Justin Andrew Johnson, Sumil Thakrar

Understanding how economic systems and ecosystems interact across space is crucial to ensure societal needs are met without compromising environmental quality. Spatially explicit economic models usually describe human activities and ensuing land-use dynamics at a resolution that this is too coarse (typically 10-1000 regions) to understand how these affect many biophysical processes, including ecosystem service supply (which requires billions of 300m or finer grid-cells). Several land-use change models exist that allocate coarse land-use changes at finer spatial scales, but there is no general, global, computationally tractable tool for downscaling to the spatial scale appropriate for understanding landscape and ecosystem change across different land-system scenarios. Here, we present a new model, the Spatial Economic Allocation Landscape Simulator (SEALS), which generates global, high resolution land-use, land-cover maps from coarse projections of land-use change. SEALS advances land-use change modeling research in several ways: it uses a novel machine-learning algorithm to empirically calibrate its parameters; it is a generalized approach that can be applied to very different types of projections within earth-economy models; and it is computationally efficient, generating global results at 300m resolution in approximately 1 hour on a laptop computer. We develop SEALS, and apply it to downscale 6 global scenarios of land-use change from the Shared Socioeconomic Pathways (SSPs) to 300m resolution, which we make publicly available, allowing researchers to project global, fine-scale changes in ecosystem services. For carbon storage and sequestration, we find that, from 2021 to 2100, the SSPs considered result in reductions in carbon storage and sequestration by 3-21%, and baseline results are 9.17% lower than conventional estimates using low-resolution inputs. Consistent with prior estimates, we find that SSPs 3 and 4, which lead to large amounts of tropical deforestation, result in the largest carbon losses, further exacerbating the climate impacts.