Landscape change in the Ci Kapundung upper water catchment area over recent decades has increased the volume of rainfall runoff, increasing the incidence of flooding in the Bandung Basin. At the same time, climate change affects rainfall variability in Indonesia, causing higher frequencies of extreme rainfall and drought in almost all regions of the country. This study develops and assesses four scenarios for the future spatial plan in the catchment area, which is part of a research project on flood risk in the Bandung Basin. The scenarios were created based on recent land use changes within the area, current spatial policies, ecological design principles, and the geodesign framework. All scenarios were simulated using Land Change Modeler (LCM), which applies a combined cellular automata and Markov model (CA-Markov), and a multilayer perceptron (MLP) neural network, to project future landscape forms for 2030. In this simulation, the results were assessed to see the variations of land cover composition. CA-Markov models have been widely used to predict urban growth. Modeling applications for forest cover change simulation have rarely been explored. Results from LCM show that there is no significant difference in the percentages of developed areas, mixed plant communities, and conifers in 2030 in all scenarios. However, the spatial arrangement of land cover varies with each scenario. In the first scenario, for example, the disperse settlement pattern is projected to occur in the watershed in 2030, including in the areas with steep slopes and near the rivers, whereas in other scenarios, specific areas are restricted to be built. The study suggests that further analysis of hydrological impacts of each scenario is needed to ascertain which scenario can effectively reduce flood risk.