Prediction of land use changes via the cellular automata-Markov model in East Luwu Regency

Abstract

The increasing involuntary demand for land caused by the aggrandization and economic expansion of inhabitants has accelerated land conversion in many developing regions. This study aims to project land use substitutions in East Luwu Regency by 2043 via a CA-Markov modeling approach. Spaceborne imagery from the Landsat 7 ETM+ and Landsat 8 OLI/TIRS datasets was visually interpreted to generate historical land use maps for 2003, 2013, and 2023. The projection analysis involved two key phases: calculating transition probabilities through the Markov chain method, followed by spatial modeling of future land use patterns via a cellular automata framework. Several driving factors were considered, including proximity to roads, settlements, and rivers; elevation; population density; and slope gradient. Projections show that forest areas will continue to decline until 2043 due to increased land conversion by local communities. Conversely, mining, open land, and settlements are projected to expand, especially in areas with high accessibility and high population density. The CA-Markov representation demonstrated a high-pitched Kappa prognostic value of 0.8943 and proved effective in analyzing and forecasting the spatial-temporal kinetics of land-use change. These discoveries accommodate influential perspicacities for community planners and policymakers in underdeveloped sustainable spatial planning and forest management strategies that balance community needs with environmental conservation in East Luwu Regency.