Impact of precipitation changes on runoff and soil erosion in Korea using CLIGEN and WEPP

Abstract

The quality of spatially and temporally distributed weather information is critical in soil erosion model results because of the primary influence of rainfall on runoff and soil movement. Detailed climate data for the Water Erosion Prediction Project (WEPP) model can be generated by a climate generator (CLIGEN) based on long-term statistical parameters for more than 4,000 locations in the United States. The objectives of this study were to apply CLIGEN and WEPP and examine the effects of changing storm frequency, storm depth, or a combination of the two on predicted rainfall, runoff, and soil loss. Two different sites, Chun-Cheon and Jeon-Ju, were studied and compared for the period 1966 to 2005. Chun-Cheon is located at a higher altitude and is surrounded with forest, while Jeon-Ju is located in the plains. CLIGEN was used to generate 100-year climate sequences with daily climate data e.g., temperature, precipitation, wind, and solar radiation for a representative climate station in the study sites to predict runoff and soil loss with WEPP. Three precipitation change scenarios were examined in this study: (1) adjusting the number of days with rainfall, (2) adjusting the mean amount of rainfall on a wet day, and (3) a combination of 1 and 2. Observed mean annual precipitation at Chun-Cheon (1,305 mm [50.9 in]) was similar to Jeon-Ju (1,291 mm [50.4 in]). CLIGEN simulated mean annual precipitation depths in Chun-Cheon and Jeon-Ju were very close to the observed data. The WEPP model predicted runoff in Jeon-Ju was 48.8% higher than that in Chun-Cheon and estimated soil loss in Chun-Cheon was 55.6% higher than that in Jeon-Ju. Precipitation change scenario 3 that combined changes in precipitation occurrence with changes in rainfall storm depths showed the largest impacts on predicted runoff and soil loss. A combined 20% increase in these precipitation parameters resulted in increases of 44%, 54%, and 52% in generated average annual precipitation, predicted runoff and predicted soil loss, respectively, at Chun-Cheon, while increases at Jeon-Ju were 44%, 60%, and 27%. Increases in rainfall due to future climate change may thus potentially result in substantial and nonlinear increases in runoff and soil loss in Korea.