How will snow alter exposure of organisms to cold stress under climate warming?

Aim: The aim was to test the capacity of an ecologically focused microclimate model to capture the interaction between snow cover and soil temperature and use it to assess how historical and future climate warming affects exposure of organisms to stressfully cold conditions in shallow soil. Location: Continental USA. Time period: 1980–2017. Methods: A snow heat budget algorithm was developed and integrated with the general-purpose microclimate model of the NicheMapR package. The gridMET daily historical weather grids were used as environmental forcing. The model was tested against hourly observations of soil temperature and snow cover for 590 Soil Climate Analysis Network/SNOw TELemetry sites across the USA. It was then used to simulate cold stress exposure of ectotherms and hibernation costs for endotherms. Results: The model captured soil and snow observations to within c. 10%–15% of the range (r c. near 0.85). Air temperature exhibited a mean warming rate of 0.37°C per decade across sites, but in snow-affected regions the predicted shallow-soil warming rates were one-third lower. Biophysical analyses predicted an increase in the activity time of ectotherms as a result of contemporary climate change but little change in cold stress at most sites. For endotherms, hibernation costs were predicted to decline in general. However, both reduced and increased cold stress was simulated to occur for ectotherms and endotherms under simulated air temperature warming of ≤3 °C. Main conclusions: Many of the direct biological consequences of climate warming will be mediated through soil temperature. Microclimatic processes linked to snow cover mean that exposure to cold stress for both ectotherms and endotherms might stay constant, decrease or even increase under climate warming, depending on the local circumstances and species-specific biology. The idiosyncratic and season-specific decoupling of above- and below-ground environments under climate warming will have important ecological consequences in snow-affected regions and can now be computed mechanistically