Interpretable Dual-Stream Learning for Local Wind Hazard Prediction in Vulnerable Communities

TL;DR

This paper introduces an interpretable dual-stream learning framework combining numerical weather data and textual narratives to improve localized wind hazard prediction in vulnerable communities, enhancing transparency and operational trust.

Contribution

It presents a novel dual-stream model integrating structured and unstructured data for wind hazard prediction tailored to underserved communities, with interpretability features.

Findings

Significant performance improvements over traditional models

Enhanced transparency through sensitivity and ablation studies

Effective risk assessment at the community level

Abstract

Wind hazards such as tornadoes and straight-line winds frequently affect vulnerable communities in the Great Plains of the United States, where limited infrastructure and sparse data coverage hinder effective emergency response. Existing forecasting systems focus primarily on meteorological elements and often fail to capture community-specific vulnerabilities, limiting their utility for localized risk assessment and resilience planning. To address this gap, we propose an interpretable dual-stream learning framework that integrates structured numerical weather data with unstructured textual event narratives. Our architecture combines a Random Forest and RoBERTa-based transformer through a late fusion mechanism, enabling robust and context-aware wind hazard prediction. The system is tailored for underserved tribal communities and supports block-level risk assessment. Experimental results show significant performance gains over traditional baselines. Furthermore, gradient-based sensitivity and ablation studies provide insight into the model's decision-making process, enhancing transparency and operational trust. The findings demonstrate both predictive effectiveness and practical value in supporting emergency preparedness and advancing community resilience.