Clustering of Provinces in Indonesia Based on Land Cover and Environmental Indicators using PAM Clustering

Land cover and environmental conditions in Indonesia play a strategic role in maintaining ecosystem sustainability, biodiversity, and supporting sustainable development. However, pressures arising from economic development, population growth, land conversion, and ecosystem degradation have resulted in significant environmental disparities across provinces. These variations necessitate the mapping and clustering of regions based on environmental indicators so that the characteristics, levels of pressure, and management needs of each province can be understood in a more systematic and structured manner. This study aims to classify 33 provinces in Indonesia based on land cover and environmental indicators, including the percentage of protected forest, mangrove realization, land cover quality index, conservation land area, forest rehabilitation, and hotspot density as an indicator of environmental pressure. The Partitioning Around Medoids (PAM) method was applied to standardized data due to its ability to produce clusters that are more robust to the presence of outliers, which are commonly found in environmental data. The optimal number of clusters was determined using the Elbow and Silhouette methods. The results indicate that Indonesian provinces can be grouped into six clusters with distinct environmental characteristics, ranging from provinces with relatively good land cover and conservation conditions to those experiencing high environmental pressure. Overall, this clustering provides a more comprehensive representation of the patterns and heterogeneity of environmental conditions across provinces and may serve as a basis for formulating more specific, targeted, and regionally characteristic-based environmental management policies.

Keywords: cluster analysis, environmental indicators, land cover, Partitioning Around Medoids (PAM)

Copyright (c) 2026 Journal of Dinda : Data Science, Information Technology, and Data Analytics

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.