The project investigates the impact of rainfall on groundwater levels, considering factors such as population, water usage, and climate conditions. Rainfall significantly influences groundwater levels, particularly in karst areas, where the rapid response of groundwater to rainfall is observable. The study explores the fractal behavior of rainfall and its subsequent effects on groundwater recharge and level fluctuations. By analyzing the relationship between rainfall patterns, groundwater exploitation, and climate variables, this project aims to understand the processes driving groundwater dynamics. The Northern Huangqihai Basin, where both rainfall changes and groundwater extraction contribute to groundwater level fluctuations, serves as a case study. Here, rainfall changes account for 22.08%, while groundwater exploitation contributes 77.92%. The project employs advanced machine learning algorithms, including Random Forest, Gradient Boost Regressor, and Linear Regression, to model and predict groundwater availability. Additionally, techniques like Explainable AI (e.g., SHAP and LIME) will be used to interpret the model predictions and offer insights into the key factors influencing groundwater levels. Exploratory Data Analysis (EDA) will further inform the modeling process, uncovering patterns and relationships between rainfall, groundwater usage, and other climatic variables. The expected output is a predictive tool that estimates future groundwater availability, aiding in effective water resource management. This system will be developed with a Python backend, while the frontend will utilize HTML, CSS, and JavaScript for user interaction and visualization.

Keywords

Rainfall, Groundwater levels, Machine learning, Random Forest, Gradient Boost Regressor, Linear Regression, Explainable AI, SHAP, LIME, EDA, Water management, Hydrology.