Quantifying the model structural error in carbon cycle data assimilation systems

Abstract

Abstract. This study explores the impact of the structural error of biosphere models when assimilating net ecosystem exchange (NEE) measurements or CO 2 concentration measurements to optimise uncertain model parameters within carbon cycle data assimilation systems (CCDASs). This error has been proven difficult to identify and is often neglected in the total uncertainty budget. We propose a simple method which is derived from the model-minus-observation mismatch statistics. This diagnosis is applied to a state-ofthe-art biogeochemical model using measurements of the net surface CO 2 flux at twelve sites located in temperate, deciduous, broadleaf forests. We find that the structural model error in the NEE space has a standard deviation of 1.5 to 1.7 gC m −2 d −1 , without a significant correlation structure beyond the lag of a few days, and a large spatial structure that can be approximated with an exponential decay of efolding length of 500 km. In the space of concentrations, its characteristics are commensurate with the transport errors, both for surface air sample measurements and total column measurements. The inferred characteristics are confirmed by complementary optimality diagnostics performed after sitescale parameter optimisations.

Constraining a land-surface model with multiple observations by application of the MPI-Carbon Cycle Data Assimilation System V1.0

“…Conceptually similar systems have been built for other, more complex, global biosphere models. Applications of these alternative systems include, for example, constraining the phenology of the JULES model with the MODIS collection five-leaf area index product (Luke, 2011) and carbon fluxes in the ORCHIDEE model using observations from several FLUXNET sites (Kuppel et al, 2012(Kuppel et al, , 2013. Previous studies with these systems focussed on the effect of different (in situ and satellite) FAPAR observations at selected sites on simulated phenology with the ORCHIDEE model (e.g.…”

Technical Note: Comparison of ensemble Kalman filter and variational approaches for CO<sub>2</sub> data assimilation

“…For the CO 2 inverse problem, however, the dynamics are embedded within the atmospheric transport and are not used explicitly to inform the temporal evolution of the state vector. A few recent studies have attempted to use an explicit dynamical flux model (e.g., Kuppel et al, 2013), but they note that model errors significantly reduce the performance of the inversion in terms of the quality of the estimated fluxes. Currently, in most carbon flux estimation studies, dynamical models are not used, resulting in a loss of potentially valuable information to the DA system.…”

Diagnostic methods for atmospheric inversions of long-lived greenhouse gases

“…The attribution to these two components of error is based on leveraging differences in their spacetime structure, however, and is made easier when the two sources of error have features that are statistically distinct (e.g., Desroziers et al, 2005). Alternatively, some of the statistics may be well known from some other information source and can then play the role of a fixed point to deduce the other ones (e.g., Kuppel et al, 2013). It is important to remember, however, that diagnostics cannot bring original information to the problem, but rather provide a framework for interpreting available information.…”