Aim of the study:
Drought is one of the most significant environmental challenges of modern times, greatly affecting the availability of groundwater—a crucial resource in water management. The aim of this study is to demonstrate the effectiveness of kernel estimators in determining the distribution of minimum groundwater levels.

Material and methods:
The study analyzes the distribution of minimum groundwater levels in the Wielkopolska region, an area particularly vulnerable to water shortages, during the hydrological period of 2001–2020. Data from 55 monitoring stations were used to compare the effectiveness of classical parametric methods with kernel estimation techniques proposed by the author. Additionally, three distinct error metrics were applied to assess the performance of the methods.

Results and conclusions:
The results indicate that kernel estimation outperforms parametric methods in terms of accuracy for deeper wells with a confined aquifer. For unconfined aquifers, no clear advantage of either method was observed. The findings may contribute to improving forecasting methods for extreme groundwater levels and support water resource management in the context of increasing drought risk.