Independent Project at the Department of Earth Sciences Självständigt arbete vid Institutionen för geovetenskaper

2017: 12

Meteorological Conditions on Nordenskiöldbreen Glacier,

Svalbard (2009 – 2015) Meteorologiska förhållanden på glaciären

Nordenskiöldbreen, Svalbard (2009 – 2015)

Niclas Bergman

DEPARTMENT OF EARTH SCIENCES

I N S T I T U T I O N E N F Ö R

G E O V E T E N S K A P E R

Independent Project at the Department of Earth Sciences Självständigt arbete vid Institutionen för geovetenskaper

2017: 12

Meteorological Conditions on Nordenskiöldbreen Glacier,

Svalbard (2009 – 2015) Meteorologiska förhållanden på glaciären

Nordenskiöldbreen, Svalbard (2009 – 2015)

Niclas Bergman

Copyright © Niclas Bergman Published at Department of Earth Sciences, Uppsala University (www.geo.uu.se), Uppsala, 2017

Sammanfattning Meteorologiska förhållanden på glaciären Nordenskiöldbreen, Svalbard (2009 - 2015) Niclas Bergman Svalbard täcks till ungefär 60% av snö och is och har ett landskap där glaciärer har en framträdande roll. I och med att vårt nuvarande klimat är inne i en fas av uppvärmning observeras dessa förändringar snabbast i de arktiska områdena.

Syftet med denna rapport var att undersöka hur olika meteorologiska parametrar förhåller sig till varandra samt deras påverkan på glaciären Nordenskiöldbreen och angränsande terräng. Mätdata togs från en automatisk väderstation placerad centralt på glaciären. Denna data innefattade vindhastighet och vindriktning, molntäcke, albedo, snödjup, temperaturförhållanden, solinstrålning och snödrift.

Vindförhållandena uppvisade tydliga säsongsbundna variationer med en dominerande nordöst-sydväst riktning på vintern med högre vindhastigheter och en väst-öst och sydväst-nordöst på sommaren med generellt lägre vindhastigheter. Det framkom att så kallade katabatiska vindar hade mer påverkan på de lokala vindförhållandena under vintern än sommaren.

Snödrift visade sig vara vanligt förekommande på Nordenskiöldbreen där vissa områden kunde vara helt snöfria även efter ett snöfall på grund av höga vindhastigheter i nära anslutning till snöfallet. Detta kan i sin tur leda till en temporär ökning i markens reflektionsförmåga (albedo) och på så sätt ha en lokalt avkylande effekt. Det framkom även att lätta snöfall på sommaren gav en markant ökning av markens albedo. Medelvärdet av albedo under den undersökta tidsperioden var för torr snö 0.8 och för is 0.3. För åren 2010-2014 undersöktes även när snösmältningen börjar, när all snö är borta och när sedan snöackumulationen påbörjas igen. Generellt kunde det observeras att snön ligger kvar cirka en månad innan smältning börjar.

Sammanfattningsvis kan sägas att de flesta av de olika undersökta meteorologiska parametrarna är på ett eller annat sätt beroende av varandra. Primära faktorer, såsom mängden av den inkommande solinstrålningen som absorberas eller reflekteras, styrs av hur tjockt molntäcket är vilket i sin tur påverkar markens temperatur och då också påverkar om det blir barmark, snö eller is. Nyckelord: Nordenskiöldbreen, albedo, vindhastighet, glaciär, Svalbard. Självständigt arbete i goevetenskap, 1GV029, 15 hp, 2017 Handledare: Ward van Pelt Institutionen för geovetenskaper, Uppsala Universitet, Villavägen 16, 752 36 Uppsala (www.geo.uu.se) Hela publikationen finns tillgänglig på www.diva-portal.org

Abstract Meteorological Conditions on Nordenskiöldbreen Glacier, Svalbard (2009 – 2015) Niclas Bergman Glacial environments in the Arctic are a much-studied topic as well as a field of research with strong influences regarding the current and future global climate of our planet. This report is focused on the meteorological conditions on Nordenskiöldbreen glacier from 2009-2015 and how they correlate with each other, the glacier surface and the surrounding terrain. With data gathered from an automatic weather station located at the centre of the glacier, a range of meteorological parameters is examined; wind direction and velocity, snow depth, cloud cover, incoming and reflected shortwave radiation, temperature deficit, albedo and drifting snow.

Seasonal differences were discovered, especially for wind direction and velocity where winds from the northeast occurred more frequently in the winter, indicating katabatic winds, whereas winds from the west and southwest were more pronounced in the summer. The calculated temperature deficit shows that katabatic winds blow down-glacier under stably stratified conditions and are shown to increase in strength with increasing temperature deficit (atmospheric temperature minus surface temperature. The mean albedo at Nordenskiöldbreen during this period is within the expected limits, 0.8 for snow and 0.3 for ice and the cloud cover was 0.58. Additionally, it could be observed that the occurrence of dry, drifting snow is present in the winter season as snow depth shows pronounced drops during high-wind events in winter.

