Coupling Random Forest, Allometric Scaling, and Cellular Automata to Predict the Evolution of LULC under Various Shared Socioeconomic Pathways

Accurately estimating land-use demand is essential for urban models to predict the evolution of spatial morphology. Due uncertainties inherent in socioeconomic development, accurate forecasting remains a daunting challenge. The present study proposes modeling framework determine scaling relationship between population and area simulates spatiotemporal dynamics land use cover (LULC). An allometric (AS) law Markov (MK) chain are used variations LULC. Random forest (RF) cellular automata (CA) serve calibrate transition rules change LULC realize its micro-spatial allocation (MKCARF-AS). Furthermore, this research uses several shared pathways (SSPs) as scenario storylines. MKCARF-AS model changes under various SSP scenarios Jinjiang City, China, from 2020 2065. results show that figure merit (FoM) FoM improve by 3.72% 4.06%, respectively, compared with MKCAANN during 2005–2010 simulation period. For 6.28% discrepancy predicted actual over period 2005–2010, degrades 21.42%. growth permanent City follows an exponent 0.933 2005–2020, relative residual R2 0.0076 0.9994, respectively. From 2065, estimated 19.4% greater than SSP5. At township scale, different produce significantly distributions expansion rates. By coupling random scaling, substantially improves use.