Integrated High-Resolution, Continental-Scale Land Change Forecasting

Predicting future land change is crucial in anticipating societal and environmental impacts and informing responses. We developed an integrated, high resolution, land-change model and forecasted continental land change for Australia for the years 2020, 2025 and 2030 for Cropland, Forest, Grassland, and Built-up land-uses. We combined a set of drivers and trained land-use suitability models using a random forest classifier. Thirty-meter resolution, per-class suitability layers were generated for the country and used for allocating land-use. Land-use was first projected for 2015 for validation purposes, then it was projected for 2030, allocating future land demand extrapolated via compositional linear regression. Accuracy at national level was ~94%. Forecasted land change showed increases in Grassland and Built-up areas, and decreases in Forest and Cropland. Our modelling approach expands the current capabilities of large-scale land-change models and provides new multiclass land forecasts for Australia that can inform land policy at multiple scales in Australia.