Overall, it is concluded that most of the examined parameters correlate and need each other to function and act as mechanisms within the cryosphere and as such it is crucial for scientists to understand their connected relationships when attempting to study global climate changes. Key words: Nordenskiöldbreen, albedo, glacier, Svalbard, windspeed Independent project in Earth Science, 1GV029, 15 credits, 2017 Supervisor: Ward van Pelt Department of Earth Sciences, Uppsala University, Villavägen 16, SE-752 36 Uppsala (www.geo.uu.se) The whole document is available at www.diva-portal.org

Table of Contents 1. Introduction ............................................................................................................. 1

2. Study site ................................................................................................................ 2

3. Data and Methods .................................................................................................. 2

3.1 Automatic Weather Station ................................................................................ 3

3.2 Technical malfunctions and data reliability ........................................................ 4

4. Data processing ...................................................................................................... 4

4.1 Albedo ............................................................................................................... 4

4.2 Cloud cover ....................................................................................................... 4

4.3 Temperature deficit ............................................................................................ 4

5. Results and Discussion .......................................................................................... 5

5.1. Wind characteristics ......................................................................................... 5

5.1.1 Katabatic winds ........................................................................................... 6

5.1.2 Wind velocity and wind direction ................................................................. 6

5.1.3 Seasonal variability ..................................................................................... 6

5.1.4 A comparison with Kongsfjorden-Kongsvegen glacier ................................ 8

5.2 Shortwave radiation ........................................................................................... 9

5.3 Cloud cover ..................................................................................................... 10

5.4 Snow and ice ................................................................................................... 11

5.4.1 Albedo ....................................................................................................... 13

5.4.2 Drifting snow .............................................................................................. 15

6. Conclusion ............................................................................................................ 17

Acknowledgements .................................................................................................. 17

References ............................................................................................................... 17

1

1. Introduction Increasing global temperatures are affecting a wide variety of environmental and biological areas (Committee on Ecological Impacts of Climate Change, 2008). Even seemingly small variations in average temperatures may have significant impact on the intricate systems and components of the geochemical model, including the cryosphere. It is therefore vital that continuous studies of climate change and their related consequences are prioritized by governments and other policy makers. For the northern hemisphere, the Arctic Monitoring and Assessment Programme (AMAP) is continuously keeping track of rates of pollution as well as climate change in the Arctic region. One of their published reports from 2011 demonstrate that the area is currently experiencing a warming above the global average, with Arctic glaciers diminishing in size at an alarming rate.

Over the past half century, this ice melt has been the source of about 30% of the observed eustatic sea-level rise (Church et al., 2011). This is evidential for the connection between sea-level change and mass balance of glaciers and ice caps. If all glaciers and ice caps, outside of the Antarctic and Greenland ice sheets, would melt it would result in a 0.5 m increase in global sea-level (Wang et al., 2005). Model simulations predict that Arctic glaciers will continue to melt at an increasing rate due to anthropogenic influences for at least the 21st century (Meier et al., 2007; Wang et al., 2005).

Svalbard is a Norwegian archipelago located between latitude 77-81°N covering an area of 61,022 km2. It is one of the world’s northernmost inhabited areas and is thus sparsely populated, with the main settlement Longyearbyen supporting some 2000 inhabitants (van Pelt, 2013). Svalbard has an Arctic climate which is milder than other areas at the same latitude, due to the warm flow of the North Atlantic Current. Due to its northern latitude, Midnight Sun is present from April 20 until August 23 and the Polar Night from October 26 until February 15 (Norwegian Polar Institute). Average summer temperatures on Svalbard at sea-level are between 4 to 6 °C and average winter temperatures range from -12 to -16 °C (Norwegian Meteorological Institute). Nevertheless, current temperature measurements indicate that Svalbard is experiencing the warmest conditions since the start of the measurements in 1975, with each of the past 73 months being warmer than average (Anderssen, 2016; Ritter, 2016). Since the Little Ice Age, the mean annual temperature in the central Spitsbergen has increased by 4 °C (Førland & Hanssen-Bauer, 2003).

The archipelago is largely covered by glaciers and ice caps and the first complete glacier inventory, Glacier Atlas of Svalbard and Jan Mayen (Hagen, 1993) estimated that some 60% of the area is covered, which in turn would represent ~6% of the total number of global glacier covered areas (Van Pelt, 2013). However, more recent studies of the inventory of Svalbard glaciers, such as Nuth et al., 2013, identify a 7% decrease in glacial cover over the last three decades.

The aim for this project is to analyse if a correlation is present between several meteorological parameters at Nordenskiöldbreen glacier and how they influence the glacier surface as well as the surrounding terrain. The examined data consist of wind, precipitation, radiation, temperature and cloudiness and is gathered from automatic weather station (AWS) measurements on the glacier during the period of March 2009 to April 2015.

2

2. Study site As the Little Ice Age came to an end in the late 19th century a rapid warming took place over the course of the following half-century until temperatures declined and reached a minimum in the 1960s before beginning to increase again to the levels present today (Isaksson et al., 2001). The aftereffects of this change in global temperature is believed to be a strong influence on why most of the Svalbard glaciers are receding. Among those glaciers are Nordenskiöldbreen, a tidewater glacier with an area of approximately 242 km2 stretched out over 26 km (Stacke et al., 2013) in a northeast to southwest fashion as can be seen in Figure 1. Its outlet is fed by the 600 km2 Lomonosovfonna ice plateau in the NE with the foremost flow path passing between De Geerfjellet and Terrierfjellet, traversing the surface from a top elevation of 1237 m a.s.l, to Adolfbukta at sea-level in the southwest. Although the glacier is experiencing a rapid frontal retreat, an actively calving front of some 3 km is present (van Pelt, 2013). The annual mean surface velocity of Nordenskiöldbreen is between 50-60 m.a-1 (Den Ouden et al., 2010). The surface being exposed beneath the quickly retreating glacier (mean average linear retreat rate of 35 m.a-1) is strongly associated with ongoing glacial and glaciofluvial processes (Stacke et al., 2013).

