Publisher's Synopsis
In this comprehensive study, the impact of climate change on hydro-meteorological parameters in the Bagmati River Basin is rigorously examined. Climate change poses significant threats, altering the frequency of extreme hydrological events with profound societal and environmental implications. These impacts encompass human health, safety, water resources management, sediment transport dynamics, and infrastructure resilience. The research employs statistical downscaling techniques using atmospheric predictors such as temperature, humidity, and geopotential height to refine precipitation data. Monthly rainfall data from four rain gauge stations and reanalysis data spanning 33 years are utilized, alongside discharge data from a key gauging site on the Bagmati River. Statistical models, including Feedforward Neural Networks (FFNN) and Wavelet Neural Networks (WNN), are developed and validated to predict precipitation patterns. WNN emerges as particularly effective for precipitation prediction, informing projections up to 2036 using General Circulation Model (GCM) outputs. Furthermore, the study advances rainfall-runoff modeling using FFNN and WNN, enhanced through Particle Swarm Optimization (PSO) for optimal parameterization. This hybrid approach improves the robustness of runoff estimation critical for water resource management. Trend analysis employing Mann-Kendall tests and Sen's Slope estimators reveals significant climate-induced variations in rainfall and runoff patterns from 1981 to 2013, crucial for understanding basin hydrology amidst changing climatic conditions. Addressing drought, the study employs the Standardized Precipitation Index (SPI) to forecast future drought occurrences during 2014-2036, highlighting drought years in 2023, 2027, 2031, and 2033. This underscores the study's contribution to drought monitoring and mitigation strategies in the region. Overall, this research underscores the urgent need for systematic approaches to mitigate the impacts of climate change on hydrological systems.
