top of page
AdobeStock_268518669.jpeg

New Publication on Natural Disasters, Human Migration, and Explainable AI in the United States

GEAR Lab is pleased to share a new publication in Journal of Environmental Management by Xiang Li, Yi Qiang, Xiao Huang, Lei Zou, and Heng Cai.


The paper, titled “Natural Disasters and Human Migration in the United States: Insights from Automated Machine Learning and Explainable AI,” examines how major natural disasters are associated with human migration patterns across the contiguous United States. Focusing on floods, hurricanes, wildfires, and tornadoes from 2000 to 2020, the study combines statistical modeling, automated machine learning, and explainable AI to better understand how disaster impacts relate to county-level net migration rates.
The paper, titled “Natural Disasters and Human Migration in the United States: Insights from Automated Machine Learning and Explainable AI,” examines how major natural disasters are associated with human migration patterns across the contiguous United States. Focusing on floods, hurricanes, wildfires, and tornadoes from 2000 to 2020, the study combines statistical modeling, automated machine learning, and explainable AI to better understand how disaster impacts relate to county-level net migration rates.

The research shows that disaster-related migration is shaped by both environmental hazards and broader socio-economic conditions. While factors such as housing, income, and local economic conditions remain important drivers of migration, the study finds that certain disasters, especially hurricanes, floods, and wildfires, provide meaningful signals for understanding population movement over time.



By using automated machine learning and SHAP-based explainable AI, the study offers a more interpretable way to examine complex disaster-migration relationships. This approach helps identify which disaster and socio-economic variables contribute most to migration predictions, while also revealing how these relationships vary across space and time.


This work contributes to growing research on climate adaptation, disaster resilience, and population dynamics. As natural hazards become more frequent and severe, understanding how disasters influence migration can support more informed planning, risk reduction, and long-term resilience strategies for communities across the United States.


Congratulations to the authors on this important contribution to environmental management, disaster science, migration studies, and explainable GeoAI.


Citation

Li, X., Qiang, Y., Huang, X., Zou, L., & Cai, H. (2026). Natural disasters and human migration in the United States: Insights from automated machine learning and explainable AI. Journal of Environmental Management, 406, 129796. https://doi.org/10.1016/j.jenvman.2026.129796

 
 
 

Comments


GearLab-Logo-White.png

Geospatial Exploration and Resolution (GEAR) Lab

© 2026 by The GEAR Lab

bottom of page