Previous studies on Nordenskiöldbreen present the surface mass balance (Van Pelt et al., 2012, 2014), subsurface thermal and stratigraphic conditions (Marchenko et al., 2016, 2017), ice thickness (Van Pelt et al., 2013) and ice core chemistry and stratigraphy (e.g. Vega et al., 2016).

Figure 1. Nordenskiöldbreen is a glacier positioned in central Spitsbergen on Svalbard. Its centre is roughly at 78°.69’N, 17°.16’E. Topographic rock formations form a western boundary (De Geerfjellet) for the glacier and more centrally located is Terrierfjellet with Ferrierfjellet further to the south. The red dot on the map indicates the location of the Automatic Weather Station (AWS) from which data was collected (Norwegian Polar Institute).

3. Data and Methods The data used in this report is provided by continuous measurements from an Automatic Weather Station (AWS) located in the centre of the glacier at an altitude of

AWS

N W E

S

3

~524 m a.s.l. It was deployed there in March 2009 by the Institute for Marine and Atmospheric research Utrecht (IMAU) (van Pelt, 2013).

To make the extracted raw data values from the datalogger on the AWS easier to process, the information was compiled into datasets usable in MATLAB. With the help of this computer software, it was then possible to create informative plots illustrating correlations between different meteorological variables.

3.1 Automatic Weather Station The AWS is a monitoring instrument capable of measuring and storing several different meteorological data. This report is focused on the following in situ data components.

• Snow depth • Shortwave radiation (incoming and reflected) • Wind speed and wind direction • Air temperature

Figure 2. The automatic weather station on Nordenskiöldbreen (photo taken by Carleen Reijmer).

Subsequent calculations were made to attain additional values of cloud cover, albedo, surface temperature and temperature deficit. Incoming and outgoing longwave radiation data was used for calculating cloud cover and temperature deficit.

Gathering dependable and continuous data on Arctic glaciers can be problematic due to unpredictable weather conditions as well as the inaccessibility of some of them. Because of the convergence of cold air from the north and mild, humid air from the south, weather conditions can change quickly and often, with occasional strong

4

winds and wind chills of around -30 °C in the winter (Norwegian Polar Institute, 2017). These rapidly changing weather conditions make it more challenging to monitor the glaciers and poor accessibility of glacierized terrain further complicates data collection.

3.2 Technical malfunctions and data reliability During the time that the AWS was recording data, some technical issues occurred. Between fall 2012 and spring 2013 a short-circuit of the equipment caused the battery to malfunction thus causing all sensors to go offline. Additionally, accumulation of ice on the wind and radiation sensors, a process known as riming, is known to potentially cause inaccuracies in the data from automatic weather stations. In some instances, the data from the AWS require an extra layer of processing before being used in plots. For example, reflected shortwave radiation (SW) readings from the sensor pointed downward can be larger than from the sensor measuring direct, incoming SW radiation (sensor directed upwards). The reason for this can either be that the upward looking sensor is covered by snow or that a low sun angle, common at these latitudes, will cause the downward looking sensor to register radiation not only from the surface but also from the sky (van den Broeke et al, 2004). To mitigate these issues the shortwave data used in this report was processed so that values <5 W m-2 were considered as unreliable and were therefore excluded.

Data from 2009 and 2015 are largely omitted due to not being completed annual cycles.

4. Data processing 4.1 Albedo As mentioned earlier, the raw data need be processed to calculate albedo, α. This is done by dividing the measured outgoing shortwave radiation with incoming SW radiation, α = SWout/SWin (Benn & Evans, 2010).

When attempting to calculate albedo from the incoming and reflected shortwave radiation, the low angle of the sun created large diurnal differences towards the beginning and end of the polar summer. It was therefore needed to set a specific time for when to calculate the albedo. This was set to 12:00 (noon).

4.2 Cloud cover The AWS data did not provide any direct observations of cloud cover. However, it was possible to indirectly estimate cloud cover from observed incoming longwave radiation and air temperature measurements following a procedure described by Kuipers Munneke et., al (2010).

4.3 Temperature deficit Temperature deficit is the difference between the air temperature (recorded at around 3 m above the surface) and the surface temperature. However, the surface temperature is not observed directly, but can be estimated from the observed outgoing longwave radiation. To do this, the Stefan-Boltzmann law is used: LWout = sigma*Tsurf4, where LWout is the outgoing longwave radiation, sigma is the Stefan-Boltzmann constant and Tsurf is the surface temperature (De Bruin & Holtslag 1988).

5

5. Results and Discussion 5.1. Wind characteristics The following section will present wind velocities and wind direction as well as local wind features on Nordenskiöldbreen and how these meteorological parameters may differentiate from season to season.

Figure 3. Seasonal wind directions and wind velocities. The seasons are divided into spring, March, April, May (MAM); summer, June, July, August (JJA); autumn, September, October, November (SON) and winter, December, January, February (DJF). The percentage values on the angle range (degrees) of the bars are indicative of the frequency of wind originating from that wind direction range.

6

5.1.1 Katabatic winds Katabatic winds are known to form during night-time when a radiative cooling effect occurs as high elevation air begins to descend beneath warmer, less dense air due to the gravity force. The dense air layer is created by the intense surface cooling, which means that this relatively thin layer is stably stratified (Esau & Repina, 2012). This phenomenon is enhanced during winter over snow covered surfaces and after dry, clear nights. Wind velocities are commonly low (~5 m/s) in many areas but if the slope is steep and the depth of cold air large, velocities may increase (Kumar, 2011).

