Spatiotemporal Pattern of Invasive Pedicularis in the Bayinbuluke Land, China, during 2019–2021: An Analysis Based on PlanetScope and Sentinel-2 Data

The accurate identification and monitoring of invasive plants are great significance to sustainable ecological development. Pedicularis poses a severe threat native biodiversity, security, socioeconomic development, human health in the Bayinbuluke Grassland, China. It is imperative useful obtain precise distribution map for controlling its spread. This study used positive unlabeled learning (PUL) method extract from Grassland based on multi-period Sentinel-2 PlanetScope remote sensing images. A change rate model single land cover type dynamic transfer matrix were constructed under GIS reflect spatiotemporal Pedicularis. results reveal that (1) PUL accurately identifies satellite images, achieving F1-scores above 0.70 up 0.94 across all three datasets: data (seven features), (thirteen features). (2) When comparing datasets, number features more important than spatial resolution terms use extraction. Nevertheless, when compared with data, demonstrated higher level accuracy predicting (3) During 2019–2021 growing season, area decreased, was mainly concentrated northeast southeast Swan Lake. acquired pattern could potentially be aid spread or elimination, methods proposed this adopted by government as low-cost strategy identify priority areas which concentrate efforts control continue invasion.