The city of Banjarbaru has been designated as the capital of South Kalimantan Province, a status that has the potential to trigger changes in land cover as part of regional development. Therefore, a study of land cover changes supervised the classification approach. The remote sensing data used were Landsat images from 2015 and 2025, with classification classes applied to vegetation, bare land, built-up land, and water bodies. Accuracy testing was carried out using a confusion matrix, the result of which indicated that the accuracy value was within the range permitted (>75%) by the Indonesian National Institute of Aeronautics and Space, so the classification result was considered satisfactory for further analysis. The dynamics of land change in the city of Banjarbaru have been quite significant over the last 10 years. The changes are visible in built-up land, which has changed by around 307.81%, followed by a decrease in bare land area in 2025. These changes indicate that a significant amount of open land has been converted into built-up land. This phenomenon indicates the potential for classification bias. Areas with sparse vegetation or hardened soil, such as post-fire land, former mining sites, or dry rice fields, may be misclassified as built-up areas because their spectral characteristics resemble surfaces like concrete or asphalt. For example, dry rice fields during the dry season or after harvest often show high reflectance and a homogeneous texture, making them appear similar to open land. These conditions were further influenced by the El NiƱo event in 2015.
Copyrights © 2025