Key points are not available for this paper at this time.
Tropical rainforests play an important role in the global carbon cycle. They store massive amounts of biomass in their trees and soils, and contribute to climate mitigation by removing carbon from the atmosphere through photosynthesis. It is assumed that plant responses to rising atmospheric CO2 concentrations may have induced an increase in biomass and thus, increased the carbon sink in forests worldwide. Rising CO2 directly stimulates photosynthesis (the so-called CO2-fertilization effect) and tends to reduce stomatal conductance, leading to enhanced water-use efficiency, which may provide an important buffering effect for plants during adverse climate conditions and also have implications for water resources by reducing the loss of soil moisture through transpiration. For these reasons, current global climate simulations consistently predict that undisturbed tropical forests will continue to sequester more carbon in aboveground biomass. However, several lines of evidence point towards a decreasing carbon sink strength of the Amazon rainforest in the coming decades, potentially driven by nutrient limitation, droughts or other factors. Mechanistically modelling the effects of rising CO2 in the Amazon rainforest are hindered by a lack of direct observations from ecosystem scale CO2 experiments. To address these critical issues, we are currently building a free-air CO2 enrichment (FACE) experiment in an old-growth, highly diverse, tropical forest in the Brazilian Amazon and we here present our main hypotheses that underpin the AmazonFACE experiment. We focus on possible effects of rising CO2 on carbon uptake and allocation, phosphorus cycling, water-use and plant-herbivore interactions, and discuss relevant ecophysiological processes, which need to be implemented in dynamic vegetation models to estimate future changes of the Amazon carbon sink. We give an update on the state of the experiment construction, present the sampling strategy and discuss our approach to upscale tree-level responses to stand scale.
Rammig et al. (Fri,) studied this question.