Using free air CO 2 enrichment data to constrain land surface model projections of the terrestrial carbon cycle

Abstract. Predicting the responses of terrestrial ecosystem carbon to future global change strongly relies on our ability to model accurately the underlying processes at a global scale. However, terrestrial biosphere models representing the carbon and nitrogen cycles and their interactions remain subject to large uncertainties, partly because of unknown or poorly constrained parameters. Data assimilation is a powerful tool that can be used to optimise these parameters by confronting the model with observations. In this paper, we identify sensitive model parameters from a recent version of the ORCHIDEE land surface model that includes the nitrogen cycle. These sensitive parameters include ones involved in parameterisations controlling the impact of the nitrogen cycle on the carbon cycle and, in particular, the limitation of photosynthesis due to leaf nitrogen availability. We optimise these ORCHIDEE parameters against carbon flux data collected on sites from the Fluxnet network. However, optimising against present-day observations does not automatically give us confidence in the future projections of the model, given that environmental conditions are likely to shift compared to present-day. Manipulation experiments give us a unique look into how the ecosystem may respond to future environmental changes. One such experiment, the Free Air CO2 Enrichment experiment, provides a unique opportunity to assess vegetation response to increasing CO2 by providing data at ambient and elevated CO2 conditions. Therefore, to better capture the ecosystem response to increased CO2, we add the data from two FACE sites to our optimisations, in addition to the Fluxnet data. We use data from both CO2 conditions of the Free Air CO2 Enrichment experiment, which allows us to gain extra confidence in the model simulations using this set of parameters. We find that we are able to improve the magnitude of modelled productivity. Although we are unable to correct the interannual variability fully, we start to simulate possible progressive nitrogen limitation at one of the sites. Using an idealised simulation experiment based on increasing atmospheric CO2 by 1 % per year over 100 years, we find that the rate of CO2 fertilisation is much lower when Free Air CO2 Enrichment data has been in the optimisation.