments 4 (HAY) and 5 (WILD) seem the richest but

this was not significant. Some perennial species were observed only from 2010 onwards.

0

10

20

30

40

50

60

70

80

90

1 2 3 4 5

No

mb

re d

'esp

èce

s

Modalité

2010

Avril

Juin

Aout

Figure 3a,b : Total plant species richness of each of 5 modalities among the three sessions of botanical

survey (2009 and 2010)

Number of total species on the plot In 2010, 107 species were recorded in all. This number is slightly lower than 2009 but this might be due

to the change in the survey protocol (only one third of the plot was assessed). In 2010 as well the

richness of each modality seems to change during the season. All modalities show a huge reduction of

richness in August, probably due to a very dry summer, but this effect seems to differ from one treatment

to another. Species richness is lowest for RFWC with 61 species. HORT and CONT contain 70 species

each, HAY shows 76 species and WILD seems to be the richest with 78 species.

Again a large number of species is recorded in addition to the sown species. The seedbank already

present in the soil (control) is expressing itself. That is why we calculated the Hamming distance

between the CONT and the other treatments (table 3)

The distance between the control and the sown treatments is small or medium for all modalities,

whatever the session of the survey. We can conclude as a consequence that the seed bank of the soil

plays a dominant role in the groundcover diversity, for all treatments.

Table 3: Hamming distance between CONTROL and sown treatments 1, 2, 4, 5 in 2010.

Session April June August

Treatment (x) 1 2 4 5 1 2 4 5 1 2 4 5

Hamming (H) distance between CONT

and treatment x 47 47 44 54 33 38 36 42 40 45 50 28

Treatments: 1 = RFWC; 2= HORT, 4 = HAY, 5 = WILD. H<20 : very small difference, 20<H<40: small

difference, 40<H<60 : medium difference, 60<H<80 : big difference, 80<H: very big difference.

Impact on the plant vigour Measurements did not show any significant difference between the groundcover and the chlorophyll

content of the vines

0

10

20

30

40

50

60

70

80

90

1 2 3 4 5

No

mb

re d

'esp

èce

s

Modalité

2009

Juin

Juillet

Aout

Discussion

Variation inside treatment For the five different treatments, we noticed an average richness (calculated on the basis of 11 replicates)

clearly lower than the total richness. Seed bank originated plant communities do change from one

replicate to another, due to differences of soil etc. (Cornwell and Grubb, 2003). The most important

variability between replicates is recorded for treatment 3 (CONT), and confirms this hypothesis. The

treatment 4 (HAY) also shows a huge variability, this is due to a high diversity of seeds, distributed non-

homogeneously in the hay.

Variability between treatments Richness results show that RFWC is the poorest treatment: total richness of 29 and 49 species

respectively for the first and second session in 2010, average richness 11.6 and 11.9 species. This low

diversity is probably due to the strong competition of this mixture on seed bank expression.

HAY is the richest regarding the two first sessions of 2010 (63 and 62 species). Nevertheless, the

average richness (16.57 and 19.6) is not the highest among the five different treatments. This indicates

that there is a huge number of species with low frequency of occurrence. The total richness can be

explained by the rich community of the collection meadow but its distribution is heterogeneous. What is

more, the plants of this treatment are local species and therefore most likely to resist to local conditions

(dry soil, …). The sown treatments RFWC, HORT and WILD were more homogeneous among

replicates. Even if big differences were recorded between treatments, all strategies of groundcover shown

a common basis provided by the seed bank of the soil.

This experiment allowed highlighting the main characteristics of different strategies concerning

groundcover. The HAY treatment seems most interesting for the botanical diversity of groundcover.

However the CONT shows good results in terms of diversity, without any effort. Future observations (in

4-5 years) should confirm this and more knowledge is still required to follow the evolution of the

groundcover and be able to translate experimental results into technical advice.

