Basin-wide variation in tree hydraulic safety margins predicts the carbon balance of Amazon forests

Funding: Data collection was largely funded by the UK Natural Environment Research Council (NERC) project TREMOR (NE/N004655/1) to D.G., E.G. and O.P., with further funds from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—Brasil (CAPES, finance code 001) to J.V.T. and a University of Leeds Climate Research Bursary Fund to J.V.T. D.G., E.G. and O.P. acknowledge further support from a NERC-funded consortium award (ARBOLES, NE/S011811/1). This paper is an outcome of J.V.T.’s doctoral thesis, which was sponsored by CAPES (GDE 99999.001293/2015-00). J.V.T. was previously supported by the NERC-funded ARBOLES project (NE/S011811/1) and is supported at present by the Swedish Research Council Vetenskapsrådet (grant no. 2019-03758 to R.M.). E.G., O.P. and D.G. acknowledge support from NERC-funded BIORED grant (NE/N012542/1). O.P. acknowledges support from an ERC Advanced Grant and a Royal Society Wolfson Research Merit Award. R.S.O. was supported by a CNPq productivity scholarship, the São Paulo Research Foundation (FAPESP-Microsoft 11/52072-0) and the US Department of Energy, project GoAmazon (FAPESP 2013/50531-2). M.M. acknowledges support from MINECO FUN2FUN (CGL2013-46808-R) and DRESS (CGL2017-89149-C2-1-R). C.S.-M., F.B.V. and P.R.L.B. were financed by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—Brasil (CAPES, finance code 001). C.S.-M. received a scholarship from the Brazilian National Council for Scientific and Technological Development (CNPq 140353/2017-8) and CAPES (science without borders 88881.135316/2016-01). Y.M. acknowledges the Gordon and Betty Moore Foundation and ERC Advanced Investigator Grant (GEM-TRAITS, 321131) for supporting the Global Ecosystems Monitoring (GEM) network (gem.tropicalforests.ox.ac.uk), within which some of the field sites (KEN, TAM and ALP) are nested. The authors thank Brazil–USA Collaborative Research GoAmazon DOE-FAPESP-FAPEAM (FAPESP 2013/50533-5 to L.A.) and National Science Foundation (award DEB-1753973 to L. Alves). They thank Serrapilheira Serra-1709-18983 (to M.H.) and CNPq-PELD/POPA-441443/2016-8 (to L.G.) (P.I. Albertina Lima). They thank all the colleagues and grants mentioned elsewhere [8,36] that established, identified and measured the Amazon forest plots in the RAINFOR network analysed here. The authors particularly thank J. Lyod, S. Almeida, F. Brown, B. Vicenti, N. Silva and L. Alves. This work is an outcome approved Research Project no. 19 from ForestPlots.net, a collaborative initiative developed at the University of Leeds that unites researchers and the monitoring of their permanent plots from the world’s tropical forests [61]. The authros thank A. Levesley, K. Melgaço Ladvocat and G. Pickavance for ForestPlots.net management. They thank Y. Wang and J. Baker, respectively, for their help with the map and with the climatic data. The authors acknowledge the invaluable help of M. Brum for kindly providing the comparison of vulnerability curves based on PAD and on PLC shown in this manuscript. They thank J. Martinez-Vilalta for his comments on an early version of this manuscript. The authors also thank V. Hilares and the Asociación para la Investigación y Desarrollo Integral (AIDER, Puerto Maldonado, Peru); V. Saldaña and Instituto de Investigaciones de la Amazonía Peruana (IIAP) for local field campaign support in Peru; E. Chavez and Noel Kempff Natural History Museum for local field campaign support in Bolivia; ICMBio, INPA/NAPPA/LBA COOMFLONA (Cooperativa mista da Flona Tapajós) and T. I. Bragança-Marituba for the research support. Tropical forests face increasing climate risk1,2, yet our ability to predict their response to climate change is limited by poor understanding of their resistance to water stress. Although xylem embolism resistance thresholds (for example, Ψ50) and hydraulic safety margins (for example, HSM50) are important predictors of drought-induced mortality risk3-5, little is known about how these vary across Earth's largest tropical forest. Here, we present a pan-Amazon, fully standardized hydraulic traits dataset and use it to assess regional variation in drought sensitivity and hydraulic trait ability to predict species distributions and long-term forest biomass accumulation. Parameters Ψ50 and HSM50 vary markedly across the Amazon and are related to average long-term rainfall characteristics. Both Ψ50 and HSM50 influence the biogeographical distribution of Amazon tree species. However, HSM50 was the only significant predictor of observed decadal-scale changes in forest biomass. Old-growth forests with wide HSM50 are gaining more biomass than are low HSM50 forests. We propose that this may be associated with a growth-mortality trade-off whereby trees in forests consisting of fast-growing species take greater hydraulic risks and face greater mortality risk. Moreover, in regions of more pronounced climatic change, we find evidence that forests are losing biomass, suggesting that species in these regions may be operating beyond their hydraulic limits. Continued climate change is likely to further reduce HSM50 in the Amazon6,7, with strong implications for the Amazon carbon sink. Publisher PDF