A comprehensive taxonomy for forest fire risk assessment: bridging methodological gaps and proposing future directions

• Published in: Environmental monitoring and assessment 2024-09, Vol.196 (9), p.825-825, Article 825
• Main Authors: Özcan, Zühal, Caglayan, İnci, Kabak, Özgür
• Format: Article
• Language: English
• Subjects: Atmospheric Protection/Air Quality Control/Air Pollution; Classification; Conservation of Natural Resources - methods; Decision analysis; domain; Earth and Environmental Science; Ecology; Ecotoxicology; Effectiveness; Environment; Environmental changes; Environmental Management; Environmental Monitoring - methods; Fire behavior; Fire hazards; Fire prevention; Fires; Forest fires; Forest management; Forestry; Forestry - methods; Forests; Hazard assessment; Hazard identification; landscapes; Learning algorithms; Machine Learning; Modelling; Monitoring/Environmental Analysis; multi-criteria decision making; Multiple criterion; Protected species; Regional development; Review; Reviews; Risk assessment; Risk Assessment - methods; species; Taxonomy; Wildfires
• ISSN: 0167-6369; 1573-2959; 1573-2959
• DOI: 10.1007/s10661-024-12982-8

Forest fire risk assessment plays a crucial role in the environmental management of natural hazards, serving as a key tool in the prevention of forest fires and the protection of various species. As these risks continue to evolve with environmental changes, the pertinence of contemporary research in this field remains undiminished. This review constructs a comprehensive taxonomic framework for classifying the existing body of literature on forest fire risk assessment within forestry studies. The developed taxonomy categorizes existing studies into 8 primary categories and 23 subcategories, offering a structured perspective on the methodologies and focus areas prevalent in the domain. We categorize a sample of 170 articles to present recent trends and identify research gaps in forest fire risk assessment literature. The classification facilitates a critical evaluation of the current research landscape, identifying areas in need of further exploration. Particularly, our review identifies underrepresented methodologies such as optimization modeling and some advanced machine learning techniques, which present routes for future inquiry. Moreover, the review underscores the necessity for model development that is tailored to specific regional data sets but also adaptable to global data resources, striking a balance between local specificity and broad applicability. Emphasizing the dynamic nature of forest fire behavior, we advocate for models that integrate the burgeoning field of machine learning and multi-criteria decision analysis to refine predictive accuracy and operational effectiveness in fire risk assessment. This study highlights the great potential for new ideas in modeling techniques and emphasizes the need for increased collaboration among research communities to improve the effectiveness of assessing forest fire risks.