Acknowledgements

Anaïs Moison (2009) and Coralie Pineau (2010) for the botanical surveys, assisted by Earthwatch

expedition members and EW scientific board. Nova-flore for providing seeds, Jean-Baptiste Rivoal and

Coralie Laveau, Jean Jacques Brethon and Serge Labat, the Univitis / Château les Vergnes team and

‘The Earthwatch Institute’ and its scientific board for supporting this experiment.

References

Blamey, M., Grey-Wilson, C. 1991. La flore d’Europe Occidentale. Paris : Arthaud éditions.- 544p.

Bonnier G., de Layens G. 1986. Flore complète portative de la France, de la Suisse et de la Belgique. Paris : Belin

éditions.- 425p

Bonnier G. 1971.- Les noms des fleurs trouvés par la méthode simple. Paris : librairie Générale de l’enseignement.-

338p.

Cesar J. & Daget P. (1997). Recherche des limites écologiques dans une végétation. Revue Elev. Méd. vét.Pays

trop, 50 (2), pp 153-156.

Decante, D. van Leeuwen, C. & van Helden, M. (2009). "Influence of plot characteristics and surrounding

vegetation on the intra-plot spatial distribution of Empoasca vitis." Agricultural and Forest Entomology

Agricultural and Forest Entomology, pp. 377–387

Grey-Wilson C. 1994. Wild flowers of Britain and Northwest Europe. London: Dorling Kindersley handbooks.-

320p.

Hanf, M. 1982. Les adventices d’Europe : leurs plantules, leurs semences. Allemagne : BASF éditions.- 496p.

ITV France. 2002. L’enherbement permanent de la vigne. Cahiers Itinéraires d’ITV France. n°4. 16pp.

Viticulture, Landscape and Functional Biodiversity:

Agronomy, Ecology, Sociology and Economy!

Maarten van Helden1, Josépha Guenser

2, Emma Fulchin

2, Joël Rochard

3, Benjamin

Porte3

1Bordeaux Science Agro, Univ. de Bordeaux, ISVV, 1 Cours Général de Gaulle 33170 Gradignan, France.

m-vanhelden@enitab.fr; 2Vitinnov, 1 Cours Général de Gaulle 33170 Gradignan, France.

3IFV, 12, rue Sainte

Anne, 75001 Paris, France

Abstract The European Life+ program BIODIVINE focuses on the conservation of general

biodiversity in viticulture landscapes across Europe. Many such landscapes have a long history and

strong dominance of viticulture that has created remarkable landscapes. The conservation of

biodiversity in such landscapes, without impacting the landscape aesthetics and without constraints for

farmers (costs, loss of surface) is challenging. Efficient communication is needed to convince farmers

to adapt new practices. Ecosystem services such as conservation biological control can be used to

convince farmers but supporting data are not always convincing. Better results can be achieved by

focusing primarily on agronomic and economic reasons for landscape management.

The use of ground cover inside and around plots, hedgerow planting, sowing fallow plots with fodder

crops all have clear agronomic amenities that should stay the primary objective for the farmer.

Training farmers and personnel, and involving local stakeholders, help to increase awareness. Finally

efficient external communication on landscape actions can be included in marketing strategies, but

care should be taken to avoid greenwashing. Key words: Viticulture, Landscape project management, biodiversity, France.

Introduction

Conservation of biodiversity in agricultural landscapes depends on semi-natural habitats

(Duelli & Obrist, 2003, Billeter et al., 2010). In historical high quality wine production areas

(appellations d’origine controlees, AOC), sometimes event recognized as cultural landscapes

(UNESCO), semi-natural habitats are often still decreasing because of economic pressure,

resulting in an increase in vineyard surfaces. Farmers, citizens, politicians and other

stakeholders do become aware of the continuous loss of general biodiversity, but fear of

economic constraints and lack of knowledge on ecological and agronomical impacts are

inhibiting factors for conservation actions. Experience obtained through environmental action

plans in several wine-growing areas in France, presented in this paper allowed us to improve

the impact of such actions, resulting in more efficient communication, implication and

organisation of stakeholders and a considerable increase in conservation actions.

