Optimasi Model Natural Rural Electrical Cooperation Agency Untuk Memprediksi Debit Aliran Bulanan di Sub DAS Lesti

Climate change and land use change in the Lesti Sub-DAS increase the risk of flooding and land degradation, requiring reliable flow predictions to support water resource management. However, the performance of the NRECA model in predicting monthly flows in this region is still not optimal because calibration-validation strategies and the use of environmental parameters have not been systematically studied. This study optimizes the NRECA model to predict monthly discharge for the period 2011-2020 in the Lesti sub-watershed by calibrating the GWF and PSUB parameters based on rainfall, evapotranspiration, and watershed morphometry data in three data division scenarios (70:30, 80:20, and 90:10 for calibration:validation). The results show that all scenarios produce excellent performance with calibration Nash-Sutcliffe Efficiency (NSE) values between 0.99491-0.99561 and correlation coefficients (R) between 0.99746-0.99785, while validation yielded NSE values between 0.89112-0.97227 and R between 0.49959-0.81520. The best scenario was obtained with a combination of 8 years of calibration and 2 years of validation, with NSE = 0.99561 and R = 0.99785 at the calibration stage, and NSE = 0.97227 and R = 0.81512 at the validation stage, indicating the model's ability to consistently represent monthly discharge variations. The similarity between the model discharge pattern and observations during the base and peak flow periods indicates that GWF optimization specifically improves the representation of base flow response. This study contributes by presenting an optimization-based calibration-validation scheme for the NRECA model, which can be used as a reference in conservation planning and reservoir operation management in watersheds with limited data.