A Water Consumption Forecasting Model by Using a Nonlinear Autoregressive Network with Exogenous Inputs Based on Rough Attributes

A hybrid SWAT-ANN model approach for analysis of climate change impacts on sediment yield in an Eastern Himalayan sub-watershed of Brahmaputra

The current study focuses on analyzing the impacts of climate change and land use/land cover (LULC) changes on sediment yield in the Puthimari basin, an Eastern Himalayan sub-watershed of the Brahmaputra, using a hybrid SWAT-ANN model approach. The analysis was meticulously segmented into three distinct time spans: 2025-2049, 2050-2074, and 2075-2099. This innovative method integrates insights from multiple climate models under two Representative Concentration Pathways (RCP4.5 and RCP8.5), along with LULC projections generated through the Cellular Automata Markov model. By combining the strengths of the Soil and Water Assessment Tool (SWAT) and artificial neural network (ANN) techniques, the study aims to improve the accuracy of sediment yield simulations in response to changing environmental conditions. The non-linear autoregressive with external input (NARX) method was adopted for the ANN component of the hybrid model. The adoption of the hybrid SWAT-ANN approach appears to be particularly effective in improving the accuracy of sediment yield simulation compared to using the SWAT model alone, as evidenced by the higher coefficient of determination value of 0.74 for the hybrid model compared to 0.35 for the standalone SWAT model. In the context of the RCP4.5 scenario, during 2075-99, the study noted a 29.34% increase in sediment yield, accompanied by simultaneous rises of 42.74% in discharge and 27.43% in rainfall during the Indian monsoon season, spanning from June to September. In contrast, under the RCP8.5 scenario, for the same period, the increases in sediment yield, discharge, and rainfall for the monsoon season were determined to be 116.56%, 103.28%, and 64.72%, respectively. The present study's comprehensive analysis of the factors influencing sediment supply in the Puthimari River basin fills an important knowledge gap and provides valuable insights for designing proactive flood and erosion management strategies. The findings from this research are crucial for understanding the vulnerability of the Puthimari basin to climate and land use changes, and by incorporating these findings into policy and decision-making processes, stakeholders can work towards enhancing resilience and sustainability in the face of future hydrological and environmental challenges.

Application of Rainfall-Runoff Simulation Based on the NARX Dynamic Neural Network Model

The research into rainfall-runoff plays a very important role in water resource management. However, runoff simulation is a challenging task due to its complex formation mechanism, time-varying characteristics and nonlinear hydrological dynamic process. In this study, a nonlinear autoregressive model with exogenous input (NARX) is used to simulate the runoff in the Linyi watershed located in the northeastern part of the Huaihe river basin. In order to better evaluate the performance of NARX, a distributed hydrological model, TOPX, is used to simulate the discharge as a reference, and runoff classification by cluster analysis is used to further improve the accuracy of runoff simulation. Based on the four statistics indexes of the Nash–Sutcliffe efficiency (NSE), correlation coefficient (CC), root mean square error (RMSE) and mean relative bias (Bias), the NARX model is capable of simulating the rainfall-runoff dynamic process satisfactorily, although there is a little underestimation of the peak flow. After runoff classification, underestimation has been improved, and discharge simulation driven by NARX based on runoff classification (C-NARX) is well consistent with the observation. It is feasible to take it as a promising method, which also can be seen as a good reference and replacement for the current rainfall-runoff simulation.