We used the Visible Infrared Imaging Radiometer Suite (VIIRS) active fire data (375 m spatial resolution) to automatically extract multispectral samples and train a One-Class Support Vector Machine for burned area mapping, and applied the resulting classification algorithm to 300-m spatial resolution imagery from the Project for On-Board Autonomy-Vegetation (PROBA-V). The active fire data were screened to prevent extraction of unrepresentative burned area samples and combined with surface reflectance bi-weekly composites to produce burned area maps. The procedure was applied over the Brazilian Cerrado savanna, validated with reference maps obtained from Landsat images and compared with the Collection 6 Moderate Resolution Imaging Spectrometer (MODIS) Burned Area product (MCD64A1) Results show that the algorithm developed improved the detection of small-sized scars and displayed results more similar to the reference data than MCD64A1. Unlike active fire-based region growing algorithms, the proposed approach allows for the detection and mapping of burn scars without active fires, thus eliminating a potential source of omission error. The burned area mapping approach presented here should facilitate the development of operational-automated burned area algorithms, and is very straightforward for implementation with other sensors.

Climate change has substantially increased the frequency of extreme droughts in the Amazon basin, generating concern about impacts on the world's largest tropical forest, which contributes about one‐seventh of the global vegetation carbon sink. Most research to understand drought impacts has focused on the immediate influences of such events, neglecting post‐drought effects on ecosystems recovery. Since ecological processes are influenced by antecedent conditions, we analyzed whether extreme droughts affect vegetation growth (i.e., net primary productivity, NPP) recovery. Here, we evaluated the NPP in the Amazon basin from 2003 to 2020, a period in which drought frequency was almost double the decadal incidence of the last century. We show that NPP does respond to the coupled impacts of individual droughts and the post‐drought impacts during ecosystem recovery. In particular, our results reveal that the ecosystems undergoing recovery show NPP about 13% lower than reference values based on the pre‐drought state or in areas undisturbed by drought. NPP deficits have consistently increased with the extreme droughts of 2005, 2010, and 2015 due to the combined effects of disturbances magnitude and the length of recovery. If the expected increase in drought frequency and intensity does occur, reduced recovery may lead the Amazon Forest to an alternative ecosystem state with lower carbon uptake, contributing to a warming global climate.

A significant proportion of carbon (C) captured by terrestrial primary production is buried in lacustrine ecosystems, which have been substantially affected by anthropogenic activities globally. However, there is a scarcity of sedimentary organic carbon (OC) accumulation information for lakes surrounded by highly productive rainforests at warm tropical latitudes, or in response to land cover and climate change. Here, we combine new data from intensive campaigns spanning 13 lakes across remote Amazonian regions with a broad literature compilation, to produce the first spatially-weighted global analysis of recent OC burial in lakes (over ~50-100-years) that integrates both biome type and forest cover. We find that humid tropical forest lake sediments are a disproportionately important global OC sink of 7.4 Tg C yr⁻¹ with implications for climate change. Further, we demonstrate that temperature and forest conservation are key factors in maintaining massive organic carbon pools in tropical lacustrine sediments.