Estimating snow depletion curves for American River basins using distributed snow modeling

• Published in: Journal of hydrology (Amsterdam) Vol. 334; no. 1; pp. 162 - 173
• Main Authors: Shamir, Eylon, Georgakakos, Konstantine P.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 20.02.2007; Elsevier Science
• Subjects: American River basin; Distributed modeling; Earth sciences; Earth, ocean, space; energy balance; estimation; Exact sciences and technology; Freshwater; Hydrology. Hydrogeology; mathematical models; rivers; Snow cover area; Snow depletion curve; snow depletion curves; Snow water equivalent; Snowmelt; snowpack; Subgrid variability; watersheds; California; American River; Southern Europe; Italy; Marches Italy; United States; USA, California, American R; Distributed modeling; Subgrid variability; Snow depletion curve; Snowmelt; Snow cover area; Snow water equivalent; depletion; models; accumulation; Europe; ablation; surveys; energy balance; melting; North America; drainage basins; correlation; snow; prediction; temperature
• ISSN: 0022-1694; 1879-2707
• DOI: 10.1016/j.jhydrol.2006.10.007

We used a well evaluated temperature-based distributed energy balance model for snow accumulation and ablation with a 1 km × 1 km resolution to derive estimates of snow water equivalent (SWE) and snow cover area (SCA) for the American River basin during the water years 1960–1999. These model prognostic variables were then used to derive and analyze the properties of the snow depletion curves (SDC) of the drainage basins of the three Forks of the American River. Such curves are commonly used in spatially lumped model applications as a priori estimates that account for the areal heterogeneities of the snow pack. Assuming that SDC can be derived for a basin and the annual variability is small, a single SDC offers an adequate description of the snow pack area distribution. Unique model-derived SDC were obtained for each Fork but with large inter-annual variability. The curve variability introduces large uncertainty in the prediction of snowmelt timing and magnitude when spatially lumped models are used. Further analysis of the model derived SDC showed significant correlation between the maximum annual SWE of the basin and the SWE values for a given SCA. In addition, the maximum annual SWE of the basin is also correlated with the SWE from snow courses surveys conducted in the beginning of March. These empirical relationships provide a procedure to estimate the SDC at the beginning of the melting season and, therefore, reduce the uncertainty in snowmelt estimation associated with the use of SDC.