Formulating Relationships for Estimating URBS Model Parameters in Runoff Simulation for Ungauged Catchments: A Case Study in the East-Coast Gulf Basin, Thailand

Abstract

Rainfall-runoff modeling can be directly performed at catchments with streamflow gauging stations to obtain representative model parameters for the area. For ungauged catchments, a traditional approach involves applying model parameters from nearby gauged catchments as proxies. However, this method may result in significant errors if the neighboring catchments exhibit different catchment characteristics. This study aims to develop relationships between specific catchment characteristics and model parameters to estimate the parameters of the URBS rainfall-runoff model for ungauged catchments. The East-Coast Gulf Basin was selected as the study area. Hourly composite radar rainfall data from all stations covering the study area, including the Sattahip, Samut Songkhram, and Suvarnabhumi radar stations were utilized for model calibration and validation at six streamflow gauging stations: Z.11, Z.14, Z.21, Z.28, Z.45 and Z.47A. The calibration and validation results demonstrated that the URBS model effectively simulated hydrographs with go to very good performance, with NSE values ranging from 0.66 to 0.81 and RMSE values ranging from 4.19 to 41.80 m3/s. Additionally, relationships were established between the URBS model parameters and the catchment characteristics, including catchment area, main stream length, the main stream length from the centroid to the outlet and main stream slope. The study found that the correlation coefficients (r) ranged from 0.77 to 0.98, which are considered good to very good for both multiple linear regression and multiple power regression analyses. Therefore, the developed relationships are deemed suitable for application in ungauged catchments.