Artificial intelligence-based approach to study the impact of climate change and human interventions on groundwater fluctuations

Abstract

Water resource management is highly impacted by variations in rainfall, maximum and minimum temperature, and potential evapotranspiration. The rice area is also a key aspect for groundwater declination due to high-water consuming crop. Groundwater in Central Punjab has declined at an alarming rate over the last two decades. The decisions regarding water resource management need accurate information for the groundwater level. Therefore, to explore the main reason for the depletion of groundwater, it is essential that the most influential factors responsible for groundwater depletion should be addressed. A study was conducted in Central Punjab by using artificial neural network (ANN) and multiple linear regression (MLR) models during 1998–2018 to forecast the groundwater depth. ANN performed better than MLR. The sensitivity analysis showed that tubewell density, rice area, and rainfall are highly responsible for groundwater fluctuation. HIGHLIGHTS In the present study, both climatic and human-induced factors were taken for groundwater modeling.; Artificial neural network, a complex phenomenon was used to forecast groundwater depth.; Python was used for groundwater modeling.; ANN was found to be more accurate than MLR.;

Keywords

• climate
• mann–kendall
• rainfall
• sen's slope
• temperature