# Predicting groundwater storage from seasonal managed aquifer recharge: insights from machine learning and explainable AI techniques

## TL;DR

This paper uses machine learning and explainable AI to predict groundwater storage after managed aquifer recharge, helping optimize water management in drought-prone areas.

## Contribution

A novel ML approach combining U-Net and XGBoost with explainable AI to model transient MAR effects and identify key factors influencing groundwater storage.

## Key findings

- U-Net and XGBoost accurately predicted MAR response and decay coefficient with R2 > 0.82 in the Baakse Beek catchment.
- Explainable AI (SHAP values) identified key management decisions and surface water properties affecting MAR effectiveness.
- The approach enables efficient large-scale scenario testing and optimization for sustainable groundwater management.

## Abstract

Managed Aquifer Recharge (MAR) is widely used to enhance groundwater storage and support sustainable water use. To support site selection and planning, Machine Learning (ML) models are increasingly used as computationally efficient surrogates for traditional numerical models. While ML has shown promise for steady-state simulations, capturing transient responses remains a challenge, yet these are essential for understanding how recharged water is retained through dry periods. In this study, we use ML to model transient MAR effects by decomposing the groundwater storage time series after recharge ceases into two components: the MAR-response and a decay coefficient, assuming exponential storage decline. This simplified representation captures long-term storage dynamics, with U-Net and XGBoost accurately predict these components (R2 > 0.82) for the Baakse Beek catchment in the sandy, drought-sensitive soils of the Netherlands. The trained models are computationally efficient, making large-scale scenario testing and optimization feasible. Explainable AI techniques, specifically SHAP values, were used to identify key site management decisions and surface water properties that control MAR effectiveness. These findings illustrate the potential of explainable AI and ML surrogate models to enhance the planning and optimization of MAR.