Estimating the Spatial and Temporal Distribution of Snow Water Equivalent Within a Watershed

Abstract

The goal of this research was to develop a spatial-temporal model to forecast the snow water equivalent (SWE) within a watershed. This model defined the relationship between the physical parameters of a watershed and the spatial distribution of SWE within that watershed. Many models of snowmelt runoff rely on snow depletion curves, which describe the seasonal decline of snow covered fraction and SWE, assuming some interannual uniformity of basin response. The null hypothesis associated with this research was that the spatial and temporal variability of SWE within a watershed is a random process that is independent of the physical characteristics of the watershed.

A conceptual model of spatially distributed SWE accumulation and depletion that can be calibrated and validated with spatially distributed observations was created. The effects of the physical variables and parameters on the SWE distribution within a watershed were demonstrated, and both significant and insignificant physical variables and parameters were identified. How data were used in the calibration/validation of the model was demonstrated, including showing the benefit of additional data on model accuracy. Finally, the proper calibration and validation of the model using an actual watershed was demonstrated on three different watersheds.

Results of the research were mixed in terms of accepting or rejecting the null hypothesis. Created SWE maps and satellite images of the Upper Rio Grande Watershed visually suggested that similar SWE patterns exist for this watershed. An analysis of the data from the SNOTEL sites within the Upper Rio Grande Watershed also suggested the existence of similar interannual SWE patterns within the watershed. This analysis supports the Depletion Curve Theory.

However, an analysis of the SWE distributions for the three watersheds, performed utilizing the Kolmogorov-Smirnov Two-Sample Nonparametric Test, suggested that consistent interannual SWE patterns do not exist for the watersheds studied. This analysis contradicts the Depletion Curve Theory.