What controls variation in carbon use efficiency among Amazonian tropical forests?

Why do some forests produce biomass more efficiently than others? Variations in Carbon Use Efficiency (CUE: total Net Primary Production (NPP)/ Gross Primary Production (GPP)) may be due to changes in wood residence time (Biomass/NPPwood), temperature, or soil nutrient status. We tested these hypotheses in 14, one ha plots across Amazonian and Andean forests where we measured most key components of net primary production (NPP: wood, fine roots, and leaves) and autotrophic respiration (Ra; wood, rhizosphere, and leaf respiration). We found that lower fertility sites were less efficient at producing biomass and had higher rhizosphere respiration, indicating increased carbon allocation to belowground components. We then compared wood respiration to wood growth and rhizosphere respiration to fine root growth and found that forests with residence times 40 yrs. A comparison of rhizosphere respiration to fine root growth showed that rhizosphere growth respiration was significantly greater at low fertility sites. Overall, we found that Amazonian forests produce biomass less efficiently in stands with residence times >40 yrs and in stands with lower fertility, but changes to long-term mean annual temperatures do not impact CUE.This work is a product of the Global Ecosystems Monitoring (GEM) network (gem.tropicalforests.ox.ac.uk) the Andes Biodiversity and Ecosystems Research Group ABERG (andesresearch.org) and the Amazon Forest Inventory Network RAINFOR (www.rainfor.org) research consortia, and was funded by grants from the UK Natural Environment Research Council (Grants NE/D01025X/1, NE/D014174/1), grants to YM and OP from the Gordon and Betty Moore Foundation, and a grant from the EU FP7 GEOCARBON (283080) project. We thank the Servicio Nacional de Areas Naturales Protegidas por el Estado (SERNANP) and personnel of Manu National Park who provided logistical assistance and permission to work in the pro- tected areas in Peru, the Explorers’ Inn at Tambopata, ACCA for use of the Wayqecha Research Station, and IIAP for use of the Allpahuayo Research Station, the Museo Goeldi for access to the Caxiuan~a Research Station, and IPAM for the access to the Tan- guro plots. We also gratefully acknowledge LBA support CNPQ grant 411 457914/2013-0/MCTI/CNPq/FNDCT/LBA/ESE- CAFLOR to ACLD, and NERC and ARC support to PM (NE/ J011002/1, DP170104091). YM is supported by an ERC Advanced Investigator Award GEM-TRAIT (321131) and by the Jackson Foundation. CED is supported by the John Fell Fund and Google