ABSTRACT This study examines the interplay between climate change, violent conflict and forced migration in the Middle East and North Africa (MENA), focusing on asylum flows to the European Union (EU). By integrating high‐resolution climate, conflict and socioeconomic data spanning 2000 to 2023, we develop a comprehensive empirical framework to identify the key drivers of cross‐border migration. Using a machine learning approach with a Random Forest Model (RFM), we compare its predictive performance against the traditional Gravity Model (GM). The RFM, which captures nonlinear relationships and variable interactions, significantly outperforms the GM, explaining over 53% of the variance in migration patterns. Our findings highlight the predominant influence of conflict and economic instability as primary predictors, while climate‐related stressors, particularly drought severity and agricultural decline, function as threat multipliers, exacerbating existing vulnerabilities and amplifying displacement pressures. The analysis demonstrates that climate conditions alone are insufficient to explain migration surges; rather, it is the convergence of environmental, political and socio‐economic fragilities that drives forced mobility. This research offers critical insights for migration governance and underscores the importance of integrated policy responses that address both immediate humanitarian needs and long‐term structural resilience in the face of accelerating climate change. A key limitation is that the analysis focuses exclusively on asylum applications from MENA countries to the EU, does not capture internal or regional displacement and relies on country‐level, predictive modelling that cannot establish causal relationships or reflect subnational climate–conflict heterogeneity.
Mathbout et al. (Mon,) studied this question.