Mapping photovoltaic power plants in China using Landsat, random forest, and Google Earth Engine

Abstract. Photovoltaic (PV) technology, an efficient solution for mitigating the impacts of climate change, has been increasingly used across world to replace fossil fuel power minimize greenhouse gas emissions. With world's highest cumulative and fastest built PV capacity, China needs assess environmental social these established plants. However, a comprehensive map regarding plants' locations extent remains scarce on country scale. This study developed workflow, combining machine learning visual interpretation methods with big satellite data, plants China. We applied pixel-based random forest (RF) model classify from composite images in 2020 30 m spatial resolution Google Earth Engine (GEE). The resulting classification was further improved by approach. Eventually, we 2020, covering total area 2917 km2. found that most were situated cropland, followed barren land grassland, based derived national map. In addition, installation generally decreased vegetation cover. new dataset is expected be conducive policy management, assessment, photovoltaic plant distribution available public at https://doi.org/10.5281/zenodo.6849477 (Zhang et al., 2022).