Advancements in Wildfire Detection and Prediction: An In-Depth Review
Reem Salman1, Ali Karouni2, Elias Rachid3, Nizar Hamadeh4
1Reem Salman, Lebanese University, EDST, Lebanon, Beirut.
2Ali Karouni, Lebanese University Faculty of Technology, Lebanon, Saida.
3Elias Rachid, Saint-Joseph University, Ecole Supérieure D’ingénieurs de Beyrouth, Lebanon, Beirut.
4Nizar Hamadeh, Lebanese University Faculty of Technology, Lebanon, Saida.
Manuscript received on 06 December 2023 | Revised Manuscript received on 27 December 2023 | Manuscript Accepted on 15 January 2024 | Manuscript published on 30 January 2024 | PP: 6-15 | Volume-13 Issue-2, January 2024 | Retrieval Number: 100.1/ijitee.B97741320124 | DOI: 10.35940/ijitee.B9774.13020124
Open Access | Editorial and Publishing Policies | Cite | Zenodo | OJS | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Wildfires pose a significant hazard, endangering lives, causing extensive damage to both rural and urban areas, causing severe harm for forest ecosystems, and further worsening the atmospheric conditions and the global warming crisis. Electronic bibliographic databased were searched in accordance with PRISMA guidelines. Detected items were screened on abstract and title level, then on full-text level against inclusion criteria. Data and information were then abstracted into a matrix and analyzed and synthesized narratively. Information was classified into 2 main categories- GIS-based applications, GIS-based machine learning (ML) applications. Thirty articles published between 2004 and 2023 were reviewed, summarizing the technologies utilized in forest fire prediction along with comprehensive analysis (surveys) of their techniques employed for this application. Triangulation was performed with experts in GIS and disaster risk management to further analyze the findings. Discussion includes assessing the strengths and limitations of fire prediction systems based on different methods, intended to contribute to future research projects targeted at enhancing the development of early warning fire systems. With advancements made in technologies, the methods with which wildfire disasters are detected have become more efficient by integrating ML Techniques with GIS.
Keywords: Wildfire, Detection, Prediction, Machine Learning, GIS
Scope of the Article: Machine Learning