Creating a Rainfall-Runoff Model for Measuring Streamflow from First-Order and Ungauged Headwaters of the Tropically Medium-Sized Watersheds in Northern Thailand.

This study investigated the suitable Initial abstraction ratio (λ) for estimating the depth of runoff in northern Thailand. The Initial abstraction ratio (λ) and curve number (CN) are two catchment parameters in the Natural Resources Conservation Service-Curve Number method (NRCS-CN). In addition, the λ is considered a regional parameter, therefore, a rainfall-runoff model for estimating streamflow in first-order and ungauged headwater of the tropically medium-sized watersheds in northern Thailand should be created to gain the more accurate runoff for this particular region. NRCS-CN was employed to determine the λ in five watersheds, evaluating the performance of the modified λ values and determining the correlation for potential maximum retention (S0.2) from the original Curve Number (CN) equation with the revised Sλ value. A set of instruments comprised a rain gauge and a water stage recorder installed at each watershed. The initial abstraction (Ia), λ and CN of 80 storm events in 2018 and the Sλ-S_0.20 correlation equation were calculated. The runoff model assessment from 72 selected events in 2019 was validated. The results revealed a median λ of 0.045. The runoff model assessment used the data from the measuring station compared to those calculated from the created rainfall-runoff model provided more accurate runoff and better responded to Nash-Sutcliff coefficient (NSE), coefficient of determination (R²) and percent of bias (PBIAS) (0.92, 0.92 and 1.91 respectively). Regarding the Watershed-Specific S correlation equation based on the S correlation equation between S0.045 and S0.20 values, R² was 0.97. It is recommended that the created model with the λ of 0.045 should be used for future simulations in the 100 to 250 km2 tropically medium-sized watershed. [ABSTRACT FROM AUTHOR]