Spatially explicit predictions of land cover change using the Density-Functional Fluctuation Theory (DFFT) approach

Many existing models for predicting land cover change (LCC) rely on extensive spatial drivers and complex calibration. This study introduces Density-Functional Fluctuation Theory (DFFT) as an alternative, low-data and computationally efficient approach for spatially explicit LCC prediction. DFFT operates on binary landscapes consisting of focal and background cells and represents a site as a grid of blocks, each described by the proportion of focal cells it contains. From a single observed land-cover snapshot, the method infers a block-level density functional that encodes the statistical structure of local spatial configurations. Future density-grid patterns are generated using a stochastic Monte Carlo procedure that updates block densities toward configurations consistent with this functional and subject to a prescribed future global density. Predictions for multi-categorical LCC are obtained using a sequential binary procedure applied to individual land-cover categories. The approach does not rely on explicit transition rules, suitability surfaces, or spatial covariates. The method was tested on four National Land Cover Database (NLCD) sites between 2001 and 2019, covering urban expansion, two types of agricultural change, and wetland dynamics. Across multiple block sizes, predicted density grids show strong agreement with observed data, with mean absolute errors ranging from 0.012 to 0.101. Compared with commonly used approaches such as CA-Markov, DFFT is highly parsimonious and scalable, making it particularly suitable for large domains, data-scarce settings, and scenario-based forecasting.