Katabatic winds are especially strong in Antarctica, where speeds of up to 40 m/s is not uncommon, due to the large elevation difference and the large distance between the icy inland with altitudes beyond 4000 m and the shore (Parish & Cassano, 2001). 5.1.2 Wind velocity and wind direction The first observation to make from the wind rose diagram shown in Figure 3 is that the most dominant wind direction is the one originating from the northeast, and this is true for all seasons of the year. This would be well in line with the expected behaviour of katabatic winds as they characteristically blow down glacier and in the case of Nordenskiöldbreen that would be sloping from northeast towards southwest. However, although the dominant wind direction is clearly from the northeast, it is not the only direction present. Higher wind velocities (>10 m/s) are more commonly originating from the east but both northwest and southwest show quite significant wind peaks although with mostly lower wind velocities than that from the east.

Furthermore, the most frequent wind velocities (illustrated with blue and dark blue colours) may be observed as being of lower velocity (<5 m/s).

The annual mean wind velocity for the examined period was 4.6 m/s. 5.1.3 Seasonal variability When looking for differences and similarities in Figure 3 regarding wind velocity and wind direction per season, it is obvious that there exists a large divergence between the summer months (June, July, August) and the winter months (December, January, February), where spring and autumn largely act as transitional periods for wind velocities and wind direction frequency to reach minimum and maximum values.

Summer and winter, being at opposing ends from each other, have clear differences in frequency of low and high wind velocities. In the summer season, the wind speeds are principally of lower velocity, <5 m/s, with roughly half of the winds from the northeast being <2.5 m/s. On the other hand, for the winter season there is a significantly larger amount of higher wind velocities, especially in the range of 5-7.5 m/s from the northeast and 10-15 m/s from the east.

It is not only wind velocity that is demonstrating a different pattern between summer and winter but also the wind direction. Both the directions of NW as well as SW have more frequent winds in the summer season, showing twice (SW) and three times (NW) the amount of wind frequency than that of the corresponding directions in the winter season. However, the winter season have a significantly larger increase in high velocity winds originating from the east.

To examine why these two seasons show such radical differences in wind velocity and direction it is needed to take a step back and look closer at a few components mentioned earlier in this report.

7

First, the topography as shown in Figure 1 appear to be a strong influence on how winds originating from certain directions behave in the area. Katabatic winds travelling northeast to southwest down the flowline of the glacier are mostly channelled into a tunnel shaped area between De Geerfjellet, MaCabefjellet and Flemingfjellet in the west and Terrierfjellet in the east, all four with elevation peaks of ~1000 m. An interesting observation can be made here and directly linked to the northwest wind direction, and that is the mountain passage between Flemingfjellet and McCabefjellet. This opening (Fig. 1) is most likely the reason why there is a “peak” in wind frequency and wind velocity towards the northwest as shown in the summer season of Figure 3. It can be assumed that the decrease in elevation provided by this passage allow west-easterly winds traversing the flat neighbouring glacier of Ebbabreen to be funnelled through the opening and increase their velocity, thus resulting in significant AWS recordings.

Furthermore, in summer katabatic winds are less dominant due to air and surface temperature changes thus giving way to large-scale wind systems to influence observed wind characteristics at the weather station. This makes the northeast direction less dominant and the other directions more dominant.

For the eastern direction in the winter season, large scale wind systems are likely to occur and be recorded as there is less topography blocking their path from that direction (east) in contrast to the west. The lack of topography would also be a contributing factor for the winds originating in the southwest as they travel across Adolfbukta and up the glacier. In winter, the katabatic winds are stronger, which means that the likelihood of winds coming from any other direction becomes smaller.

Lower wind speeds from the northeast in the summer season is an indicator that the atmosphere is in a more stable condition during this time of the year (less storm related events) resulting in decreased turbulence and mixture between the atmosphere and the surface. This would on its own be an indicator for the development of katabatic winds as a stably stratified atmosphere would induce katabatic wind flow. However, in the summer, increasing amounts of shortwave radiation from the sun heats the surface, weakening the stratified air layers which reduces the temperature difference between the atmosphere and the surface. As a result, katabatic winds are less likely to develop in summer.

In winter, there are long periods with no incoming solar radiation, which causes the surface to be much colder, stimulating katabatic winds. This may be further observed in Figure 4 where the temperature deficit, i.e. the temperature difference between the surface and the atmosphere, is plotted. Figure 4 shows that high wind speeds during large-scale storms lead to vertical mixing of the boundary layer, which causes the temperature deficit to drop. On the other hand, during calm large-scale conditions, katabatic winds can develop, which tend to become stronger with increasing temperature deficit (see the black cloud in Fig. 4, right panel).

8

Figure 4. This temperature deficit plot demonstrates the varying concentrations of wind velocities relative to certain directions as well as amount of incoming SW radiation. 5.1.4 A comparison with Kongsfjorden-Kongsvegen glacier Previous glacier studies in Svalbard have discussed similar occurrences as mentioned above regarding glacier wind characteristics, supporting the notion of consistent wind channelling features and the relevance of katabatic winds for the overall wind climatology in the region. However, in the research article by Esau and Repina (2012) the influence of katabatic winds as the main wind driving mechanism for valley glaciers is questioned.

Their study of the wind climate of the Kongsfjorden-Kongsvegen glacier (hereafter referred to only as KK) valley in western Svalbard shed light on interesting aspects, some of which show similarities to the conditions supported by the AWS data gathered at Nordenskiöldbreen.

With the data from Nordenskiöldbreen being derived purely from an AWS, the KK location used additional observations from the Integrated Global Radiosonde Archive (IGRA). This data made it possible to analyse wind climate over a longer period as well as at higher elevations through measurements of a vertical depth of the wind from the surface to a set of chosen atmosphere elevation levels (pressure gradients).

