Uncertainty in Temperature and Precipitation Datasets over Terrestrial Regions of the Western Arctic

Abstract

A better understanding of the interannual variability in temperature and precipitation datasets used as forcing fields for hydrologic models will lead to a more complete description of hydrologic model uncertainty, in turn helping scientists study the larger goal of how the Arctic terrestrial system is responding to global change. Accordingly, this paper investigates temporal and spatial variability in monthly mean (1992–2000) temperature and precipitation datasets over the Western Arctic Linkage Experiment (WALE) study region. The six temperature datasets include 1) the fifth-generation Pennsylvania State University–National Center for Atmospheric Research Mesoscale Model (MM5); 2) the 40-yr European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERA-40); 3) the Advanced Polar Pathfinder all-sky temperatures (APP); 4) National Centers for Environmental Prediction– National Center for Atmospheric Research (NCEP–NCAR) reanalyses (NCEP1); 5) the Climatic Research Unit/University of East Anglia CRUTEM2v (CRU); and 6) the Matsuura and Wilmott 0.5° × 0.5° Global Surface Air Temperature and Precipitation (MW). Comparisons of monthly precipitation are examined for MM5, ERA-40, NCEP1, CRU, and MW. Results of the temporal analyses indicate significant differences between at least two datasets (for either temperature or precipitation) in almost every month. The largest number of significant differences for temperature occurs in October, when there are five separate groupings; for precipitation, there are four significantly different groupings from March through June, and again in December. Spatial analyses of June temperatures indicate that the greatest dissimilarity is concentrated in the central portion of the study region, with the NCEP1 and APP datasets showing the greatest differences. In comparison, the spatial analysis of June precipitation datasets suggests that the largest dissimilarity is concentrated in the eastern portion of the study region. These results indicate that the choice of forcing datasets likely will have a significant effect on the output from hydrologic models, and several different datasets should be used for a robust hydrologic assessment.