Geographia Technica, Vol 21(2), Special Issue: Artificial Intelligence Applications in Geography, 2026, pp. 51-74
AN INTEGRATED ANN-CA MODEL FOR LAND USE CHANGE PREDICTION AND FLOOD RISK MITIGATION: A CASE STUDY OF ENREKANG REGENCY, INDONESIA
Uca SIDENG 
, Nurul Afdal HARIS , Mustari S. LAMADA
ABSTRACT: Flood disasters in Enrekang Regency, South Sulawesi Province, have caused significant material losses and disrupted community activities due to the region’s unique geographical characteristics with undulating and mountainous topography. The Saddang River, as one of the main rivers in South Sulawesi, flows through this area, making it highly vulnerable to flooding, especially during the rainy season. Rapid land cover changes due to human activities such as settlement expansion, agriculture, and deforestation have increasingly elevated flood risks. Land conversion from forests to agricultural and settlement areas reduces water absorption capacity and increases surface runoff. Currently, flood management in Enrekang Regency remains reactive, with budgets allocated more for post-disaster response than for mitigation and prevention. This research develops an integrated Artificial Neural Network Cellular Automata (ANN-CA) model for land use change prediction and flood risk mitigation. The model integrates remote sensing technology, ANN-CA modeling, and Geographic Information Systems (GIS) to predict future land use changes and identify high flood-risk areas. The methodology involves satellite image acquisition (2010-2020), land cover change extraction, ANN training, CA configuration, model validation (Accuracy >85%, Kappa >0.8), and integration with flood risk factors. Results show that the model can effectively predict land use changes with high accuracy, providing valuable spatial information for flood mitigation planning. The predicted land use map for 2030 indicates significant expansion of built-up areas in flood-prone zones, necessitating immediate policy interventions. This research contributes to the development of predictive and preventive flood management approaches, offering a scientific basis for spatial planning and disaster risk reduction in mountainous regions.
Keywords: Artificial Neural Network; Land use change prediction; Flood risk mitigation; Remote sensing; Spatial analysis.
Full article here