Mapping the global potential of onshore field-scale solar PV using positive-unlabeled deep learning

Accelerating the deployment of low-carbon solar photovoltaics (PV) can contribute to global decarbonization and mitigation of climate change. Yet the rapid expansion of solar PV could bring negative impacts on environment, biodiversity, and food security. A spatially explicit potential map of solar PV worldwide is beneficial in optimizing future deployments of solar PV projects to meet sustainable development goals. Here we assess the land suitability for PV development using positive-unlabeled deep learning and estimate the potential of generating capacity and decarbonization for onshore field-scale solar PV worldwide. We estimate that the area of suitable land for PV development is about 21,459,552 km2 globally, the potential generating capacity is 169.52 (±3.21) TW, and the potential annual carbon reduction afforded by solar PV is about 12 Gt. Rangeland contributes the largest proportion of estimated capacity (57.3 %), followed by cropland (28.3 %). About 68 % of the potential annual carbon reduction is contributed by China, the United States, India, Russia, and Japan. Our results can facilitate decision-makers to plan PV projects in a sustainable way and provide important input for investigating the long-term impacts of PV expansion under different scenarios.