Rangelands are extensive ecosystems, providing important ecosystem services while undergoing continuous change. As a result, there is a need for improved monitoring technologies that better characterize vegetation changes over space and time. Satellite remote sensing has proven effective in this regard, tracking vegetation dynamics at both broad and fine scales. Advancements in technology provide the opportunity to improve monitoring efforts and to better capture subtle yet ecologically significant changes. We leveraged the enhanced spatial, spectral, and temporal resolution of Sentinel-2 satellites to estimate fractional cover and the size distribution of plant inter-canopy gaps across rangelands of the western United States. We developed a one-dimensional convolutional neural network, trained on extensive field data, to predict cover of plant functional types and select genera, and canopy gap size classes. We produced annual, 10 m resolution estimates from 2018 to 2024, providing an unprecedented resource for monitoring rangeland condition and assessing the effectiveness of management strategies. ### Competing Interest Statement The authors have declared no competing interest.
原文链接
分享
引用
