Elevation-dependent sensitivity of the snow accumulation period to climate warming in mountain basins

The seasonal snow accumulation period – approximated as the number of days between the autumn onset and spring end of sub-freezing temperatures – varies across elevation in mountain basins, with colder, higher elevation areas accumulating snow for longer relative to warmer, lower-elevation areas. Climate warming will reduce the sub-freezing duration throughout a year and alter the timing and magnitude of snow accumulation and melt, with profound implications for downstream ecosystems, infrastructure, and societies; however, it is not well known how such changes will vary across elevation gradients in mountain basins.

Here we present a novel idealized conceptual model to analytically explain how and why the annual sub-freezing period responds differently to climate warming across elevation gradients in mountain basins. We use this model to demonstrate theoretically under what climactic conditions the sub-freezing duration is more sensitive to warming at low elevation areas relative to high elevations, and vice-versa. Both the strength (i.e. the magnitude of change of sub-freezing duration) and the shape (i.e. whether high- or low-elevation regions are more sensitive to warming) of this sensitivity vary non-linearly as a function of mean temperature, temperature seasonality, temperature lapse rate, and the basin elevation range.

We then use ERA5-Land climate reanalysis data to apply this novel framework to mountain basins across North America. We find that in basins with warmer climates (i.e. those with mean annual temperatures greater than freezing), the annual number of days below freezing is more sensitive to warming at low elevations. In contrast, in basins with colder climates (i.e. those with mean annual temperatures less than freezing), the annual number of days below freezing are more sensitive to warming at high elevations. We evaluate both the present sensitivity of the sub-freezing duration, as well as observed changes since 1950.

We detail two case studies in which we apply our methodology. First, we present how the snow accumulation period may decrease by substantially more in some glacierized areas than others (i.e. Canadian Rocky Mountains vs Coast Mountains). We find that in many glacierized basins, the snow accumulation period is more sensitive to warming in the high-elevation glacierized areas relative to lower-elevation non-glacierized areas. Second, we adapt our methodology to describe the period of the year when heatwaves, defined as persistent periods hotter than the seasonally-varying 90^th percentile of temperature, may be warmer than freezing and thus potentially able to modify basin hydrology. We show that the potential for heatwave-driven ablation in mountain basins changes non-linearly across elevation with warming, and that heatwaves may rapidly emerge as a more prominent driver of high-elevation melt. Overall, our study presents a novel modelling framework to assess and project changes to melt-driven hydrological dynamics in mountain basins.