Wildfire Drivers, Impacts, and Predictive Modeling: Opportunities in Sensing and Machine Learning
GeoHealth · 2026-11 · 10 (11)
DOI: 10.1029/2025gh001734 ↗Kurz gefasst
Übersichtsarbeit in der Zeitschrift GeoHealth, online erschienen am 11.11.2026. Verfasst von Jasmin Yujen Tung und 7 weiteren Autor:innen. Beteiligte Einrichtung: Zachry Department of Civil and Environmental Engineering Texas A&M University College Station TX USA. Die Arbeit nennt 447 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation.
Abstract (Originalsprache)
Wildfire prediction and management remain inherently challenging because fire behavior is governed by complex, nonlinear interactions among the atmosphere, biosphere, geosphere, and human influences. Intensified by climate change and anthropogenic activities, wildfires are becoming more severe in portions of the world, degrading air quality, threatening public health, generating substantial economic losses, and producing secondary hazards such as debris flows and water contamination. This review synthesizes recent advances in multi‐sensor remote sensing and machine learning approaches for wildfire management, focusing on three key domains: pre‐fire risk monitoring, active‐fire detection, and post‐fire severity assessment. It highlights how emerging sensing technologies and predictive models integrate diverse environmental variables, remotely sensed spectral indices, and geospatial data sets to improve wildfire forecasting, situational awareness, and decision‐making. Despite these advances, important limitations remain, including constrained physical interpretability and operational reliability in purely data‐driven models, limited transferability of empirical approaches across regions, and sensor‐related constraints such as sparse revisit times, radiometric inconsistencies, and atmospheric interference from clouds and smoke that affect wildfire monitoring and post‐fire assessment. Overall, the review concludes that bridging the gap between research and operational implementation will require hybrid physics‐informed modeling frameworks and multi‐platform geospatial data fusion systems capable of delivering more accurate, interpretable, scalable, and operationally robust wildfire prediction and management.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Zachry Department of Civil and Environmental Engineering Texas A&M University College Station TX USA.
- Department of Civil Engineering Boise State University Boise ID USA.
- Institute for Water, Environment, and Health (UNU-INWEH) United Nations University Richmond Hill ON Canada.
- Department of Civil & Environmental Engineering University of California Irvine CA USA.
- Department of Earth System Science University of California Irvine CA USA.
- Department of Management of Complex Systems University of California, Merced Merced CA USA.
ORCID-Profile: Jasmin Yujen Tung, Ashok Mishra, Mojtaba Sadegh, Amitesh Sabut, Amir Aghakouchak, John Abatzoglou, Naresh Devineni, A Sankarasubramanian
Förderung
- National Science Foundation
Zitieren
APA
Tung, J. Y., Mishra, A., Sadegh, M., Sabut, A., Aghakouchak, A., Abatzoglou, J., Devineni, N., & Sankarasubramanian, A. (2026). Wildfire Drivers, Impacts, and Predictive Modeling: Opportunities in Sensing and Machine Learning. GeoHealth, 10(11). https://doi.org/10.1029/2025gh001734
Vancouver
Tung JY, Mishra A, Sadegh M, Sabut A, Aghakouchak A, Abatzoglou J, et al. Wildfire Drivers, Impacts, and Predictive Modeling: Opportunities in Sensing and Machine Learning. GeoHealth. 2026;10(11). doi:10.1029/2025gh001734
BibTeX
@article{Tung2026,
author = {Tung, Jasmin Yujen and Mishra, Ashok and Sadegh, Mojtaba and Sabut, Amitesh and Aghakouchak, Amir and Abatzoglou, John and Devineni, Naresh and Sankarasubramanian, A},
title = {Wildfire Drivers, Impacts, and Predictive Modeling: Opportunities in Sensing and Machine Learning},
journal = {GeoHealth},
year = {2026},
volume = {10},
number = {11},
doi = {10.1029/2025gh001734}
}
EUVIMED