APPLICATION OF WEB-BASED L-THIA 2012 DIRECT RUNOFF AND POLLUTANT AUTO-CALIBRATION MODULE USING GENETIC ALGORITHM

摘要

Urbanization in a watershed leads to increased runoff and pollutant loads due to increase inthe impervious area. In this study, our approach was to calibrated surface runoff and nonpointsource (NPS) pollution loads by minimizing the differences between the observed and modelsimulated data. The model “Long-Term Hydrology Impact Assessment (L-THIA)” was adoptedfor the study. Here, we used default Event Mean Concentrations (EMCs) of BOD, total nitrogen-TN, and total phosphorus-TP (as the default values to L-THIA) collected at various regions inSouth Korea corresponding to the classifi cations of different rainfall intensities and land usefor improving predicted NPS pollutions. The study region was the Yeoju-Gun and Icheon-Sisites, located within the Paldang watershed (South-Korea) that suffered by NPS pollutantsgenerated by frequent overfl ows at outfalls of storm sewers within urbanized areas during thesummer monsoon periods. We calibrated the GA-based L-THIA model using the measureddirect runoff and pollutant loads. Our approach was tested only during the simulation period in2011, because of the limited EMCs. The EMC values, as the default values used for L-THIA,have the different ranges of less than 10, 10-30, 30-50, and more than 50 mm, respectively.Our model outputs using the sub-daily (ten minute intervals) rainfall and classifi ed EMCscorresponding to the various rainfall intensities and landuse matched reasonably well with theobservations, although the results at the Icheon-Si site have more uncertainties (especially forTP with the negativeNSE value). the Web-based L-THIA model could provide easy accesstousers across the world. In our future works, we plan to develop an improved L-THIA modelthat can estimate runoff and NPS pollutions using open source GIS programs.