Hydrological drought prediction and its influencing features analysis based on a machine learning model
Authors:
Open Access
- 4 November 2025
- journal article
- research article
- Published by Copernicus GmbH in Natural Hazards and Earth System Sciences
- Vol. 25 (11) , 4299-4316
Abstract
Predicting future drought conditions are crucial for effective disaster management. In this study, a machine learning framework is proposed to predict hydrological drought in the Huaihe River Basin, China. The Extreme Gradient Boosting (XGBoost) model is applied to predict four drought categories in 28 grid regions for one-month prediction, using 26 features for monthly and 18 for seasonal predictions. The framework also integrates the Shapley Additive Explanation (SHAP) variable importance index to infer drought prediction features. The model achieves 79.9 % accuracy in classifying droughts, with the Standard Precipitation Index (SPI) being the most influential feature. The SHAP values of SPI are 0.360, 0.261, 0.169, and 0.247 for spring, summer, autumn, and winter, respectively. Soil moisture content and evapotranspiration are particularly affected in spring and autumn, while large-scale climatic features are more significant in summer and winter. Overall, this study offers valuable decision support for regional drought management and water resource allocation.Keywords
All Related Versions
Funding Information
- Natural Science Foundation of Jiangsu Province (BK20250906)
This publication has 59 references indexed in Scilit:
- AI-driven U.S. drought prediction using machine learning and deep learningClimate Dynamics, 2025
- Impact of driving factors on drought propagation: perspectives on rainfall deficit and excessive evaporationClimate Dynamics, 2025
- Delayed Response of Soil Moisture and Hydrological Droughts to Meteorological Drought Over East AsiaInternational Journal of Climatology, 2025
- AMO modulation of interdecadal background of persistent heavy rainfall in summer over the Huaihe River BasinClimate Dynamics, 2024
- Spatiotemporal characteristics of extreme hydrometeorological events and its potential influencing factors in the Huaihe River Basin, ChinaStochastic Environmental Research and Risk Assessment, 2023
- Artificial intelligence application in drought assessment, monitoring and forecasting: a reviewStochastic Environmental Research and Risk Assessment, 2021
- Deep Learning and Machine Learning in Hydrological Processes Climate Change and Earth Systems a Systematic Review2020
- Spatiotemporal characteristics of droughts and floods in northeastern China and their impacts on agricultureStochastic Environmental Research and Risk Assessment, 2018
- Drought forecasting in Luanhe River basin involving climatic indicesTheoretical and Applied Climatology, 2016
- Erratum: Drought under global warming: a reviewWires Climate Change, 2012