A remote sensing derived dataset for agricultural plastic greenhouses in China of 2019

This paper provides a remote sensing derived dataset for agricultural plastic greenhouses in China of 2019, which has spatial resolution of 30 meters. The dataset is based on the Google Earth Engine (GEE) cloud computing platform, where random forest classification model is utilized to classify the Sentinel-2 remote sensing images. Specifically, ground-truth samples are collected through field survey and visual interpretation, and then are randomly divided into training set and test set. Afterwards, feature extraction is performed such as spectral and texture features to construct a multi-dimensional feature space. Finally, the trained random forest model is utilized to classify national-scale remote sensing images through parallel computing to acquire the spatial distribution of China’s agricultural plastic greenhouses. The accuracy evaluation shows that the average classification accuracy is 87.45%, which indicates that the proposed dataset can accurately reflect the spatial distribution of agricultural plastic greenhouses across the whole country of China. In addition, in order to better visualize the national greenhouse distribution data, this paper also calculates the proportion of the greenhouse area in the 5km grid. Above all, this data set is the first publicly released thematic data on the spatial distribution of China’s agricultural plastic greenhouses across the country, which can provide data references for scientific researchers in related fields.