Climate prediction downscaling of temperature and precipitation in the Great Basin region

The primary objective of this study is to investigate the model biases in three global climate model (GCM; NCAR-CCSM3, ECHAM5, and CSIRO-Mk3.5) outputs and their ensembles under designated Intergovernmental Panel on Climate Change (IPCC) climate change scenarios to assess future hydrological resources and their variability, uncertainties, and socio-economic impact in the semi-arid and mountainous terrain of Nevada as well as the entire southwest U.S. region. The study addresses downscaling methodology for the surface variables (2-m air temperature and precipitation) from GCM horizontal grid resolutions (100 km or more) to regional scales (10 km or less) appropriate for hydrologic impact studies.

Preliminary hindcast analysis for a 50-year period (1950-2000) indicated that the precipitation rates extracted from the GCMs at 46 individual stations in Nevada show correct seasonal trends, but the monthly mean precipitation rates are significantly overestimated, especially in the Humboldt River watershed region (an area of 44030 km2) of Nevada. The areal mean precipitation rate is considerably biased by about 5 mm day-1 as compared to observations (Western Regional Climate Center, DRI; WRCC), and National Climate Data Center (NCDC) and Parameter-elevation Regressions on Independent Slopes Model (PRISM) climate data sets. The daily mean surface air temperature from the GCMs and a regional climate model (RCM) using Weather Research and Forecasting (WRF) forced by the CCSM3 outputs is generally under-predicted, with root-mean-square errors as large as 6o K on an annual scale.

The present study employs bias correction and spatial disaggregation (BCSD) models to improve representation of synoptic-scale seasonal and extreme events at local and regional scales. Recognizing the non-stationarity in the climatic and hydrologic processes, an ensemble approach is used to better represent the range of possible outcomes under different IPCC greenhouse gas emissions scenarios. The study also will contribute to further improvements of the convective and microphysical parameterizations in the Weather Research Forecasting regional climate model (WRF-RCM).