Evaluation and Prediction of Land Cover Changes in the Jangkok Watershed Using the Approach Ca-Ann

Jangkok watershed forest area evaluation Landsat-8 OLI land cover change

Authors

May 11, 2026

Downloads

This study addresses the increasing concern over deforestation and land cover changes in watershed areas, particularly in the Jangkok Watershed, West Nusa Tenggara, Indonesia. Rapid land conversion driven by population growth, agricultural expansion, and accessibility has contributed to forest degradation, potentially disrupting hydrological functions and ecosystem stability. Therefore, this research aims to evaluate land cover changes from 2019 to 2024 and to predict future changes until 2039 using an integrated modeling approach. The study employs remote sensing data derived from Landsat-8 OLI (2019) and Landsat-9 OLI-2 (2022 and 2024), processed through image pre-processing, classification, and accuracy assessment using ground checks, error matrix, and Kappa statistics. Furthermore, land cover change analysis is conducted using transition matrices, while the direction of change is evaluated using Curve Number (CN) and Soil Hydrology Group (KHT). Future predictions are generated using the Cellular Automata–Artificial Neural Network (CA–ANN) model based on driving factors such as distance to roads, rivers, and slope. The results indicate a significant decrease in primary and secondary dryland forests, accompanied by a substantial increase in mixed garden areas. Negative land cover changes dominate, leading to increased surface runoff and reduced infiltration capacity. The CA–ANN model demonstrates good predictive performance (Kappa = 0.78), projecting continued forest decline until 2039. In conclusion, land cover changes in the Jangkok Watershed are predominantly driven by human activities and pose risks to environmental sustainability. Strategic land management and conservation policies are essential to mitigate further degradation and preserve watershed functions.