Material and methods

Project site description and results

Landscape action plans for biodiversity conservations were launched in several wine production

areas in France since 1997. Table 1 lists 8 such actions and gives information on total surface

and vineyard surface, project leaders, territorial limit of the project, primary goal, conservation

actions (types of actions and quantity achieved so far), stakeholders involved as pilots or

consultants in the action, the type of communication used in the project and a (subjective) note

on the efficiency of each of these projects to initiate conservation actions.

Table 1: List of French ‘landscape scale’ projects in viticulture and their main characteristics.

Results and Discussion

From the information on these different projects we can extract a certain quantity of information

in order to improve efficiency of such action, useful for future or ongoing actions. A more

thorough scientific analysis could be done if more projects are included, but this was not tempted

here. We consider here the amount of conservation actions achieved by the farmers (in a certain

Farmers

Policy makers

interbranch

org.

Landscape

architect

Scientists

Nat. cons.

Org.

Hedges

(km

)

Fallo

w

plo

ts a

nd

Meadow

(ha)

Vin

eyard

Gro

und

cove

r (h

a)

Inte

rnal

Exte

rnal

Ba

rba

nne

650

Wate

rshed

Wate

r +

Pestic

ide r

ed.

2000-2

004

Pla

nt

pro

tectio

n

serv

ice

-P

P-

--

--

-P

--

Eng

ranne

13000

(8000)

Wate

rshed

Specie

s

conserv

atio

n7

2008-

ongoin

g

Cham

ber

of

agricultu

reC

PP

--

C2

20 (

150

1)

0B

8-

-/+

Sa

um

ur-

Cha

mp

igny

6000

(1500)

Appella

tion

Functio

nal

bio

div

ers

ity

2006-

ongoin

g

Farm

ers

unio

nP

-P

-P

C20

-2 3

BP

8B

P+

+

Sa

int E

mili

on

412000

(8000)

Appella

tion

Genera

l

bio

div

ers

ity

2009-

ongoin

g

Farm

ers

unio

ns

PP

CP

PC

10

82

AB

8B

++

+

Lim

oux4

42000

(7800)

Appella

tion

Natu

re

conserv

atio

n

2005-

ongoin

g

Win

e tra

de

unio

nP

-P

--

P2

Not ye

t4

BB

+

Co

stiè

res d

e

Nîm

es

4

15000

(4500)

Appella

tion

Specie

s

conserv

atio

n5

2008-

ongoin

g

Farm

ers

unio

nC

-P

C-

C1

0(9

01

)3

BB

+

Vill

eg

oug

e972

(380)

Natu

re 2

000

Specie

s

conserv

atio

n6

2010-

ongoin

g

Nat.C

ons.

Org

.-

P-

--

P0,3

Not ye

tN

ot ye

tB

B?

Bo

urg

og

ne

4800 (

600)

Mic

ro-

regio

n

Genera

l

bio

div

ers

ity

2010-

ongoin

g

Indiv

idual

farm

ers

P-

CC

PC

Not ye

tN

ot ye

tN

ot ye

tA

BN

ot ye

t?

1 =

Exi

sting

me

ad

ow

s w

ere

inclu

de

d fo

r lo

w inp

ut m

ana

ge

me

nt co

ntr

acts

2 =

Org

anis

atio

n im

plie

d in la

ter

sta

ge

duri

ng

pro

ject.

3 =

gro

und

co

ver

ma

na

ge

me

nt w

as e

xpe

rim

ente

d, lo

we

r m

iow

ing

fre

que

ncie

s w

ere

ob

se

rve

d

4 =

Site

s p

art

icip

ating

in the

Life

+ B

iod

ivin

e p

rog

ram

(m

ostly

alre

ad

y im

plie

d in p

roje

cts

be

fore

)

5 =

Little

Busta

rd (

Te

tra

x te

tra

x) ; 6

= b

ats

; 7

=E

uro

pe

an m

ink (

Mu

ste

la lu

tre

ola

) a

nd

Euro

pe

an fre

shw

ate

r cra

yfis

h (

Au

str

op

ota

mo

biu

s p

allip

es

)

