Fig. 1.
CO2 fluxes from June to December in TgC.degree−2 over the MCI from the SiBcrop vegetation model (a), our reference case TR0, i.e. the inverse system using the entire network of observation sites (b), using only the sites outside of the Corn Belt area (c), using the sites only within the Corn Belt area (d), using a sparser network (e) and using a minimal configuration of two sites (one in the Corn Belt and one out) (f).
Table 1. Regional CO2 flux balance from June to December 2007 in TgC over the MCI (first line), and averaged fluxes over the corn-dominated area (which corresponds to 23% of the domain) and out of it in gC m−2, for Sibcrop (prior fluxes), using the complete observation network (posterior or TR0), using towers around the corn belt (NON-CORN), using towers within the corn belt area (CORN), using a sparser network of towers (SPARSE), and using only two towers (MIN)
Fig. 2.
CO2 flux correction from June to December in TgC.degree−2 over the MCI using the SiBcrop prior fluxes, with our reference case, i.e. the inverse system using the entire network of observation sites (a), with the entire network but the flux error correlation is built on an exponentially decreasing model only (b), using only the sites out of the Corn Belt area (c), using the sites only within the Corn Belt area (d), using a sparser network (e) and using a minimal configuration of two sites (one in the Corn Belt and one out) (f).
Fig. 3.
Error reduction in % using all the towers and prior flux errors with ecosystem-based standard deviations (RMS) and spatial correlations based on ecosystems and distances between pixels (case TR0) (a) and the second case considering correlations with distance only (exponentially decaying model) (TRD) (b).
Table 2. Mixing ratio residuals in ppm for each tower, averaged over the 7 months (mean) and their related RMSE at the hourly time scale before and after inversion. The three lines correspond to the initial mismatch between modelled and observed mixing ratios (a priori), posterior residuals after inversion using the eight towers (a posteriori) and residuals from each inversion excluding the tower indicated in the first line, used for validation (LOOCV)
[i] Residuals in LOOCV correspond to cross-validation of the inverse fluxes retrieved from the Leave-One-Out experiments. Values closer to zero compared with the a priori mismatch indicate an improvement.