As with what was supported by the data from Nordenskiöldbreen regarding seasonal wind behaviour, the KK area displayed the typical characteristics of a valley glacier with the strongest surface winds for all months channelling along the glacier axis (northeast to southwest) accompanied by a less distinct wind flow during the summer months.

9

Nevertheless, the authors were critical that previous studies did not consider the; “strongest thermal circulations, namely the circulation caused by the horizontal difference in the land and sea surface temperatures” (p.3); and,

“The wind field in the valley is likely created by a complex interplay of different mechanisms where the commonly accepted katabatic wind mechanism may not be leading or even important. Presence of large mountain glaciers in the valley is frequently referred to as one of the reasons to invoke the katabatic wind mechanism for explanation of observed and simulated wind features” (p.3). Esau and Repina also questioned the use of a quantitative analogy between katabatic winds in mountain glaciers, as observed by Oerlmans and Grisogono (2002), to winds in Antarctica and Greenland due to the large difference in glacier length (~50 km for mountain glaciers). The issue of glacier length and wind speed was further emphasized by England and McNider (1993) as they; “derived a quadratic asymptotic estimation of the maximum katabatic wind speed in the neutrally stratified atmosphere as a function of the glacier length” (p.3, Esau & Rapina). This would mean that England and McNider’s theory links temperature deficit (Fig. 4) with the maximum katabatic wind speed of the location, and according to their formula, katabatic wind speeds for Svalbard glaciers should be <5 m/s. This wind velocity can be partly supported by Figure 4 for Nordenskiöldbreen where a positive trend is shown near the 5 m/s value with a large concentration of darker coloured dots (representative of the 0-50° range) from the direction of ~40° (northeast), indicating winds blowing down glacier.

Additionally, Esau and Repina argue that one parameter that has been neglected in previous Arctic studies is the thermally driven land-sea breeze circulation caused by horizontal temperature difference between the open fjord and the glacier. Local topography is likely to increase this difference and thus reinforce the circulation. This is probably less pronounced for Nordenskiöldbreen than for KK since the former is situated in a more protected area inland.

5.2 Shortwave radiation Particles (photons) and waves are the two ways in which all materials radiate electromagnetic energy, with cooler materials radiating longer wavelengths and hotter materials shorter wavelengths (Benn & Evans, 2010). The driving force, and primary source of energy for our climate and any biological process on the planet is the sun and more specifically, the shortwave (SW) radiation emitted from the sun. SW radiation and longwave radiation are the most significant mechanisms of the energy balance on most glaciers. Other energy fluxes comprise the sensible heat flux, latent heat flux, ground heat flux and the heat supplied by rainfall. The amount and strength of radiation reaching the surface is influenced by the optical depth of the atmosphere, which is a function of distance, humidity, cloudiness and concentration of other particles. Additionally, the grid orientation, solar elevation angle and shading are important factors that affect incoming solar radiation (Benn & Evans, 2010).

SW radiation at Nordenskiöldbreen demonstrates expected behaviour for a location at these latitudes where the appearance of the sun is both strongly seasonal over a whole year Midnight Sun in the summer and Polar Night in the winter. This cyclic behaviour can be observed in Figure 5 where the white gap between the “bell-

10

shapes” of the plot indicate winter season with no sun and the small white “bumps” are due to continuous incoming SW radiation in the summer, resulting in SW radiation never reaching 0. The incoming radiation is always higher than the reflected as some radiation is absorbed by the surface causing a warming effect while the rest is reflected, depending on the albedo of the surface.

Figure 5. The upper figure illustrates the annual cycle of incoming and reflected shortwave radiation. The bottom plot is an example for a typical year from April to September, where incoming SW is larger than reflected SW.

5.3 Cloud cover The mean cloud cover for the examined period was calculated to 0.58, where clear sky conditions would have a value of 0 and cloudy or overcast sky 1. This value on its own is not revealing any specific trend seen over the whole period. However, Malecki (2015) examined the cloud cover of the neighbouring glacier Ebbabreen from 2008-2010 and found that the mean value for cloud cover at this location was 0.78 over the 3-year period. This is a fairly large difference, suggesting that two locations, although in close proximity to each other, can experience quite different weather conditions in terms of cloudiness perhaps due to topographic boundaries. Cloud cover is plotted against incoming SW radiation in Figure 6 to illustrate how they correlate with each other. With low cloud cover values corresponding to higher amounts of incoming SW (in the summer) and vice versa.

11

Figure 6. Cloud cover is directly related to the amount of SW radiation able to enter the atmosphere and reach the surface. The top panel illustrate the cloud cover and SW radiation over a year (2014), with high SW values in the summer and low to none in fall, spring and winter. The bottom panel provides a more detailed view on the actual relation between the two variables.

5.4 Snow and ice Figure 7 illustrates the annual cycle of snow accumulation and snow melt for the examined period. The ladderlike steps indicate where the AWS malfunctioned and no data was recorded.

Dates for when the snow melt starts (Table 1) were taken when a rapid snow depth drop coincides with above zero temperatures. Furthermore, the snow accumulation date was estimated as the date after which the snow depth (for that year) did not reduce to zero anymore, and snow events were counted if the snow cover was <3 cm and if it subsequently dropped to 0 cm after the event.

Looking at Table 1, it can be observed that for all years except 2012, the snow remained on the surface for about one month counted from the start of the snow melt (in mid to late June) until the snow depth reached 0 (early to mid-July).

Further down on the glacier, ice will be exposed earlier on in the summer season thus lowering the albedo in that area as the snow will gradually melt away more quickly in the spring.

