Snow Cover and Climate Change on Cairngorm Mountain: A Report for the Cairngorms National Park Authority
Snow Cover and Climate Change on Cairngorm Mountain
A report for the Cairngorms National Park Authority
Mike Rivington (The James Hutton Institute)
Mike Spencer (SRUC)
28th April 2020
The James Hutton Institute
SRUC
Executive Summary
This report details research findings on the historical changes in snow depth and number of days of snow on Cairngorm Mountain and how snow may respond under climate change. It is an extension to the report ‘Snow Cover and Climate Change in the Cairngorms National Park: Summary Assessment’ produced by ClimateXChange in 2019¹.
Key Findings: Observed changes
• There has been a decrease in the observed maximum and average snow depth since the beginning of records (1983−84 winter). Maximum snow depth has declined by c. 10cm and the average by c. 3cm.
• There has been an observed decrease in the number of days when snow depth exceeds specific amounts. The largest decreases have occurred for shallower depths (>2<5, >5<10 cm) of c. 10 days since 1983.
• The mean snow depth per month has decreased in January and February since 1983. Depth per month has been highly variable but the observed trend has been downwards. Other months have different trends: March has had a slight decrease whilst November has been consistent and December a slight increase.
• For all months there is a clear increasing warming trend in observed maximum and minimum temperature between 1960 and 2019. The largest increases have occurred in April. The main snow fall months of January and February have had a relatively small increase in temperature.
• There has been an increasing trend of mean monthly precipitation amount for November, December and January since 1960, whilst March’s amount has decreased.
• There has been an increase in mean monthly solar radiation (MJ m² day¯¹) in February, March and April since 1994, implying greater heat energy input at the ground surface.
Key Findings: Future projections
• Likely to be a decline in snow cover days per year from the 2030s for Aviemore, the Cairngorm Chairlift meteorological station and Ptarmigan Restaurant on Cairngorm Mountain. This trend will continue through to the 2080s.
• There will be large variation between years and there are likely to still be some years comparable with past amounts of snow cover, but these will be less frequent.
• These findings are in line with results from the UK Meteorological Office and Inter-governmental Panel on Climate Change.
• Temperatures are projected to continue increasing, with a higher probability of having more days when the temperature is above a threshold of 2°C for snow formation.
• There is an increasing probability of more heat energy input on ground surfaces with an increasing snow melting affect.
• Snow is complex to model and project in the future, especially in temperate regions like Scotland with its strong maritime (Atlantic Ocean) climatic influence. Changes in seasonal variability will depend on how air flow over the UK (e.g. location of the jet stream) is affected by global scale ocean-atmosphere circulation processes. Our findings are a good indicator of future trends, but there remain substantial uncertainties at Cairngorm Mountain that need to be considered in making this a more detailed assessment of future snow cover.
Conclusions:
Warming will continue meaning snow cover and depth is likely to decrease on Cairngorm Mountain from the 2030s. There are likely to be some years with snow comparable to the past but overall there will likely be a decrease.
Contents
Introduction 5 Report structure 5 Summary of Cairngorms Snow Cover Report 6 Approach and Methods 7 Temperature, precipitation and solar radiation 7 Future snow modelling 8 Results 8 Observed Temperature, Precipitation and Solar Radiation trends 8 Temperature 11 Precipiation 11 Solar Radiation 11 Snow Depth 12 Snow Depth Days 14 Mean monthly snow depth 15 Results: Future Projections 16 Snow modelling 16 Translating estimates of future snow cover to snow depth 19 Conclusions 19 References 20 Appendix 20 Thermal Time Accumulation 29
Table of Figures
Figure 1. Days of lying annual average snow (1981−2010). Source: UK Met Office 6 Figure 2. January temperature 9 Figure 3. February temperature 10 Figure 4. Snow depth (cm) per winter by month, Cairngorm Chairlift meteorological station 12 Figure 5. Maximum snow and average snow depth (cm) at Cairngorm Chairlift meteorological station 13 Figure 6. Count of days at different snow depth ranges (2−5, 5 – 10, 10 – 15, 15 – 20 and +20 cm) per year, Cairngorm Chairlift 14 Figure 7. Mean monthly snow depth and trends, Cairngorm Chairlift (1982 – 2020) 15 Figure 8. Map of Cairngorm National Park and model grid cells 16 Figure 9. Annual snow cover (days per year) at three elevation ranges for Aviemore, the Cairngorm Chairlift (Base) and Ptarmigan visitor centre 17 Figure 10. November temperature 21 Figure 11. December temperature 22 Figure 12. March temperature 23 Figure 13. April temperature 24 Figure 14. Monthly mean maximum (Tmax) and minimum (Tmin) temperatures per year and trends (1982 – 2018) for Cairngorm Chairlift meteorological station 25 Figure 15. Cairngorm Chairlift location (1km resolution interpolated data) mean monthly precipitation (1960 ‑2018) 26 Figure 16. Mean monthly solar radiation (MJ m2 day‑1) and trends over time (1994 – 2017) 27 Figure 17. Daily snow depth (cm) per winter 28 Figure 18. Snow depth and observed thermal time accumulation per winter 30
Acknowledgements:
This report has been supported by the Cairngorms National Park Authority.
