Automatic calibration tool for snowmelt runoff simulation using shuffled complex evolution algorithm

In recent years, hydrological models have gained significant popularity in water resource planning and management activities. In 2019, a spatially distributed snow melt runoff model (SDSRM) was developed (Rajkumari et al.) and applied successfully for various data-scarce Himalayan River basins (Kiba et al.). However, the model calibration process is currently done manually, which can be time-consuming and prone to errors. To address this issue, the study focuses on the development and application of an automated tool for SDSRM calibration. The tool was developed in the VB.NET programming language. In this tool, SDSRM was integrated with the Shuffled Complex Evolution Algorithm (SCE-UA) to produce a pool of eligible parameter sets. The developed tool was tested for the Mago river basin of Tawang district in Arunachal Pradesh, India. These parameter sets were then analyzed based on the performance metrics such as NSE, R-square, PBIAS, and RSR for each method. The NSE value ranged from 0.755 to 0.79 and the maximum NSE was obtained for the Energy Balance method. The optimum parameter set also yielded peak runoff rate and runoff volume for each method. The Radiation Temperature Index method yielded the closest values for simulated runoff rate and runoff volume as 106.5 m³s⁻¹ and 10293.02 m³ to the observed values as 136.64 m³s⁻¹ and 10293.08 m³. The showcased applications show that SDSRM-SCE holds the potential to serve as a valuable tool for the calibration of SDSRM.