Land Change Prediction Model in the Mamasa Watershed, Powerhouse of South Sulawesi

Image: article cover, from Utami A et al. (2025), Jurnal Wasian, licensed CC BY 4.0.

A research team from Universitas Sulawesi Barat has predicted land cover changes in the Mamasa Sub-watershed, located on the border of West Sulawesi and South Sulawesi, up to 2031. Covering approximately 116,108 hectares, this area supplies water to the Bakaru Hydroelectric Power Plant (PLTA), which is the main source of electricity for South Sulawesi. The findings were published in the Jurnal Wasian in 2025.

The study compared two machine learning-based prediction methods: Artificial Neural Network-Cellular Automata (ANN-CA) and Logistic Regression-Cellular Automata (LR-CA). The ANN-CA method achieved an accuracy rate of 83.67 per cent, slightly higher than LR-CA, which reached 82.35 per cent. Both models projected land cover changes for 2026 and 2031, but their predictions remained below actual values for large land classes such as forest, shrubland, paddy fields, and grassland. Conversely, accuracy was higher for small land classes such as swamp forest, plantations, settlements, and teak forest.

According to the study, land cover changes from forest, shrub, and grassland to agricultural land can disrupt the water balance in the watershed. If this condition continues, water supply to the Bakaru PLTA could decline and affect electricity production. The research team noted that their model did not yet incorporate socio-economic variables such as population size and income levels, which also influence land change. Further research is expected to analyse the influence of these variables.

Source and how to cite

This story is based on the research article Will Indonesia's Forests Survive Development Pressure? Machine Learning Predictions for Energy-Critical Tropical Watersheds by Widyanti Utami A, Irlan Irlan, Nur Hilal A Syahrir and Rosmaeni Rosmaeni, published in Jurnal Wasian Vol. 12 No. 01 (2025). The full article is free to read on the Jurnal Wasian website.

Cite the original article (APA 7):

Utami A, W., Irlan, I., Syahrir, N. H. A., & Rosmaeni, R. (2025). Will Indonesia's Forests Survive Development Pressure? Machine Learning Predictions for Energy-Critical Tropical Watersheds. Jurnal Wasian, 12(01), 35–43. https://doi.org/10.62142/hjs6a555

If you use these findings in coursework, a thesis or other writing, cite the original article above, not this news story. The original article is licensed CC BY 4.0.