We would like to acknowledge the UK Meteorological Office for the use of the MIDAS observed data, the gridded observed weather data and the UKCP18 climate projections. The capability to use the climate data and bias correct it to higher spatial resolution has been developed under the Scottish Government’s RESAS Strategic Research Programme 2011 – 2016 and 2016 – 2021.
This report was prepared by:
Mike Rivington, The James Hutton Institute, Aberdeen.
Mike Spencer, SRUC, Edinburgh.
28th April 2020.
Please Cite as:
Rivington M and Spencer M (2020) Snow Cover and Climate Change on Cairngorm Mountain: A report for the Cairngorms National Park Authority. The James Hutton Institute, Aberdeen, UK.
Introduction
The spatial extent of snow cover, here defined as its duration (number of days per year snow is on the ground) and its quantity (depth) is an essential part of the ecology and hydrology in the Cairngorms National Park (CNP) and is a key factor influence winter sports and tourism activities. It also influences greenhouse gas emissions and sink potential from peatlands. Whilst there has been large inter-annual variation in the past, there are substantial concerns that, as a result of climate change, there may be significant decreases in snow cover, quantity and spatial extent, in the future.
This study explored the likelihood of these decreases in snow cover and depth in the future on Cairngorm Mountain. The aim was to assess the general past trends and plausible possible future snow cover. This study builds upon the work undertaken in 2019 to assess snow cover for the whole Cairngorms National Park. For full details on methods, results, key messages and caveats please refer to Rivington et al 2019.
There are many weather factors that determine the creation of snow. The UK’s heaviest snowfalls are associated with temperatures between 0 and 2°C. Above 2°C snowflakes will likely either not form, or if they do, melt and fall as sleet or rain. Some years have experienced large snowfalls, e.g. the winter of 2009⁄10 when a blocking high pressure system meant cold air from the north mixed with warm moist air from the Atlantic. This gave the coldest December on record since 1910². There are also many factors determining how long it snows for and what happens to it once on the ground (e.g. movement by wind, ground temperature). These are beyond the scope of this study and there are many uncertainties associated with projecting these other factors and the conditions they create in the future.
In this study we use a well-established assumption that snow cover is more correlated to temperature than precipitation, based on good evidence indicating temperature is a primary influencing factor as it influences formation and controls depth and duration (Harrison et al 2001, Beniston et al 2003, Trivedi et al 2007). A study at the Ben Lawers National Nature Reserve found that snow cover duration at mid to upper altitudes (600 – 900 m) responds most strongly to variation in mean daily temperature: a 1 °C rise in temperature can correspond to a 15-day reduction in snow cover at 130 m and a 33-day reduction at 750 m (Trivedi et al 2007).
Much of the analysis focusses on the drivers of changes in snow amount, primarily temperature but also precipitation and solar radiation, as there is adequate observed data for this, whereas there are issues of data quality and period of coverage for snow depth and areal coverage. Hence the results are best interpreted by combining an understanding of how the key drivers have changed up to now, and what the modelling of future conditions indicate.
Report structure
This report first presents the summary of the findings from the study of snow cover for the whole Cairngorms National Park area (Rivington et al 2019) and an overview of the methods used. Results are presented of analyses of observed weather data to identify historic trends in temperature, precipitation and solar radiation. These are important indicators of factors likely to influence the formation and accumulation of snow. For temperature we show future projections to 2080. The results of snow cover modelling under climate projections and conclusions are then provided. Supporting information is included in an appendix.
Days of Snow Lying Annual Average 1981 – 2010
Average Value (days)
60 40 to 60 30 to 40 20 to 30 10 to 20 5 to 10 <5
Figure 1. Days of lying annual average snow (1981−2010). Source: UK Met Office.
