Forest and Land Fire Vulnerability Assessment and Mapping using Machine Learning Method in East Nusa Tenggara Province, Indonesia
Dublin Core
Title
Forest and Land Fire Vulnerability Assessment and Mapping using Machine Learning Method in East Nusa Tenggara Province, Indonesia
Subject
East Nusa Tenggara, forest and land fires, feature selection, machine learning, mapping
Description
Forest and land fires (FLF) severely damage forest ecosystems and reduce their functionality.
Predicting areas prone to fires is crucial for effective management and prevention. Machine learning
(ML) has shown potential in this field. By 2022, East Nusa Tenggara (NTT) experienced the highest
incidence of fires in Indonesia, with 70,637 hectares burned. This study evaluates NTT's FLF
vulnerability using seven ML methods: Gaussian Naive Bayes, Support Vector Machine, Logistic
Regression, Artificial Neural Network, Random Forest, Gradient Boosting Machine, and Extreme
Gradient Boost (XGB). A geospatial dataset combining NTT's 2022 fire data and fourteen fire-related
factors was developed with ArcGIS. Using the Information Gain Ratio for feature selection, twelve
key features were identified: Elevation, Slope angle, Slope Aspect, Plan Curvature, Land Cover,
NDVI, Distance to Road, Distance to Buildings, Annual Rainfall, Average Temperature, Wind Speed,
and Relative Humidity. The XGB model performed best, with AUC values of 0.959 for training and
0.743 for testing. The resulting vulnerability map revealed key fire factors: low elevation, gentle
slopes, curved terrain, forest cover, poor vegetation health, human activity, distant firefighting
resources, low rainfall, high temperatures, high wind speeds, and low humidity. Recommendations
include land management, fire-resistant vegetation, policy enforcement, community education, and
infrastructure enhancement.
Predicting areas prone to fires is crucial for effective management and prevention. Machine learning
(ML) has shown potential in this field. By 2022, East Nusa Tenggara (NTT) experienced the highest
incidence of fires in Indonesia, with 70,637 hectares burned. This study evaluates NTT's FLF
vulnerability using seven ML methods: Gaussian Naive Bayes, Support Vector Machine, Logistic
Regression, Artificial Neural Network, Random Forest, Gradient Boosting Machine, and Extreme
Gradient Boost (XGB). A geospatial dataset combining NTT's 2022 fire data and fourteen fire-related
factors was developed with ArcGIS. Using the Information Gain Ratio for feature selection, twelve
key features were identified: Elevation, Slope angle, Slope Aspect, Plan Curvature, Land Cover,
NDVI, Distance to Road, Distance to Buildings, Annual Rainfall, Average Temperature, Wind Speed,
and Relative Humidity. The XGB model performed best, with AUC values of 0.959 for training and
0.743 for testing. The resulting vulnerability map revealed key fire factors: low elevation, gentle
slopes, curved terrain, forest cover, poor vegetation health, human activity, distant firefighting
resources, low rainfall, high temperatures, high wind speeds, and low humidity. Recommendations
include land management, fire-resistant vegetation, policy enforcement, community education, and
infrastructure enhancement.
Creator
Hans Timothy Wijaya and Aniati Murni Arymurthy
Source
http://dx.doi.org/10.21609/jiki.v18i1.1304
Publisher
Faculty of Computer Science Universitas Indonesia
Date
2025-02-08
Contributor
Sri Wahyuni
Rights
e-ISSN : 2502-9274 printed ISSN : 2088-7051
Format
PDF
Language
English
Type
Text
Coverage
Jurnal Ilmu Komputer dan Informasi (Journal of Computer Science and Information)
Files
Collection
Citation
Hans Timothy Wijaya and Aniati Murni Arymurthy, “Forest and Land Fire Vulnerability Assessment and Mapping using Machine Learning Method in East Nusa Tenggara Province, Indonesia,” Repository Horizon University Indonesia, accessed May 23, 2025, https://repository.horizon.ac.id/items/show/8937.