Abstract. Forests play a pivotal role in global carbon cycling and biodiversity conservation, yet they face increasing disturbances from both anthropogenic and natural drivers. This study presents the first high-resolution (30 m) global forest disturbance dataset (GFD) for 2000–2020, classifying 11 disturbance types by integrating Landsat-based Continuous Change Detection and Classification (CCDC) time-series analysis with spatial metrics and machine learning. A total of 57 000 expert-validated sample points were used to train and validate a decision tree model, achieving an overall accuracy of 99 %. The results reveal that forestry replanting (44 %), shifting cultivation (24 %), and forest fires (11 %) dominate global forest loss. There are regional differences in global forest disturbance, such as farmland expansion in South America and Africa, forest fires in northern regions, and shifting cultivation in tropical regions. Disturbed forests span 1247.06 Mha, accounting for 31 % of the global forest area. Notably, 3 % of global forests were newly established, primarily in China, India, and Brazil. The spatial consistency analysis (R2= 0.93) highlights a strong overall agreement between the GFD product and other datasets, while the GFD product offers superior spatial resolution. The GFD dataset advances our understanding of forest dynamics and underscores the need for targeted conservation strategies in an era of escalating environmental change. The 30 m resolution GFD generated by this study is openly available at https://doi.org/10.6084/m9.figshare.28465178 (Liu et al., 2025a).
Wang et al. (Mon,) studied this question.