Statistically downscaled projections of snow/rain partitioning for Alaska
Adaptation planning in Alaska, as in other snowy parts of the world, will require snow projections, yet snow is a challenging variable to measure, simulate and downscale. Here we describe the construction and evaluation of 771-m-resolution gridded historical and statistically downscaled projections of snow/rain partitioning for the state of Alaska at decadal temporal resolution. The method developed here uses observational data to describe the relationship between average monthly temperature and the fraction of wet days in that month receiving snow, the snow-day fraction. Regionally and seasonally specific equations were developed to accommodate variability in synoptic scale climatology of rain and snow events. These equations were then applied to gridded decadal temperature data and projections. The gridded products provide a reasonable characterization of snow-day fraction throughout the state. However, there are local deviations from the regional relationships, particularly in the topographically complex areas ringing the Gulf of Alaska and Cook Inlet. When applied to questions about changing precipitation regimes in northern, western and south-eastern Alaska, these data demonstrate the potential for marked changes from snow-dominated to mixed precipitation regimes and also exhibit a wide range of potential future conditions.
McAfee, SA, Walsh, J, and Rupp, TS.. 2014. Statistically downscaled projections of snow/rain partitioning for Alaska. Hydrological Processes. 28(12): 3930-3946. http://onlinelibrary.wiley.com/doi/10.1002/hyp.9934/abstract?deniedAccessCustomisedMessage=&userIsAuthenticated=false. DOI: doi: 10.1002/hyp.9934.