Adapting Historical Climate Network Precipitation and Temperature Data for Water-Balance Modeling

Abstract:

Monthly water-balance modeling is sometimes applied to estimate changes in runoff, snowmelt, and other water-balance variables associated with climate fluctuations and change. Accuracy of the estimates is a function of the ability of the selected water-balance model to approximate the important physical processes in the water balance. Accuracy also logically depends on how closely the input precipitation (P) and temperature (T) time series track natural climate fluctuations. This paper explores the U.S. Historical Climate Network (USHCN) as a source of precipitation (P) and temperature (T) input for water-balance modeling of the Truckee-Carson River Basin (TCRB), which drains eastward from the Sierra Nevada of California and Nevada. USHCN station monthly data are spatially interpolated to water-balance-model grid cells by inverse distance weighting of standardized anomalies, which are then converted to original data units by restoring means and standard deviations adjusted for grid-cell elevation. In this paper, grid-cell P and T over the TCRB from USHCN stations are compared with grid-cell P and T sampled from a widely used publicly available spatially interpolated climate data source (PRISM: Parameter-elevation Relationships on Independent Slopes Model). The goal is to identify key differences in the resultant grid-cell input fields of P and T by the two methods in preparation for water-balance modeling in the context of a tree-ring reconstruction of discharge in the TCRB. Results highlight a trade-off between the temporal stability of the USHCN and the rich data coverage of PRISM. Both networks give broadly similar patterns of basin-wide fluctuations in climate in the 20th and 21st centuries. This work was supported by NSF Award 1903561.