The weather station at the Cairngorm chairlift has the highest average number of days of snow falling, with snow falling on 76.2 days throughout the year (based on 1981 – 2010 averages), while the station at Aviemore records 66 days³.
Summary of Cairngorms Snow Cover Report
To assist in interpreting the results presented here, the following are the key findings from the CXC report on snow cover for the whole of the Cairngorms National Park¹.
1) There has been an overall decline in observed snow cover in the Cairngorms National Park (1969−2005). This trend conforms to those seen across other mountain areas and the Arctic and is in keeping with the observed global warming trend. However, some variability can also be seen with significant snow events and a possible increase in snow cover in the last decade. The overall declining snow cover trend is projected to continue and accelerate in the future. 2) A warming trend has been observed at meteorological stations (Balmoral) in the CNP since 1918 for both maximum and minimum temperature. There is variation between months: a) October and November show approximately 1.6°C + maximum temperature and 0.8 °C minimum temperature rises. This may influence the likelihood of when seasonal snow forms and cover becomes established. b) March, April and May show a warming trend indicating likelihood of earlier onset of snow melting. c) Precipitation (measured as rainfall and snow or ice) per month is variable between years with no strong trend observed. 3) There is a clear observed decrease in the number of days of snow cover at all elevation levels over the 35 winters between 1969⁄70 and 2004⁄05, with higher elevations having a larger proportional decrease (Snow Survey of Great Britain, Whitehillocks observing location, south- east CNP).
4) In the near-term, our estimates indicate the potential for a continuation of snow cover at the current range of variation, but with a substantial decline from the 2040s. These findings are in line with results from the UK Meteorological Office and Inter-governmental Panel on Climate Change (IPCC 2019). There will be some years in the future when the weather conditions create snow and enable lying snow that may be comparable to the past, but such occasions will become fewer. This applies to all elevations, but with larger proportional decreases at higher levels. Results indicate a likelihood of some years with very little or no snow by 2080.
Approach and Methods
We first assessed past trends using observed weather and snow depth records from the UK Meteorological Office to identify possible correlations between observed weather and snow depth trends. We then used climate model projections from the Met Office in a snow model to estimate future snow cover responses. Full details are available in the CXC report, including reviews of uncertainties and caveats.
An initial assessment of the data available from the UK Meteorological Office for the Cairngorm Chairlift site showed that the data for the count of days of snow, sleet or hail were either missing or of insufficient quality to be used in this study. Our analysis was therefore restricted to snow depth.
It is important to note that snow depth and cover are different, but with our current snow modelling projection capabilities we are not able to simulate snow depth. Hence there is need for interpretation on how much changes in snow cover can be correlated with changes in snow depth. The assumption is that more extensive snow cover implies greater depth and vice-versa.
For the future projections we used the UK Climate Projections 2018 (UKCP18) daily data for the RCP8.5 emission pathway (current emissions trajectory). The UKCP18 data is produced by a Regional Climate Model (HadRM3). This is run twelve separate times with variations in the model parameters that result in variations in the model estimates. This is done to capture the range of uncertainty in the parameters and provides a probabilistic range of possible future climate conditions. Each of the 12 model runs is referred to as an ensemble member. The ensemble mean is the average across the 12 members and represents the 50% (mid-range) probability level.
These are the only daily data released (the snow model used here needs daily data), hence this is just one possible future scenario. Scenarios with lower greenhouse gas emissions may reduce the likelihood of snow cover loss, but the world is locked into some global warming already in the next 30 – 40 years due to past emissions.
Temperature, precipitation and solar radiation
We examined daily observed weather data from the Cairngorm Chairlift meteorological station (1980−2019), plus daily data from a 1km resolution gridded data set (1960−2019) to assess evidence of the past trends for maximum and minimum air temperature and precipitation. This analysis was repeated using the UKCP18 climate model data to assess how past rends may align with future projections. We analysed mean maximum and minimum air temperature to assess the change in range and potential consequences of differences in rates of change and levels of variation between them.
We also analysed solar radiation data estimated from satellite data (1994 – 2018)4 to assess potential changes in the micro-climate energy inputs. Solar radiation is an indication of the energy input at the surface level and determines the temperatures in micro-climates. Air temperatures (at
4 See SolarGIS: https://solargis.com/
1.5m above ground, as measured by met stations) are influenced by wider movements of air masses. Increases in ground level receipt of solar radiation can be interpreted as additional drivers of snow melt through direct energy transfer (e.g. through dark surfaces on or near snow).
