Evaluation of Derived Monthly Statistical Parameter Values of CLIGEN for Grid-based Climate Database

Abstract

CLIGEN, a stochastic weather generator, is often used in conjunction with the Water Erosion Prediction Project (WEPP) model for runoff and soil erosion predictions. CLIGEN requires a parameter file as input that contains monthly statistical parameter values of weather variables derived from the point-scale US Historical Climatology Network (USHCN) database. To apply WEPP in many locations across the U.S., grid-based products are generally used for climate interpolations. This study derived the monthly statistical parameter values for CLIGEN parameter files from Daymet, PRISM, MACA, and MarkSim gridded products using the daily precipitation and temperature values. CLIGEN was used to generate a stochastic climate for the comparison of monthly values of precipitation and temperature. Monthly statistical parameter values of point-scale CLIGEN parameter files were compared to those of gridded climate data products. Baseline climate scenario comparisons were made for Daymet, PRISM, MACA, and MarkSim for 20 selected locations in the contiguous US (Figure 1). Future climate scenario comparisons were also made for similar locations between MACA and MarkSim separately for early-, mid-, and late-21^st century climate scenarios. <fig><graphic xlink:href=23053_files/23053-03.jpg id=ID_1adf644f-f558-40c7-b72f-b0029158fdbb><alt-text>DiagramDescription automatically generated</alt-text></graphic></fig> <b>Figure 1. </b>Twenty selected locations in the contiguous US used in this study. Results of this study show that for the baseline scenario, Daymet and PRISM closely matched the monthly statistical parameter values of CLIGEN while MACA and MarkSim were statistically different. Monthly precipitation and temperature values derived from stochastically-generated climate from respective parameter files were similar among all the climate products. For future climate scenarios, MACA and MarkSim values were similar.