Abstract ID: 395

A comparison of simulated and observed CO₂ above Amazonia using regional and global transport models.

One of the most important ways in which we can assess the large-scale accuracy of models of terrestrial CO₂ flux is to test them against atmospheric CO₂ data. Although these flux models, like CASA, SiB, and IBIS can be tuned or checked against eddy-covariance CO₂ fluxes at particular sites, the spatial representativeness of these sites is unknown. In contrast, atmospheric data collected between the surface and 5 km altitude naturally integrates CO₂ fluxes over regional (10⁵ - 10⁶ km²) scales. In this study, we compare atmospheric CO₂ simulated using surface flux models in combination with atmospheric transported models. In particular, we will test the CASA-GFED and SiB3 biosphere models during 2004 using CO₂ observations made on light aircraft above Santarem. The level of agreement between observation and simulation will provide insight into the accuracy of the flux models. Furthermore, we will conduct these tests using both global and regional transport models. First, we will use the TM5 global transport model using global 6 deg. x 4 deg. resolution, but with a 1 x 1 deg. zoom over South America. Then we will use the BRAMS mesoscale model at 20 km resolution over the Amazonia, but forced with lateral boundary conditions from TM5. This extra axis of information will help us assess the importance of accurately simulated meteorology in reproducing atmospheric CO₂ variations.

Session:  Carbon - Scaling carbon fluxes to the region from measurements in plots, towers, and aircraft.

Presentation Type:  Poster