Improving snow water equivalent simulations in an alpine basin by blending precipitation gauge and snow pillow measurements. Issue 1 (28th January 2023)

• Record Type: Journal Article
• Authors: Pelak, Norman; Sohrabi, Mohammad; Safeeq, Mohammad; Conklin, Martha
• Is Part Of: Hydrological processes. Volume 37:Issue 1(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2023-01-28
• Subjects: high‐elevation precipitation -- precipitation gauges -- snow modelling -- snow pillows -- snow water equivalent -- temperature sensitivity; Hydrology -- Periodicals; Hydrology -- Research -- Periodicals; Hydrologic models -- Periodicals; Hydrological forecasting -- Periodicals; 631.432
• DOI: 10.1002/hyp.14796
• Languages: English
• ISSNs: 0885-6087

Abstract

Abstract: Mountain snowpack is a critical freshwater reservoir that sustains ecosystems and provides water for hydropower, agriculture, and urban uses. Accurate basin‐scale snowpack estimates are essential for ecosystem and water resources management. Our ability to predict snowpack evolution is hampered by the low density of precipitation gauges in high mountain regions, as most precipitation gauges are installed in easily accessible areas, which are often below the snowline. These measurements are typically spatially interpolated to fill data gaps in high mountain areas, resulting in a range of gridded products with considerable differences. On the other hand, natural resources agencies rely on a few strategically placed snow pillows and snow courses to infer basin‐scale snow water equivalent (SWE). In the Kings River Basin, California, there is an opportunity to combine precipitation and snow pillow measurements to improve basin‐scale snowpack estimates. In this study, we blend precipitation gauge observations with snowpack measurements to force a spatially distributed, process‐based snowpack evolution model in order to improve basin‐scale precipitation and snowpack predictions. We test the blended precipitation (Blended scenario) against other gridded precipitation products, which are derived from precipitation gauges only (Gauge scenario) and the PRISM precipitation product (PRISM scenario). Our results show that the Blended scenario had improved snowpack predictions (NSE = 0.85) over the Gauge and PRISM scenarios (NSE = 0.37 and 0.81) when compared to observed SWE, which can be attributed to better representation of precipitation at high elevations. The Blended scenario was also the only scenario to produce sufficient surface water input (SWI) to account for the combined estimates of full natural flow and evapotranspiration in the basin. We also examine the model sensitivity to warming and discuss the impact of warming on SWE and SWI and implications for water management. The results underscore the need for improved snow observation networks at high elevations that can be used to improve the representation of precipitation patterns, which is critical for the hydrologic modelling of mountainous basins. Abstract : Blending inferred precipitation from snow pillows with gauge measurements in an alpine basin leads to larger depths of modelled SWE at high elevations, which are in the upper right section of (a), than using gauges alone. The additional SWE results in more realistic basin‐wide water balance estimates for the Blended compared to the Gauge or PRISM precipitation scenarios (b).