A remote sensing classification dataset of Beidagang Wetland in Tianjin from 2019 to 2022

The remote sensing classification dataset of Beidagang Wetland in Tianjin from 2019 to 2022 is based on Google Earth Engine, using the random forest model to classify Sentinel-1 and Sentinel-2 images. This dataset processed and analyzed 313 Sentinel-1 images from May to October in each year from 2019 to 2022, and 275 Sentinel-2 images with cloud coverage less than 30%. In the random forest classification, the sample points are obtained from field sampling and UAV images, and the classification features comprised of NDWI, MNDWI, EVI, VH, and the phenological index of NDVI. There are 5 classes: permanent water, temporary water, perennial vegetation, annual vegetation, and barren. The dataset includes 4 TIF images of remote sensing classification in Beidagang Wetland from 2019 to 2022.