Get started
Work

Assessment of forest fire vulnerability prediction in Indonesia: Seasonal variability analysis using machine learning techniques

Year: 2025
Type: article
Abstract: Forest fires significantly threaten Indonesia’s tropical forests, driven by complex interactions between natural and human factors. Large wildfires in 2015 and 2019, have highlighted the urgent need to understand the determinants of fire occurrence to effectively mitigate risks. This research aims to identify and analyze the main factors influencing forest fires in Kalimantan, Sumatra, and Papua during the rainy and dry seasons using machine learning techniques to create vulnerability prediction maps and categorize risk zones. Utilizing eight years (2017–2021)of historical forest fire data, this study considered several human, environmental, meteorological, and land use conditioning factors. Random forest (RF) and eXtreme Gradient Boosting (XGB) machine learning models were used to train and validate the dataset through hyperparameter tuning and 10-fold cross-validation for accuracy assessment. The findings suggest that both human and natural factors can trigger forest fires. Additionally, we identified ten key factors that significantly impact forest fires in the three study areas. These factors include altitude, population density, rainfall, soil moisture, NDMI, NDVI, distance from settlements, distance from roads, land surface temperature, and peatlands. This research highlighted the diverse characteristics of factors that determine forest fires and examined their impact on fire occurrences. The implications of our study can guide strategies to prevent forest fires, protect biodiversity, reduce carbon emissions, and allocate resources more effectively.
DOI: 10.1016/j.jag.2025.104435
Language: English
Open Access status: gold
Estimated APC paid: $3,550
OpenAlex ID: W4408063547
FWCI: 5.884
Cites: 83
Cited by: 14
Related to: 10