Evaluating the spatial variability of snowpack properties across a northern Colorado basin.

摘要

Knowledge of seasonal mountain snowpack distribution and estimates of its snow water equivalent (SWE) can provide insight for water resources forecasting and earth system process understanding, thus, it is important to improve our ability to describe the spatial variability of SWE at the basin scale. The objectives of this thesis are to: (1) develop a reliable method of estimating SWE from snow depth for the Cache la Poudre basin, and (2) characterize the spatial variability of SWE at the basin scale within the Cache la Poudre basin. A combination of field and Natural Resource Conservation Service (NRCS) operational-based snow measurements were used in this study. Historic (1936–2010) snow course data were obtained for the study area to evaluate snow density. A multiple linear regression model (based on the historical snow course data) for estimating snow density across the study area was developed to estimate SWE directly from snow depth measurements. To investigate the spatial variability and observable patterns of SWE at the basin scale, snow surveys were completed on or about April 1, 2011 and 2012 and combined with NRCS operational measurements. Bivariate relations and multiple linear regression models were developed to understand the relation of SWE with physiographic variables derived using a geographic information system (GIS). SWE was interpolated across the Cache la Poudre basin on a pixel by pixel basis using the model equations and masked to observed SCA (from an 8-day MODIS product).;The independent variables of snow depth, day of year, elevation, and UTM Easting were used in the model to estimate snow density. Calculation of SWE directly from snow depth measurement using the snow density model has strong statistical performance and model verification suggests the model is transferable to independent data within the bounds of the original dataset. This pathway of estimating SWE directly from snow depth measurement is useful when evaluating snowpack properties at the basin scale, where many time consuming measurements of SWE are often not feasible. Bivariate relations of SWE and snow depth measurements (from WY 2011 and WY 2012) with physiographic variables show that elevation and location (UTM Easting and UTM Northing) are most strongly correlated with SWE and snow depth. Multiple linear regression models developed for WY 2011 and WY 2012 include elevation and location as independent variables and also include others (e.g., eastness, slope, solar radiation, curvature, canopy density) depending on the model dataset. The final interpolated SWE surfaces, masked to observed SCA, generally show similar patterns across space despite differences in the 2011 and 2012 snow years and differing estimation of SWE magnitude between the combined dataset of field-based and operational-based measurements (modelO+F) and the dataset of operational-based measurements only (modelO). Within each of the model surfaces, interpolated volume of SWE was greatest within Elevation Zone 5 (3,043–3,405 m). The percentage of the total interpolated SWE volume for each model was distributed similarly among elevation zones.