Future snow modelling
To estimate future snow cover we ran a snow model over all the 5 km grid cells covering the whole National Park (226 cells in total, see Figure 8). The model estimates snow cover based on daily temperature and precipitation. Data from the 3 5km cells covering Aviemore, the Cairngorm Chairlift and Cairn Gorm summit were then extracted. Input future daily weather data (temperature and precipitation) to the model were from the UKCP18 climate projections. When temperature is below a threshold, precipitation accumulates as snow and when temperature rises above the threshold the snow melts. For more information on the model and calibration see Spencer 2016.
Results
Observed Temperature, Precipitation and Solar Radiation trends.
The analysis of temperature and precipitation for the observed and climate model projected future period are shown for January and February in Figures 2 and 3 (see Appendix Figs. 10 – 13 for other winter months). Interpretations can be summarised as:
• For all months there is a clear increasing trend in maximum and minimum temperature observed data between 1960 and 2019 (Figures 2, 3 and 10 – 13).
• There is large inter-annual observed variation.
• Climate model projections indicate a continuation of this warming trend and variation.
• The climate model ensemble mean projected maximum temperature fits well to the extended estimated observed trend line. For minimum temperature the climate model estimates are greater than the extended estimated observed trend line. Previous evaluation of the climate model (Rivington et al 2008) established that the model tends to over-estimate minimum temperature. However, the climate model data used here has been through a bias correction method that aims to correct for such errors. Our interpretation is that minimum temperature may increase by a greater amount than maximum temperature under future climate conditions.
• The number of years in the future when the minimum temperature is above 0°C each month is much greater. December, January and February are key months for snowfall (see Figure 4).
• December: the climate model ensemble mean estimates minimum temperature is likely to be consistently greater than 0°C by mid-2030’s. However, the extended observed trend indicates this may not occur until the mid-2060’s. Maximum temperatures are projected to be consistently greater than the 2°C temperature recognised as the upper limit for snow creation and when snowflakes melt5.
• January: future minimum temperatures are likely to be consistently greater than 0°C by mid- 2030’s. However, the extended observed trend indicates this may not occur until the mid- 2060’s. Maximum temperatures are likely to be consistently greater than 2°C.
• Precipitation has been highly variable between years, with the level of variability estimated to continue in the future. Since 1960 the trends have been: November, December and January have had an increase in mean monthly precipitation (Appendix Figure 15) of approximately 40mm; February and April show no trend; but March and May have a small decrease.
5 Air temperatures at 1.5m above the ground are used, whereas snow creation will likely be at higher altitudes where temperatures are lower. The ground level temperatures indicate probability of snowflakes melting.
Temperature (°C) ‑2 ‑4 12 700 January 10 600 8 6 4 2 500 400 300 200 100 0 Precipitation (mm) 2030 2025 2020 2015 2010 Year 2005 2000 1995 1990 1985 ‑6 ‑8 1960 19655 1980 1975 1970 Observed period Figure 2. January temperature: 1km interpolated gridded observed mean maximum (red line) and minimum (blue line) temperature (°C) where dotted lines are the observed trends extended by 60 years to 2080, and precipitation (blue bars). The future projection period data are: climate model ensemble mean (large solid line) and individual ensemble members (thin dashed line), lowest (dark blue bars) and largest (light blue bars) precipitation estimates for an ensemble member model. Black lines are the observed temperatures measured at the Cairngorm Chairlift meteorological station.
2075 2070 2065 2060 2055 2050 2045 2040 2035 Future projection period
Temperature (°C) 12 700 February 10 600 8 500 6 4 2 0 ‑2 ‑4 ‑6 ‑8 0 400 300 200 100 Precipitation (mm) 2030 2025 2020 2015 2010 Year 2005 2000 1995 1985 1980 1975 1970 Observed period 1990 1965 1960 Figure 3. February temperature: 1km interpolated gridded observed mean maximum (red line) and minimum (blue line) temperature (°C) where dotted lines are the observed trends extended by 60 years to 2080, and precipitation (blue bars). The future projection period data are: climate model ensemble mean (large solid line) and individual ensemble members (thin dashed line), lowest (dark blue bars) and largest (light blue bars) precipitation estimates for an ensemble member model. Black lines are the observed temperatures measured at the Cairngorm Chairlift meteorological station.
2035 Future projection period 2075 2070 2065 2060 2055
Temperature
Key Finding: The mean monthly maximum (Tmax) and minimum (Tmin) temperatures have increased for all winter months since 1982 (Figures 2 – 3, Appendix Figs. 10 – 13, 14). The largest increases have occurred in April. The main snow fall months of January and February have had a relatively small increase in mean Tmax and Tmin.
