Near real-time deforestation monitoring in tropical ecosystems using satellite image time series

摘要

Monitoring forest disturbances is critical for addressing its impact on carbon storage and fluxes, biodiversity, and other socio-ecological processes [1, 2]. Satellite remote sensing enables cost-effective and accurate monitoring at frequent time steps over large areas. Several methods are available to detect disturbances within satellite image time series [3, 4]. However, methods to detect changes within newly acquired satellite images are lacking. There is a critical need for methods that enable rapid and automated analysis of satellite image time series to detect forest disturbances in near-real time, especially in regions of less developed countries (e.g. Asia, Africa) [2].