Does soil pyrogenic carbon determine plant functional traits in Amazon Basin forests?
Honorio Coronado, EN
Ter Steege, H
Reason for embargo
Amazon forests are fire-sensitive ecosystems and consequently fires affect forest structure and composition. For instance, the legacy of past fire regimes may persist through some species and traits that are found due to past fires. In this study, we tested for relationships between functional traits that are classically presented as the main components of plant ecological strategies and environmental filters related to climate and historical fires among permanent mature forest plots across the range of local and regional environmental gradients that occur in Amazonia. We used percentage surface soil pyrogenic carbon (PyC), a recalcitrant form of carbon that can persist for millennia in soils, as a novel indicator of historical fire in old-growth forests. Five out of the nine functional traits evaluated across all 378 species were correlated with some environmental variables. Although there is more PyC in Amazonian soils than previously reported, the percentage soil PyC indicated no detectable legacy effect of past fires on contemporary functional composition. More species with dry diaspores were found in drier and hotter environments. We also found higher wood density in trees from higher temperature sites. If Amazon forest past burnings were local and without distinguishable attributes of a widespread fire regime, then impacts on biodiversity would have been small and heterogeneous. Alternatively, sufficient time may have passed since the last fire to allow for species replacement. Regardless, as we failed to detect any impact of past fire on present forest functional composition, if our plots are representative then it suggests that mature Amazon forests lack a compositional legacy of past fire.
We gratefully acknowledge the financial support provided to KGM, BSM, BHMJ, EAO and TRF by the Coordination of Improvement of Personnel in Higher Education, Brazil (CAPES) through a Science without Borders grant to TRF (PVE 177/2012). Sample analysis was supported by a grant from the University of Exeter to TRF. The National Council of Science and Technology, Brazil (CNPq) is acknowledged for a productivity grant (bolsa produtividade PQ) to BHMJ and BSM, for a Postdoctoral fellowship to DSN, and the financial support to the projects PELD (403725/2012-7) and PPBio (457602/2012-0). We acknowledge the financial support provided to DSN, OLP, and BSM by CNPq through a Science without Borders grant to OLP (PVE 401279/2014-6). OLP is also supported by an ERC Advanced Grant (T-FORCES) and is a Royal Society-Wolfson Research Merit Award Holder. Tree data used in this analysis are available online at the Tropical Ecology Assessment and Monitoring (TEAM) Network of Conservation International, funded by the Gordon and Betty Moore Foundation, and ForestPlots.net database. The RAINFOR forest monitoring network has been supported principally by the Natural Environment Research Council (Grants NE/B503384/1, NE/D01025X/1, NE/I02982X/1, NE/F005806/1, NE/D005590/1, and NE/I028122/1), the Gordon and Betty Moore Foundation, and by the EU Seventh Framework Programme (GEOCARBON-283080 and AMAZALERT-282664).
This is the author accepted manuscript. The final version is available from the publisher via the DOI in this record.
Vol 218, Iss 9, pp 1047–1062