12

Figure 7. Observed annual cycle of snow accumulation and snow melt from 2009-2015.

Table 1. Snow and snowfall during 2010-2014 on Nordenskiöldbreen.

1Missing data January 1 – March 20 2Missing data October 29 – April 23

Year Snow amount (cm) Jan 1

Snow melt date

No snow Snowfall event

Snow accumulation date

2010 N/A1 June 19 July 19 None September 26

2011 ~20 June 13 July 15 August 13, September 21

September 20

2012 ~40 June 5 August 9 September 12

N/A2

2013 N/A2 June 20 July 9 August 14, September 6

October 1

2014 ~30 June 27 July 18 None September 20

13

5.4.1 Albedo Albedo is a measure for reflectance or optical brightness (Coakely, 2003) and is in this report applied to properties of a snow or ice-covered surface. The amount of incoming shortwave radiation being absorbed for a specific wavelength is termed ‘spectral albedo’ and several previous studies have been carried out regarding how different wavelengths act regarding certain parameters, e.g. sun angle, snow age and amount of existing impurities or “whiteness” (Conway et al., 1996; Gardner & Sharp, 2010; Goelles et al., 2015). For the scope of this report, albedo as referred to the whole spectrum of solar radiation (i.e., “broadband albedo”), will be examined.

Summer snowfall events have a major impact on surface melt, as this snow temporarily increases surface albedo thus, for a time, reflecting a larger proportion of the incoming shortwave radiation which will have a local cooling effect. Table 1 together with Figure 8 provide information on snowfall events which are causing a significant increase in temporary albedo in the summer, observed especially clearly in 2011 and 2013. Figure 8 further illustrates the snow accumulation and snow melt and its correlation to albedo values over five years. The dates were set from April to September since it is only during these months that there is any incoming SW radiation, thus it is only then that albedo can be calculated.

Looking at Table 1 and Figure 8 it can be determined that the snow depth is not an important factor for readings on albedo values. A snow depth of ~3 cm (August 13th 2011) result in a near identical albedo of approximately 0.8 as a snow depth of ~14 cm (August 14th 2013). However, snow presence is important, since not much snow is needed to raise the albedo to typical snow albedo values.

The mean albedo for snow at Nordenskiöldbreen was calculated to 0.80 for dry snow and 0.30 for ice. In Van Pelt et al. (2012) an ice albedo of 0.39 and snow albedo of 0.85 was found for Nordenskiöldbreen, so that matches quite well. For all years, albedo will drop below 0.50 at some point in July (Fig. 8). This value would then indicate that the snow is melting and exposing the underlying glacier ice.

As global temperatures have climbed steadily over the past decades, the Arctic have experienced the most significant temperature changes through the process known as Arctic amplification (Screen & Simmonds, 2010). This feedback process is generated by the melting and disappearance of sea ice and seasonal snow cover as previously reflective white surfaces are converted to darker (dirtier) ice, ocean water or vegetation. These dark surfaces absorb more solar radiation, leading to higher air temperatures which leads to even more rapid melting, and so on.

14

Figure 8. Albedo and snow depth April to September 2010-2014

15

5.4.2 Drifting snow One occurrence that may influence the recordings for snow depth on glaciers as well as the albedo are drifting snow events. These take place when snow is eroded, transported and deposited by wind thus creating snow depths at certain locations that would not be indicative of the general snow-cover in the area.

However, for these events to occur, there are some conditions which must be present; a minimum wind speed (threshold) for initiating or sustaining saltation of snow, along with fresh, non-compacted snow precipitated within a specific timeframe. The longer the window between the snowfall and potential snow erosion, the more time the snow will have to settle, thus increasing the static threshold wind speed at which particles start moving. Dry snow becomes mobile at wind speeds of 7.7 m/s and is defined as snow that has not received temperatures of 0 °C, or wet precipitation since the last snowfall (Li and Pomeroy, 1997) and as such it is important to consider the innate properties of the snow, its crystalline structure affecting the cohesion as well as the wind threshold and time-frame for when snow drift events may be expected to happen.

As mentioned earlier, if the wind velocities are too low, there will not be enough force to re-distribute the snow. If a longer period pass between the snowfall and the high enough wind velocities, the snow is likely to become too compacted, potentially developing a snow crust, making snowdrift impossible. Additionally, if the snow become wet, i.e. if the surface temperature rises above 0 °C shortly after a snowfall, the cohesion of the snow crystals is drastically increased due to thin layers of liquid water on the crystals (Li and Pomeroy, 1997). This is more likely to occur in the summer months when temperatures more frequently remain above 0 °C.

Furthermore, it may be assumed that the drifting behaviour is different over a snow-covered surface than over ice. This was seen by Bintanja and Reijmer (2001) in their studies of similar snowdrift conditions on Antarctica, where snowdrift over blue ice commonly took place even during very low wind velocities (low threshold). He concluded that this was due to the characteristics of the snow particles; they were not able to adhere to the ice very well.

For Nordenskiöldbreen, drifting of snow is a prominent landscape feature as there are many occasions during the winter season when the snow depth is decreased, even to the extent of exposing bare ground or ice. This can be illustrated by Figure 9 and 10 where high wind speeds re-distribute the snow in such way that the snow depth for a period decreases to zero (Fig. 10). This wind drift of snow most often coincides with high-wind events.

16

Figure 9. Drifting snow event in March 2011 where high wind velocities reduce the snow cover.

Figure 10. Drifting snow event starting on October 19th with a snowfall (wind velocities below 8 m/s). As the wind increases to a maximum of ~26 m/s, the snow depth is reduced to only a few cm. With wind velocities once again dropping, the snow depth increases. Subsequently, the wind velocity increase yet again, this time reducing the snow cover to 0 on October 23rd.

