Bastin et al .’s estimate (Reports, 5 July 2019, p. 76) that tree planting for climate change mitigation could sequester 205 gigatonnes of carbon is approximately five times too large. Their analysis inflated soil organic carbon gains, failed to safeguard against warming from trees at high latitudes and elevations, and considered afforestation of savannas, grasslands, and shrublands to be restoration.

Fires have played an important role in creating and maintaining savannas over the centuries and are also one of the main natural disturbances in forests. The functional role of fires in savannas and forests can be investigated through examining sedimentary charcoal in order to reconstruct long-term fire history. However, the relationship between charcoal and vegetation structure in tropical grassy ecosystems remains to be elucidated. Here, we compared recent charcoal records from lake sediments in three tropical ecosystems (forest, savanna, and forest–savanna mosaic) with land cover inferred from remote-sensing images. Charcoal width-to-length (W/L) ratio is a good proxy for changes in fuel type. At one of the lakes, a significant W/L modification from values >0.5 (mainly wood) to <0.5 (~grass) was recorded simultaneously with changes in land cover. Indeed, a significant deforestation was recorded around this lake in the remote-sensing imagery between 1984 and 1994. The results also indicate that a riparian forest around a lake could act as a physical filter for charcoal accumulation; we used the mean charcoal size as a proxy to evaluate this process. Charcoal Accumulation Rates (CHAR), a burned biomass proxy, were combined with W/L ratio and the mean charcoal size to investigate the land-use history of the landscapes surrounding the study sites. This combined approach allowed us to distinguish between episodic slash-and-burn practices in the forest and managed fields or pastures burning frequently.

Significance: We develop a biogeographic approach to analyzing the presence of alternative stable states in tropical biomes. Whilst forest–savanna bistability has been widely hypothesized and modeled, empirical evidence has remained scarce and controversial, and here, applying our method to Africa, we provide large-scale evidence that there are alternative states in tree species composition of tropical vegetation. Furthermore, our results have produced more accurate maps of the forest and savanna distributions in Africa, which take into account differences in tree species composition, and a complex suite of determinants. This result is not only important for understanding the biogeography of the continent but also, to guide large-scaled tree planting and restoration efforts planned for the region.

Although lacustrine sedimentary charcoal has long been used to infer paleofires, their quantitative reconstructions require improvements of the calibration of their links with fire regimes (i.e. occurrence, area, and severity) and the taphonomic processes that affect charcoal particles between the production and the deposition in lake sediments. Charcoal particles >150 µm were monitored yearly from 2011 to 2016 using traps submerged in seven head lakes situated in flat-to-rolling boreal forest landscapes in eastern Canada. The burned area was measured, and the above-ground fire severity was assessed using the differentiated normalized burn ratio (dNBR) index, derived from LANDSAT images, and measurements taken within zones radiating 3, 15, and 30 km from the lakes. In order to evaluate potential lag effects in the charcoal record, fire metrics were assessed for the year of recorded charcoal recording (lag 0) and up to 5 years before charcoal deposition (lag 5). A total of 92 variables were generated and sorted using a Random Forest-based methodology. The most explanatory variables for annual charcoal particle presence, expressed as the median surface area, were selected. Results show that, temporally, sedimentary charcoal accurately recorded fire events without a temporal lag; spatially, fires were recorded up to 30 km from the lakes. Selected variables highlighted the importance of burned area and fire severity in explaining lacustrine charcoal. The charcoal influx was thus driven by fire area and severity during the production process. The dispersion process of particles resulted mostly of wind transportation within the regional (<30 km) source area. Overall, charcoal median surface area represents a reliable proxy for reconstructing past burned areas and fire severities.