Snow cover duration and extent for Great Britain in a changing climate: Altitudinal variations and synoptic‐scale influences

Abstract

Snow cover is an important indicator of climate change but constraints on observational data quality can limit interpretation of spatial and temporal variability, especially in mountain areas. This issue was addressed using archived data from the Snow Survey of Great Britain to infer key climate relationships which were then used to reference larger‐scale patterns of change. Data analysis using nonlinear (logistic) regression showed average changes in yearly snow cover were strongly related to mean temperature rather than precipitation values. Inferred change shows long‐term decline in average yearly snow cover with greatest declines in some mountain areas, notably in northern England, that can be related to their position on the most temperature‐sensitive segment of the logistic curve. Further declines in snow cover were projected in the future: a central ensemble projection from HadRM3 climate model showed average yearly snow cover predominantly confined to Great Britain mountain areas by the 2050s. However, inter‐annual variability means some years can deviate significantly from average snow cover patterns. Site‐based analysis showed this variability has distinctive geographical variations and different influences for mountains compared to adjacent valleys. Comparison of inter‐annual variability with Lamb weather‐type frequency and North Atlantic Oscillation index shows the influence of large‐scale airflow patterns on snow cover duration. Most notable is the role of northwesterly and northerly flows in explaining snowy years on mountains exposed to that direction, compared to influence of easterly flows at lower levels. Future changes will therefore depend on dominant annual/decadal circulation patterns in addition to long‐term declines from climate warming.

Snow cover variability in Great Britain during a changing climate

“…The same analysis seemed to find no significant relationship for average annual snow cover based on precipitation, either expressed as total precipitation or inferred snowfall precipitation (Brown, ). However, this may also be related to the difficulties incurred in accurately measuring and interpolating precipitation for the uplands, especially when estimating snowfall amounts.…”

“…Archived SSGB data extending from the Brecon Beacons to the Scottish Highlands has recently been used to further investigate altitudinal and geographic variations in average annual snow cover through a general relationship with mean temperature (Brown, ). This work used a standard lapse rate to interpolate mean temperature at the same location and elevation as SSGB snow cover data (base station, 750m and summit elevation).…”

“…Non‐linear relationship between snow cover duration and mean temperature (Oct–May), defined by best‐fit logistic regression based on SSGB data (after Brown, ).…”

“…Similarly, Brown () used the SSGB to investigate inter‐annual variability for both upland and lowland locations compared to synoptic‐level atmospheric patterns represented by the NAO index and frequency of Lamb Weather Types. This study found that, although longer snow cover duration in the lowlands was strongly associated with easterly type airflows, indicating prolonged incursion of polar continental air, this pattern was less evident in mountain locations with the exception of those particularly exposed to winds from that easterly direction.…”

“…As a consequence, it appears that some mountain ranges may have different synoptic relationships when compared to the adjacent lowlands regarding the influence of dominant seasonal circulation patterns, including against aggregate measures such as the NAO index. It has therefore been suggested, based on archived SSGB data, that a positive NAO index may have been associated with increased snow duration on some UK mountains (Brown, ), although whether this relationship still persists is unknown. Nevertheless, these topographic and synoptic‐scale interactions, often occurring on a multiyear and interannual timescale, may explain apparently anomalous trends found in shorter time series data, as with the recent observations of increasing snow cover in the western Cairngorms (Andrews et al ., ).…”