Mean maximum and minimum temperature for the Cairngorm Chairlift meteorological station have been highly variable between years but there has been an increased trend in all winter month. November and March have seen an increase of Tmin above 0°C since 1982. The trend lines for Tmin have all remained below the 2°C temperature threshold indicator for snow creation, but recently there are more years when in some months the Tmin has been above 2°C.
The mean Tmin in the main snow fall months of January and February has increased by c. 0.5°C. The mean Tmax in April has increased by c. 1.2°C.
Precipiation
Key Finding: There has been an increasing mean monthly precipitation amount trend for November, December and January since 1960, whilst March’s amount has decreased (see Appendix Figure 15).
Precipitation has also been highly variable between years. January’s mean monthly total precipitation has increased by c. 50mm and February’s by c.35mm. This coupled with increases in mean, Tmax and Tmin temperatures indicates greater probability of melting of accumulated snow.
Solar Radiation
Key Finding: There has been an increase in mean monthly solar radiation (MJ m² day¯¹) in February, March and April since 1994 (see Appendix Figure 16). Mean monthly solar radiation for the other winter months has remained constant.
Snow Depth (cm) 0 10 20 30 40 Snow Depth
Snow depth is highly spatially and temporally variable, hence assessing spatial distribution using data from single meteorological station is problematic.
The siting of a met station (e.g. on a ridge or hollow) can have substantial effects on how well data represents the wider area. The Cairngorm Chairlift station (pictured opposite) is located on a raised ridge area hence may under-represent surrounding snow depth. However, trends can be identified from the time-series data enabling relative interpretations to be made about the wider surrounding areas.
When snow depth accumulates is temporally highly variable (Figure 4 and Appendix Fig. 17). For example, in 2006 the entire snow depth was accumulated in March only. For many years most depth accumulation occurs in January and February.
50 60 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 2005 Year Nov Dec Jan Feb Mar Apr May 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Figure 4. Snow depth (cm) per winter by month, Cairngorm Chairlift meteorological station (Note data for 2003 is missing).
There have been many periods of snow depth accumulation followed by melting and subsequent further snow events leading to accumulation again (e.g. winter of 2009-10 had the largest snow depth recordings but these consisted of four distinction time periods, see Appendix Figure 17). Detecting trends within this mix of variation is problematic.
To address this, the average and maximum snow depths per winter provide some evidence of observed trends. Figure 5 indicates a decreasing trend in maximum snow depth (considering substantial yearly variation) of approximately 10 cm since 1983. Average depth has also decreased.
60
50
40 Maximum Snow Depth (cm) 30
20
10
0 10 9 8 7 6 5 4 3 2 1 0 Average Snow Depth (cm) Winter
Maximum snow depth Average depth Linear (Maximum snow depth) Linear (Average depth)
Figure 5. Maximum snow and average snow depth (cm) at Cairngorm Chairlift meteorological station.
Key Finding: There has been a decrease in the observed maximum and average snow depth since the 1983 – 84 winter (Figure 5). Maximum snow depth has declined by c. 10cm and average by c. 3cm.
These decreases may be correlated with the increases in temperature and changes in precipitation amounts, but it should be noted here that snow depth is also a function of wind speed and direction. This study has not considered these factors, but they are likely to be an important influence on snow depth at the location of the Cairngorms Chairlift meteorological station.