8 =

p

rese

nce

of a

mo

de

rato

r d

ed

ica

ted

to

the

pro

ject

Com

munic

atio

n

A =

Agro

nom

y

B =

Bio

div

ers

ity

P =

Pestic

ides

Co

nse

rva

tio

n a

ctio

ns

achie

ved

sin

ce

sta

rt y

ea

r

'Success rate"

Org

anis

atio

ns im

plie

d in p

roje

ct

ma

na

ge

me

nt (

P =

Pilo

t, C

=

Co

nsult)

Site

Total surface and

(vineyard) (ha)

Project leader

Main goal

Duration

Territorial limit

amount of time) as a measurement of ‘success’, even though this is clearly not a guarantee of the

actual success to achieve the goal of the action (biodiversity, pesticide reduction ...).

Territorial limit: In all ‘landscape’ actions the final objective is to change (improve) the

management of the landscape through the action of local actors: mainly wine growers that

generally are organised locally in organisations such as appellations. An efficient territorial and

social anchoring is necessary to ensure local commitment. Therefore existing and well identified

territorial networks are more suitable than ‘ecological’ (scientific) landscape units such as

watershed or nature conservation areas and even reinforce local identity!

Main Goal: Farmers in France are strongly pushed towards more sustainable management.

This often is felt as an additional imposed constraint and costs. Therefore it is of primary

importance that the action should not be felt as a constraint. Farmers should be able to identify

themselves in the action as a positive ambition. Pesticide reduction (negative goal) or Species

conservation of ‘cryptic’ species such as bats (Villegouge) or european mink (Engranne) are not

efficient goals to motivate farmers since they are felt as externally ‘imposed’ constraints. General

biodiversity or functional biodiversity are more efficient to raise awareness and interest.

Project leader: The same element of (negative) constraint versus (positive) ambition is linked

to the choice of the project leaders. If the farmers or farmers’ representatives are present in the

project from the very start, preferably as project leaders or even initiators, this clearly increases

the motivation of farmers to play an active role in the project rather than to ‘submit’ themselves.

Pilot committee and steering committee: In most projects a limited number of organisations

are involved in the management of the project to allow efficient management. It is important that

the actual landscape managers (farmers, land owners, land managers) are present in the pilot

committee. We are convinced that the presence of scientists and landscape architects is a positive

contribution to the credibility of projects. This pilot committee should be assisted by a ‘scientific

and technical steering’ committee that includes all other stakeholders and allows these to

contribute as ‘consultants’ providing ideas and suggestions, while avoiding the risk of

‘imposing’ constraints to the landscape managers, leading to desistance or excluding partners.

Conservation actions: AOC Viticulture is a financially rewarding activity. This means that

farmers tend to plant as much surface as possible. Acceptable conservation actions should not

consume vineyard surface, and useable surface for such actions should be identified and accepted

by the farmers. Because of the existing landscape (topography, history) and soil characteristic

AOC areas often are composed of many small plots, resulting in an important amount of

interstitial space (often > 10%) slopes, roads etc. (Porte et al. 2011 this meeting), that can be

partially used for actions such as hedgerows and grass strips. Normally a certain amount of plots

(often around 4%) is laying fallow between uprooting and replanting on which seed mixtures can

be implemented. Intra-plot ground cover is also contributing to general biodiversity. Other areas

can sometimes be identified such a garden and parks of wineries and specific conservation

actions can be developed.

Communication: General biodiversity is not a major concern for farmers. Even though they

are aware of the possible negative impact of their management, they are not ready to adapt their

techniques just for the sake of biodiversity. Therefore internal communication should focus on

the agronomic benefits of conservation actions, and the possibility to adopt management

strategies that are a real direct benefit for the farmer and which can also have a positive impact

on biodiversity. Example: Sowing legume fodder crops on fallow plots has several agronomic

benefits (erosion, N fixation etc.) and also provides pollen and nectar for bees and other insects.

