| Description |
A 26-year high-resolution dynamical downscaling over the Wasatch Mountains of Utah, USA was performed using the Weather Research and Forecasting (WRF) model with initial and boundary conditions derived from Climate Forecast System Reanalysis (CFSR). Precipitation validation was conducted on the inner (4-km resolution) domain with Snowpack Telemetry (SNOTEL) and Parameter-elevation Regressions on Independent Slopes Model (PRISM) data sets. Analysis of seasonal performance reveals the model's overall good skill at reproducing the spatial distribution of precipitation. Annual precipitation validates within ∼20% of SNOTEL. The largest monthly biases occurred in December-January (∼+30%). Composite analysis of cold season days with large positive or negative precipitation biases reveals two distinct synoptic regimes with significantly different moisture, temperature, and circulation patterns that respectively enhanced geopotential height and moisture biases consistent with the sign of their mean precipitation biases. The number of cold season days with large (>5 mm) positive precipitation bias was negatively correlated with El Ni˜no (r = −0.55), indicating storm track-related effects on the sign of the bias consistent with the distinct synoptic regimes revealed by the above-noted composite analyses. This historical simulation was compared with a pseudo-global warming simulation of climate change to evaluate the roles of temperature and precipitation in spring snowpack (S) variability across the western United States. In both historical and future climate, the negative correlation between S and temperature weakens linearly with elevation whereas the correlation between S and precipitation increases logarithmically with elevation. The curvilinear relationship in the latter case was not visible in prior studies because of the observation networks' limited range. In the historical simulation, there is a range of threshold elevations (1574-2119 m) above which precipitation is the main driver of snowpack variability and below which temperature is the main driver. Under a moderate end-of-century climate change scenario, these thresholds increase by 239-447 m across six mountainous regions (317 m on average). These rising thresholds indicate increasing spatial and elevational vulnerability of western U.S. spring snowpack along with associated impacts to hydrologic and ecologic systems. |
