Local Climate Zone Mapping Using Multi-Source Free Available Datasets on Google Earth Engine Platform

As one of the widely concerned urban climate issues, heat island (UHI) has been studied using local zone (LCZ) classification scheme in recent years. More and more effort focused on improving LCZ mapping accuracy. It become a prevalent trend to take advantage multi-source images mapping. To this end, paper tried utilize freely available datasets: Sentinel-2 multispectral instrument (MSI), Sentinel-1 synthetic aperture radar (SAR), Luojia1-01 nighttime light (NTL), Open Street Map (OSM) datasets produce 10 m result Google Earth Engine (GEE) platform. Additionally, derived MSI data were also exploited classification, such as spectral indexes (SI) gray-level co-occurrence matrix (GLCM) datasets. The different dataset combinations designed evaluate particular dataset’s contribution classification. was found that: (1) synergistic use SAR can improve accuracy classification; (2) multi-seasonal information Sentinel good (3) OSM, GLCM, SI, NTL have some positive when individually adding them seasonal datasets; (4) is not an absolute right way by combining many possible. With help GEE, study provides potential generate accurate large scale, which significant for development.