17

6. Conclusion In this report, I studied meteorological and surface conditions at a weather station site on the glacier Nordenskiöldbreen, Svalbard, between 2009 and 2015. A set of parameters has been analysed including wind speed, wind direction, solar radiation, albedo, cloud cover, snow depth and snow drift.

Acting as an overarching conclusion in this report is the notion that that the factors shaping glacial environments correlate and are very much connected to the cryosphere system where they are inextricably entwined. In other words, there is a strong connection between a glacier’s microclimate, surface snow/ice conditions and the surrounding topography. Therefore, changes occurring in one primary parameter will be chained to a secondary and possibly a third (e.g. shortwave radiation, cloud cover and albedo) thus having the potential to alter the balance within the system. For future studies, it would be interesting to further examine seasonal wind velocities and wind directions along with temperatures at a vertical depth further from the surface at Nordenskiöldbreen, potentially also over open water. This could be done to investigate if a sea-land breeze is present and how wind velocities at higher altitudes influence cloud cover at the location.

Acknowledgements I would like to direct my thanks to my supervisor, Ward van Pelt, for his invaluable guidance and support throughout the process of writing this report as well as for being patient when answering all of my questions.

References AMAP (2017). Snow, Water, Ice and Permafrost. Summary for Policy-makers. Arctic Monitoring and Assessment Programme (AMAP), Oslo, Norway. Bintanja, R. and Reijmer, C.H. (2001). Meteorological conditions over Antarctic blue

ice areas and their influence on the local surface mass balance. Journal of Glaciology, 47(156), pp.37-50.

Benn, D. and D.J. Evans (2014). Glaciers and glaciation. Routledge. Church, J.A., White, N.J., Konikow, L.F., Domingues, C.M., Cogley, J.G., Rignot, E., Gregory, J.M., van den Broeke, M.R., Monaghan, A.J., Velicogna, I. (2011) Revisiting the Earth’s Sea-Level and Energy Budgets from 1961 to 2008. Geophysical Research Letters, 38, no. 18. doi:10.1029/2011GL048794. Coakley, J.A., 2003. Reflectance and albedo, surface. Encyclopedia of the Atmosphere, pp.1914-1923. Elsevier Science. Conway, H., Gades, A. and Raymond, C.F. (1996). Albedo of dirty snow during

conditions of melt. Water resources research, 32(6), pp. 1713-1718. doi:10.1029/96WR00712

Den Ouden, M., C. Reijmer, L. Pohjola, R. van de Wal, J. Oerlemans, and W. Boot. (2010). Stand-alone singlefrequency GPS ice velocity observations on Nordenskiöldbreen, Svalbard, The Cryosphere, 4(4), pp. 593–604. doi: 10.5194/tc-4-593-2010

Esau, Igor, and Irina Repina. (2012). Wind Climate in Kongsfjorden, Svalbard, and Attribution of Leading Wind Driving Mechanisms through Turbulence-Resolving Simulations. Advances in Meteorology. doi:10.1155/2012/568454.

18

England, D.E. and McNider, R.T. (1993). Concerning the limiting behavior of time dependent slope winds. Journal of the atmospheric sciences, 50(11), pp.1610-1613. doi: 10.1175/1520-0469(1993)

Førland, E.J. and Hanssen‐Bauer, I. (2003). Past and future climate variations in the Norwegian Arctic: overview and novel analyses. Polar research, 22(2), pp.113-124.

Gardner, A.S. and Sharp, M.J. (2010). A review of snow and ice albedo and the development of a new physically based broadband albedo parameterization. Journal of Geophysical Research: Earth Surface, 115(F1).

Gölles, T., Bøggild, C.E. and Greve, R. (2015). Ice sheet mass loss caused by dust and black carbon accumulation. The Cryosphere, 9(5), pp.1845-1856. doi:10.5194/tc-9-1845-2015

Holtslag, A.A.M. and De Bruin, H.A.R. (1988). Applied modeling of the nighttime surface energy balance over land. Journal of Applied Meteorology, 27(6), pp. 689-704. doi: 10.1175/1520-0450(1988)

Isaksson, E., Pohjola, V., Jauhiainen, T., Moore, J., Pinglot, J.F., Vaikmaäe, R., van de Wal, R.S., Hagen, J.O., Ivask, J., Karlöf, L. and Martma, T., 2001. A new ice core record from Lomonosovfonna, Svalbard: viewing the 1920-97 data in relation to present climate and environmental conditions. Journal of Glaciology, 47(157), pp. 335-345. Kuipers Munneke, P., Reijmer, C.H. and Van den Broeke, M.R. (2011). Assessing the retrieval of cloud properties from radiation measurements over snow and ice. International journal of climatology, 31(5), pp. 756-769. doi:10.1002/joc.2114 Kumar, A. (2011). Katabatic Wind: In Relation With Snow and Glaciers.

Encyclopedia of Snow, Ice and Glaciers, edited by Vijay P. Singh, Pratap Singh, and Umesh K. Haritashya, pp. 671–72. Dordrecht: Springer Netherlands.

