How much will plants increase their photosynthetic uptake as atmospheric carbon dioxide (CO2) rises? This is one of the most pressing questions we face, with wide-ranging implications for the rate of climate change, allowable CO2 emissions and the water cycle.

In theory, we should be able to answer this question by now as we have carried out decades of CO2 manipulation experiments across the world, but it turns out we’re still limited by our experimental data. The first generation of free-air CO2 enrichment (FACE) experiments was limited geographically, perhaps most notable by an absence of an experiment in the tropics and specifically the Amazon rainforest.

The Amazon FACE project: Exploring the impact of climate change on the rainforest

How will the Amazon forest respond to elevated CO2? An earlier model-based study suggested that we should expect quite a large positive response due to an interaction between the warmer temperatures and the biochemistry of photosynthesis.

However, this prediction did not consider any interactions with nutrients and, as a result, has long been questioned. It is important to note that this prediction is embedded in the vast majority (if not all) of CMIP5 models, as only one land scheme considered a nutrient feedback (and then, only with respect to nitrogen and not phosphorus).

In this paper, we set about tackling this question in advance of the much hoped for AmazonFACE experiment. We used a new generation of terrestrial biosphere models that consider more sophisticated nutrient feedbacks than captured in the CMIP5 ensemble.

Our study showed that phosphorus availability reduced the projected CO2-induced biomass carbon growth by about 50% over 15 years compared to estimates from carbon and carbon-nitrogen models. Moreover, we showed that the variation in the biomass carbon response among the phosphorus-enabled models was considerable, ranging from 5 to 140 g C m−2 yr−1, owing to the contrasting plant phosphorus use and acquisition hypotheses embedded in models.

This now represents a key axis requiring further theory and model development. To reduce this considerable uncertainty in model projections for the Amazon ultimately requires the AmazonFACE experiment to take place. Unfortunately, securing funding for this much-needed experiment has proved fruitless so far.

  • Paper: Fleischer, Katrin, Anja Rammig, Martin G. De Kauwe, Anthony P. Walker, Tomas F. Domingues, Lucia Fuchslueger, Sabrina Garcia, et al. “Amazon Forest Response to CO2 Fertilization Dependent on Plant Phosphorus Acquisition.” Nature Geoscience, August 5, 2019.