Assessment of the Relationship Between Drought and Malnutrition

Understanding how hydroclimatic extremes translate into human vulnerability is essential for designing effective adaptation strategies in drought-prone regions. This study aims to investigate the relationship between drought conditions and malnutrition outcomes across multiple regions using a combination of climate diagnostics, statistical modelling, and machine learning approaches.

We start with a global assessment linking historical drought events to malnutrition indicators using open-source public-health. To support this analysis, we assemble a multi-source dataset integrating meteorological drought indices, vegetation and soil-moisture indicators, and subnational malnutrition metrics. Our methodological framework first characterizes drought variability across temporal scales to identify dominant spatial and temporal patterns of moisture deficits. We then explore the sensitivity of malnutrition indicators to drought stress using nonlinear and lag-aware statistical techniques, complemented by machine learning models to capture potential complex relationships. This approach enables us to begin isolating the pathways through which hydroclimatic anomalies may influence nutritional outcomes, while accounting for confounding socioeconomic factors. The long-term objective is to translate these insights into a prediction tool for improving anticipatory action.

This initial research effort seeks to contribute to the broader understanding of how climate extremes interact with public-health vulnerability. By developing an analytical framework and openly accessible datasets, this work aims to support disaster-risk management and health preparedness in the face of increasingly complex and escalating climate-related risks in developing more timely and targeted responses.