What drives basin scale spatial variability of snowpack properties in northern Colorado?

摘要

This study uses a combination of field measurements and Natural ResourceConservation Service (NRCS) operational snow data to understand the driversof snow density and snow water equivalent (SWE) variability at the basinscale (100s to 1000s km2). Historic snow course snowpack densityobservations were analyzed within a multiple linear regression snow densitymodel to estimate SWE directly from snow depth measurements. Snow surveyswere completed on or about 1 April 2011 and 2012 and combined with NRCSoperational measurements to investigate the spatial variability of SWE nearpeak snow accumulation. Bivariate relations and multiple linear regressionmodels were developed to understand the relation of snow density and SWEwith terrain variables (derived using a geographic information system(GIS)). Snow density variability was best explained by day of year, snowdepth, UTM Easting, and elevation. Calculation of SWE directly from snowdepth measurement using the snow density model has strong statisticalperformance, and model validation suggests the model is transferable toindependent data within the bounds of the original data set. This pathway ofestimating SWE directly from snow depth measurement is useful whenevaluating snowpack properties at the basin scale, where many time-consumingmeasurements of SWE are often not feasible. A comparison with a previouslydeveloped snow density model shows that calibrating a snow density model toa specific basin can provide improvement of SWE estimation at this scale, andshould be considered for future basin scale analyses. During both water year(WY) 2011 and 2012, elevation and location (UTM Easting and/or UTM Northing)were the most important SWE model variables, suggesting that orographicprecipitation and storm track patterns are likely driving basin scale SWEvariability. Terrain curvature was also shown to be an important variable,but to a lesser extent at the scale of interest.