Forest loss in the tropics affects large areas, but whereas full forest conversions are routinely assessed, forest degradation patters remain often unclear. This is particularly so for the world’s tropical dry forests, where remote sensing of forest disturbances is challenging due to high canopy complexity, strong phenology and climate variability, and diverse degradation drivers. Here, we used the full depth of the Landsat archive and devised an approach to detect disturbances related to forest degradation across the entire Argentine Dry Chaco (about 489,000 km²) over a 30-year timespan. We used annual time series of different spectral indices, summarized for three seasonal windows, and applied LandTrendr to temporally segment each time series. The resulting pixel-level forest disturbance metrics then served as input for a Random Forests classification which we used to produce an area-wide disturbance map, and associated yearly area estimates of disturbed forest. Finally, we evaluated disturbance trends in relation to climate and soil conditions. Our best model produced a disturbance map with an overall accuracy of 79%, with a balanced error distribution. A total of 8% (24,877 ± 860 km²) of the remaining forest in the Argentine Dry Chaco have been affected by forest disturbances between 1990 and 2017. Diverse spatial patterns of forest disturbances indicate a variety of agents driving disturbances. We also found the disturbed area to vary strongly between years, with larger areas being disturbed during drought years. Our approach shows that it is possible to robustly map forest disturbances in tropical dry forests using Landsat time series, and demonstrates the value of ensemble approaches to capture spectrally-complex and heterogeneous land-change processes. For the Chaco, a global deforestation hotspot, our analyses provide the first Landsat-based assessment of forest disturbance in remaining forests, highlighting the need to better consider such disturbances in assessments of carbon budgets and biodiversity change.

热带地区的森林流失影响大片地区，但是尽管常规评估了全部森林转化，但森林退化的模式仍然常常不清楚。对于世界热带干旱森林而言尤其如此，由于高冠层的复杂性，强烈的物候学和气候多变性以及多样化的退化驱动因素，对森林干扰的遥感具有挑战性。在这里，我们使用了Landsat档案库的全部深度，并设计了一种方法来检测与整个阿根廷Dry Chaco（约489,000 km ²）超过30年的时间跨度。我们使用了不同光谱指数的年度时间序列，总结了三个季节窗口，并应用LandTrendr对每个时间序列进行时间分段。然后将得到的像素级森林干扰度量用作随机森林分类的​​输入，我们使用该分类来生成区域范围的干扰图以及相关的受干扰森林的年度面积估计。最后，我们评估了与气候和土壤条件有关的扰动趋势。我们的最佳模型生成的干扰图总体精度为79％，误差分布均衡。总计8％（24,877±860 km ²）在1990年至2017年之间，阿根廷干旱地区查科的剩余森林已受到森林扰动的影响。森林扰动的不同空间格局表明，各种因素导致了扰动。我们还发现，受干扰地区在几年之间变化很大，在干旱年份受干扰的地区更大。我们的方法表明，可以使用Landsat时间序列可靠地绘制热带干旱森林中的森林扰动图，并证明集成方法在捕获光谱复杂和非均质的土地变化过程中的价值。对于Chaco这个全球性的森林砍伐热点，我们的分析提供了基于Landsat的对剩余森林中森林干扰的首次评估，强调需要在评估碳预算和生物多样性变化时更好地考虑此类干扰。