Satellite-Based Damage Assessment and Prioritization of Humanitarian Aid in the Canadian High Arctic Following Severe Ice Events
Authors
-
Emily White
Global Dynamics Corp.
-
Frank Brown
Pinnacle College
https://orcid.org/0009-0000-0000-0000
Abstract
Following catastrophic ice storms and blizzards, rapidly assessing damage and prioritizing limited aid distribution in the vast, remote Canadian High Arctic is logistically daunting. This study develops a robust Machine Learning (ML) algorithm specifically for rapid damage detection. The model processes Sentinel-1 Synthetic Aperture Radar (SAR) imagery, which effectively penetrates cloud cover and operates independent of daylight. The algorithm achieves a validated classification accuracy of 92% in differentiating structural damage (e.g., collapsed communication towers, compromised shelters) from benign natural ice cover. This technology enables international and national humanitarian organizations to bypass time-consuming ground reconnaissance, providing real-time, high-confidence situational awareness necessary to prioritize aid deployment to the most critically impacted areas within hours of an event's passage.
Copyright (c) 2025 Emily White, Frank Brown (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.