Frequent communication (newsletters etc.) including positive testimony of farmers on actions is

a key factor. As for external communication it is clear that the positive contribution of the

farmers to biodiversity is always put forward, allowing attracting interest of potential clients

(through wine tourism and press information). Care should be taken to avoid greenwashing in

this stage, conservation actions should be done before farmers start selling green-labelled bottles

to their clients!

Motivation = Sustainability

From the projects listed in the table it seems that the most ‘efficient’ programs are projects

managed by the farmers but that have a multi-stakeholder approach, and only if the projects are

focussing on ‘positive’ contributions of the actions on the farming activities and economic

results. Nature conservationists and scientist implied in such programs should also take this into

account in order to prioritize actions that are acceptable by the farmers. Once this first step is

made, their ideas on landscape management are much easier adopted by the farmers.

In the Saumur-Champigny project the amount of hedgerows planted per year has tripled, starting

with farmers volunteering in the first two years (2-3km / year) and then changing to actively

recruiting farmers to plant hedges at previously identified sites to improve landscape

connectivity, in the last few years (7-9km/year).

The new French ‘national biodiversity strategy’ (MEDDTL, 2011) clearly shows that policy

makers have become aware that ‘awareness raising’ and motivation of local stakeholders is THE

key factor for efficient actions, our observation fully agree with this.

Therefore landscape project fit into the classic ‘sustainability schedule’ (Adams, 2006) seeking a

good balance between economic, social and environmental impact.

Clear positive results as far as biodiversity are not always readily obtained, but these projects

have a long term planning and goal. Therefore sustainability of the action is essential and

obtained only if economic constraints are taken into account (1) and if we also aim a social

impact (2), which is to improve territorial anchoring of local stakeholders.

Side effects: Landscape actions based on a ‘positive’ entry such as biodiversity often allow

tackling other environmental risk such as pesticide transfer through ‘biodiversity’ hedgerows

reducing spray drift. This can be a useful argument for financial partnerships.

Acknowledgements

Data provided and discussed with Philippe Bourdens, Céline Forget, Lucile Chedorge, Denis

Fetzman, Lucile Stanicka, Marie-Anne Simonneau

References Adams, W.M. (2006). "The Future of Sustainability: Re-thinking Environment and Development in

the Twenty-first Century." Report of the IUCN Renowned Thinkers Meeting, 29–31 January

2006. Retrieved on: 2009-02-16.

Billeter., R., J., Liira, D., Bailey, R., Bugter, P., Arens, I., Augenstein, S., Aviron, J., Baudry,

R.,Bukacek, F., Burel, M., Cerny, G., De Blust, R., De Cock, T., Diekötter, H., Dietz, J., Dirksen,

C., Dormann, W., Durka, M., Frenzel, R., Hamersky, F., Hendrickx, F., Herzog, S., Klotz, B.,

Koolstra, A., Lausch, D., Le Coeur, J. P., Maelfait, P., Opdam, M., Roubalova, A., Schermann, N.,

Schermann, T., Schmidt, O., Schweiger, M.J.M., Smulders, M., Speelmans, P., Simova, J.,

Verboom, W.K.R.E., van Wingerden, M., Zobel and P.J. Edwards (2008), Indicators for

biodiversity in agricultural landscapes: a pan-European study, Journal of Applied Ecology, 45:

141–150.

Duelli, P. and M.K. Obrist (2003), Regional biodiversity in an agricultural landscape: The contribution

of seminatural habitat islands, Basic and Applied Ecology, 4:129-138.

MEDDTL (2011) Stratégie nationale pour la biodiversité 2011-2020.

http://www.developpement-durable.gouv.fr/IMG/pdf/SNB_2011-2020WEB.pdf

Porte, B., Rochard, J., van Helden, M., Guenser, J., Fulchin, E., 2011. GIS for planning conservation

actions in viticulture landscapes, IOBC Bulletin. this meeting.