Number of days 0 Snow Depth Days Key Finding: There has been an observed decrease in the number of days when snow depth is at specific amounts (2−5, 5 – 10, 10 – 15, 15 – 20 and +20 cm) (Figure 6). The largest decreases have occurred for shallower depths (2−5, 5 – 10 cm) of c. 10 days since 1983. This may indicate situations where snow still falls and accumulates but melts more rapidly. 40 35 30 25 20 15 10 5 1983⁄4 Count 2 – 5cm Count 15 – 20cm Linear (Count 5 – 10cm) Figure 6. Count of days at different snow depth ranges (2−5, 5 – 10, 10 – 15, 15 – 20 and +20 cm) per year, Cairngorm Chairlift. This decrease in the number of days at shallower depths may be a function of reduced number of days when it snowed, but also increased melting due to higher temperatures. Comparing Figures 5 and 6 shows that whilst 1986⁄86 and 1993⁄94 winters had maximum snow depths comparable to 2009⁄20, however the number of days at such depths were less than half. This illustrates how unusual the 2009⁄10 winter was, but this does not greatly affect the decreasing trend. 1985⁄6 1987⁄8 1989⁄90 1991⁄2 1993⁄4 1995⁄6 1997⁄8 Winter Count 5 – 10cm Count +20cm Linear (Count 10 – 15cm) 1999⁄00 2001⁄2 2003⁄4 2005⁄6 2007⁄8 2009⁄10 2011⁄12 Count 10 – 15cm Linear (Count 2 – 5cm) Linear (Count 15 – 20cm) 2013⁄14 2015⁄16 2017⁄18
Mean snow depth (cm) 0 Mean monthly snow depth Key Finding: The mean snow depth per month has decreased in January and February since 1983 (Figure 7). Depth per month has been highly variable but the observed trend has been downwards. Other months have different trends: March has had a slight decrease whilst November has been consistent and December a slight increase. 25 20 15 10 5 1982 1984 Nov Mar Linear (Jan) 1986 1988 1990 1992 1994 Year Dec Jan Apr May Linear (Feb) Linear (Mar) 1996 1998 2000 2002 2005 2007 2009 2011 2013 2015 2017 2019 Feb Linear (Dec) Linear (Apr) Figure 7. Mean monthly snow depth and trends, Cairngorm Chairlift (1982 – 2020). Figure 7 illustrates that in January and February, the main months when snow accumulates, there has been a decrease in the mean snow depth. Comparing with Figure 5, winters in the past have had high maximum depths (e.g. 1986⁄86 and 1993⁄94), but the means are relatively low. This illustrate the complexity of snow accumulation events, e.g. where there may be large snowfalls but the snow duration is shorter. Thus it is important to recognise that snow depth does not necessarily reflect snow density, e.g. accumulations may reduce differently in depth due to temperature and possibly precipitation conditions, giving situations where shallower depths may remain for longer periods of time but as dense snow and ice if the right temperature and precipitation conditions are right (e.g. some light rainfall, thawing and then hard freezing). Conversely, if large accumulations experience warm temperatures and high rainfall they may be completely melted. Wind will also redistribute fallen snow and accumulate it in gullies and hollows.
Elevation (m) 200 400 600 800 1,000 1.200 Mean Elevation (m) 0 to 200 200 to 400 400 to 600 600 to 800 800 to 1,000 Results: Future Projections Snow modelling
Figure 8. Map of Cairngorm National Park and model grid cells. Elevation is shown on a 50 m (left) and 5 km (right) grid. Contains Ordnance Survey data © Crown copyright and database right 2019.⁶
Snow cover data for the cells containing Aviemore (228m, mean grid cell elevation 402m asl), the Cairngorm Chairlift meteorological station (“Base” in Figure 9, 663 m, mean grid cell 485m) and Ptarmigan Restaurant (1097 m, mean grid cell 902m) on Cairngorm Mountain were extracted from the whole National Park area data set. Results are presented in Figure 9.
Key Finding: The future projections indicate a decline in snow cover days per year from the 2030s for Aviemore, the Cairngorm Chairlift meteorological station (Base) and Ptarmigan Restaurant on Cairngorm Mountain. This trend will continue through to the 2080s. There will be large variation between years and there are likely to still be some years comparable with past amounts of snow cover.
⁶ See Footnote 1 for original source.
Snow cover (days per year) Annual daily snow cover from modelled grid cells Cairngorms national park, 1960 — 2080 Aviemore 100 Base Ptarmigan 200 60 75- 40- 50 20- 25 150 100- 50- 0 W 1960 2000 2040 0 2080 1960 MW 2000 2040 2080 1960 2000 2040 2080 Year Figure 9. Annual snow cover (days per year) at three elevation ranges for Aviemore, the Cairngorm Chairlift (Base) and Ptarmigan visitor centre (near Cairngorm Mountain summit). Light grey lines show individual model runs with UKCP18 ensemble members and the heavy black line indicating the smoothed average of these. Note different scales on snow cover axes.
The results show that snow cover may continue to be similar to the past for the next 1 – 2 decades but will decline substantially afterwards. This applies to all elevations, but with larger proportional decreases at higher levels. These declines may be associated with passing a temperature threshold where precipitation no longer falls as snow and any lying snow melts sooner. An overview timeline is approximately: • 2020 – 2030: similar amounts and level of annual variation of snow cover to the past at all elevations. Some years likely to be similar or even possibly greater snow cover than in the past. • 2030 – 2040: declining snow cover but with similar levels