Li, L., and J. W. Pomeroy. (1997). Probability of occurrence of blowing snow, Journal of Geophysical Research., 102(D18). doi:10.1029/97JD01522 Małecki, J. (2015). Glacio− meteorology of Ebbabreen, Dickson Land, central

Svalbard, during 2008–2010 melt seasons. Polish Polar Research, 36(2), pp. 145-161. doi: 10.1515/popore−2015−0010

Marchenko, S., Pohjola, V.A., Pettersson, R., van Pelt, W.J.J., Vega, C.P., Machguth, H. , Bøggild, C.E. and Isaksson, E. (2016). A plot-scale study of firn stratigraphy at Lomonosovfonna, Svalbard, using ice cores, borehole video and GPR surveys in 2012-2014. Journal of Glaciology, 63, 237, pp. 67-78. doi:10.1017/jog.2016.118 Marchenko, S., van Pelt, W.J.J., Claremar, B., Machguth, H., Reijmer, C.H., Pettersson, R. and Pohjola, V.A. (2017). Parameterizing deep water percolation improves subsurface temperature simulations by a multilayer firn model. Frontiers in Earth Science: Cryospheric Sciences, 5, 16. doi:10.3389/feart.2017.00016 Meier, M.F., Dyurgerov, M.B., Rick, U.K., O’Neel, S., Tad Pfeffer, W.,Anderson, R.S., Anderson, S.P., Glazovsky. A.F. (2007). Glaciers Dominate Eustatic Sea-Level Rise in the 21st Century. Science, 317, no. 5841 pp. 1064–67. doi:10.1126/science.1143906. National Research Council. (2008). Ecological Impacts of Climate Change. Washington, DC: The National Academies Press. doi: 10.17226/12491 Nuth, C., Kohler, J., König, M., von Deschwanden, A., Hagen, J. O., Kääb, A.,

Moholdt, G., and Pettersson, R. (2013). Decadal changes from a multi-temporal glacier inventory of Svalbard, The Cryosphere, 7, pp. 1603-1621, doi:10.5194/tc-7-1603-2013.

Oerlemans,J.,and B.Grisogono. (2002). Glacier winds and parameterisation of the

19

related surface heat fluxes,Tellus A, 54(5), pp. 440–452. doi:10.1034/j.1600 0870.2002.201398.x Parish, T.R., and Cassano, J.J. (2003). The Role of Katabatic Winds on the Antarctic Surface Wind Regime. Monthly Weather Review, 131, no. 2 pp. 317–33. doi:10.1175/1520-0493 Screen, J.A. and Simmonds, I. (2010). The central role of diminishing sea ice in recent Arctic temperature amplification. Nature, 464(7293), pp. 1334-1337. doi:10.1038/nature09051 Stacke, V., Mida, P., Lehejček, J., Tóthová, G. and Nývlt, D. (2013). Recent landscape changes in terminoglacial area of the Nordenskiöldbreen, central Spitsbergen, Svalbard. Czech Polar Reports, 3, pp.3-6. doi: 10.5817/CPR2013-1-2 Van den Broeke, M., D. van As, C. Reijmer, and R. van de Wal. (2004). Assessing

and improving the quality of unattended radiation observations in Antarctica, Journal of Atmospheric and Oceanic Technology, 21(9), pp. 1417– 1431. doi: 10.1175/1520-0426(2004)

Van Pelt, W.J.J., J. Oerlemans, C.H. Reijmer, V.A. Pohjola, R. Pettersson and J.H. van Angelen. (2012). Simulating melt, runoff and refreezing on Nordenskiöldbreen, Svalbard, using a coupled snow and energy balance model.

The Cryosphere, 6, pp. 641-659. doi:10.5194/tc-6-641-2012 Van Pelt, W.J.J. (2013). Modelling the dynamics and boundary processes of Svalbard glaciers (Doctoral dissertation, Uitgeverij BOXPress). Van Pelt, W.J.J., J. Oerlemans, C.H. Reijmer, R. Pettersson, V.A. Pohjola, E. Isaksson and D.Divine. (2013). An iterative inverse method to estimate basal topography and initialize ice flow models. The Cryosphere, 7, pp. 987-1006. doi:10.5194/tc-7-987-2013 Van Pelt, W.J.J., R. Pettersson, V.A. Pohjola, S. Marchenko, B. Claremar and J. Oerlemans. (2014). Inverse estimation of snow accumulation along a snow radar transect on Nordenskiöldbreen, Svalbard. Journal of Geophysical Research: Earth Surface, 119, 4, pp. 816-835. doi:10.1002/2013JF003040 Vega, C.P., Pohjola, V.A., Beaudon, E., Claremar, B., van Pelt, W.J.J., Pettersson, R., Isaksson, E., Martma, T., Schwikowski, M. and Bøggild, C. E. (2016). A synthetic ice core approach to estimate ion relocation in an ice field site experiencing periodical melt; a case study on Lomonosovfonna, Svalbard. The Cryosphere, 10, pp. 961-976. doi:10.5194/tc-10-961-2016 Wang, L., Sharp, M.J., Rivard, B., Marshall, S. and Burgess, D. (2005). Melt season

duration on Canadian Arctic ice caps, 2000–2004. Geophysical Research Letters, 32(19). doi:10.1029/2005GL023962

Internet resources Norwegian Polar Institute (no date) Svalbard

http://www.npolar.no/en/the-arctic/svalbard/ [2017-04-20] Norwegian Meteorological Institute Climate statistics for Svalbard Airport observation site

https://www.yr.no/place/Norway/Svalbard/Longyearbyen/climate.html?spr=eng [2017-04-20]

Ritter, K. (2016). Svalbard sees ‘shocking’ temperatures near freezing point. https://phys.org/news/2016-11-svalbard-temperatures.html [2017-04-20]

20

Andersen, K.M. (2016). 2016: Warmest and wettest. http://svalbardposten.no/nyheter/engelsk/2016-warmest-and-wettest/19.7988 [2017-04-20]