资源环境科技发展态势分析平台

Characterizing hydrological processes on large scales is challenging due to limitations of observational networks, remoting sensing platforms, and modeling techniques. Water balances have larger uncertainties in mountain regions, where orographic processes produce high spatial variability in precipitation patterns and snow accumulation. Recent work suggests current water budgets underestimate mountain snow water storage, perhaps indicating biases in modeled precipitation. We assess whether global hydroclimate data sets underestimate precipitation for six North American watersheds, ranging from 3-70% mountainous. By selecting a single representative year for each watershed, we compare relatively high-resolution precipitation estimates from the Weather Research and Forecasting (WRF) regional climate model with four global products: Modern-Era Retrospective Analysis for Research and Applications, version 2, the Global Land Data Assimilation System, the Global Precipitation Climatology Project, and the Climate Research Unit's climate data set. Comparisons to WRF precipitation suggest that observation-based gridded data products do not produce reasonable estimates of watershed-scale cool-season precipitation, underestimating by 1-69%. The Global Precipitation Climatology Project and the Climate Research Unit data set have average biases of -26% and -38%, respectively. The Modern-Era Retrospective Analysis for Research and Applications version 2 and the Global Land Data Assimilation System show smaller underestimates relative to WRF (-17% and -21%, respectively), with nearly all mean bias from the mountains (underestimated by 27% and 39%) rather than the topographically simpler lowlands (underestimated by 5% and 2%). We suggest global products fail to capture orographic enhancement of precipitation, resulting in large underestimates of precipitation, snowfall, and snow water storage in mountains of selected North American watersheds, which highlights the need for more accurate precipitation estimates to accurately assess spatiotemporal variations in the water cycle.