Plastic-covered greenhouse (PCG) farming has been practiced worldwide, particularly in highland areas, for planting fruits and garden plants. Highland areas in south-central Vietnam have been witnessing an expansion of PCG farming in recent decades. For sustainable management of plastic-covered greenhouse farming areas, a proper and continuous monitoring of their spatial patterns is necessary. In this study, we used spaceborne multispectral imagery (Landsat-8 OLI and Sentinel-2 MSI) for automatic detection and mapping of plastic covered greenhouse farming areas in and surrounding areas of Dalat City in Lam Dong province, south-central Vietnam. Unsupervised (K-Means algorithm) and automatic delineation using spectral indices were applied to the Sentinel-2 and Landsat-8 data for mapping plastic-covered greenhouses in Dalat City and surrounding areas. The results were evaluated using field data; automatic spectral index-based mapping of PCGs using Sentinel-2 and Landsat-8 data was found as robust in terms of the overall accuracy. The overall accuracies of PCG maps derived from Sentinel-2 and Landsat-8 data by applying spectral indices were 90.8% and 88.7%, respectively, and those of unsupervised K-Means classification were estimated as 73.2% and 71.7%, respectively. This study estimated that about 2425 ha of plastic-covered greenhouses existed within in Dalat City in 2020.
Keywords: Automatic greenhouse mapping; High-tech agriculture; Plastic greenhouse farming; Plastic greenhouse index; Plastic remote sensing.
© 2022. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.