Evaluating the Green-Ampt and SCS-CN Loss Methods based on Extreme Inflow Events of the Lam Pao Reservoir

Abstract

This study aimed to evaluate two loss approaches of the HEC-HMS model, which were the Green-Ampt and SCS-CN, for simulating inflow to the Lam Pao reservoir. For direct runoff, baseflow, and routing simulations in the model, Snyder unit hydrograph, linear reservoir, and Muskingum methods, respectively, were used with the two approaches. The catchment of the reservoir was delineated in the HEC-HMS software and its characteristics were extracted from a Digital Elevation Model of 30 m x 30m.  Daily runoff data observed during the five flood events (in 2010, 2011, 2017, 2018, and 2019) at the three monitoring stations (E.65, E.76A and Lam Pao Dam) was used for calibration and validation of the model. To evaluate the model, the coefficients of determination, Nash-Sutcliffe efficiency, percentage errors in peak, percentage errors in volume were used. The results revealed that runoff simulated by the model using the SCS-CN approach was more reliable than that simulated by the model using the Green-Ampt approach. This study can be a good example for future hydrological modeling of flood forecasting in Thailand.