Water resource management and flood mitigation: hybrid decomposition EMD-ANN model study under climate change

• DOI: 10.1007/s40899-024-01048-9
• Publication Date: 2024-03-05T19:02:27Z
• Authors: Nisar Ahmad; Xu Yi; Muhammad Tayyab; Muhammad Hamza Zafar; Naureen Akhtar
• Keywords: 0208 environmental biotechnology; 02 engineering and technology

Abstract The growing population and the rise in urbanization have made managing water a critical concern around world recent years. Globally, flooding is one of most devastating natural disasters. Flood risk mitigation relies heavily on accurate consistent streamflow forecasts. Pakistan Upper Indus Basin (UIB) vulnerable to flooding. Floods become more frequent decades. UIB can be divided into sub-regions due its landscape variability, collective impact prominent Massam region. hydrological meteorological station observations been used study seasonal hydro-meteorological variations. To predict flooding, this proposes hybrid model combining artificial neural networks as multi-layer perceptron (MLPs) feed-forward mode, along with empirical mode decomposition (EMD). Data collected by surface-water hydrology project Meteorological Department from 1960 2012, 1969 1972 2012 utilized 17 locations. Statistical parameters Nash–Sutcliffe Efficiency were measured analyze model’s prowess. As result, decomposition-based models perform better than AI-based when it comes prediction accuracy. MLPQTP-EMD performed exceptionally competing AI models. results are further validated performing peak value analysis during season (June–September) achieving remarkable 91.3% score adding 5.6% increase EMD for input data 39.3–32.3% statistical indices scores.