Search for a command to run...
The existence of fine-grain climate heterogeneity has prompted suggestions that species may be able to survive future climate change in pockets of suitable microclimate, termed 'microrefugia'. However, evidence for microrefugia is hindered by lack of understanding of how rates of warming vary across a landscape. Here, we present a model that is applied to provide fine-grained, multidecadal estimates of temperature change based on the underlying physical processes that influence microclimate. Weather station and remotely derived environmental data were used to construct physical variables that capture the effects of terrain, sea surface temperatures, altitude and surface albedo on local temperatures, which were then calibrated statistically to derive gridded estimates of temperature. We apply the model to the Lizard Peninsula, United Kingdom, to provide accurate (mean error = 1.21 °C; RMS error = 1.63 °C) hourly estimates of temperature at a resolution of 100 m for the period 1977-2014. We show that rates of warming vary across a landscape primarily due to long-term trends in weather conditions. Total warming varied from 0.87 to 1.16 °C, with the slowest rates of warming evident on north-east-facing slopes. This variation contributed to substantial spatial heterogeneity in trends in bioclimatic variables: for example, the change in the length of the frost-free season varied from +11 to -54 days and the increase in annual growing degree-days from 51 to 267 °C days. Spatial variation in warming was caused primarily by a decrease in daytime cloud cover with a resulting increase in received solar radiation, and secondarily by a decrease in the strength of westerly winds, which has amplified the effects on temperature of solar radiation on west-facing slopes. We emphasize the importance of multidecadal trends in weather conditions in determining spatial variation in rates of warming, suggesting that locations experiencing least warming may not remain consistent under future climate change.