Mid-Continental Intensive Field Campaign Atmospheric CO2 Observations Compared to Forward Models

Two commonly used approaches to study source/sinks of CO2 are the “bottom-up” and the “top-down” methods. Because of the large discrepancies between these approaches, the North America Carbon Program devised the Mid-Continental Intensive field campaign (MCI). The MCI campaign aims at improving the carbon flux estimates of both approaches with a combination of atmospheric transport models, a denser network of in-situ atmospheric CO2 measurements and agricultural inventories. The first step in evaluating and improving inverse models is to compare observed CO2 concentrations and predicted concentrations from forwards models. This study shows a model-data comparison at multiple temporal and spatial scales for the 2007 growing season. In-situ tower-based observations are compared to two different forwards models: NOAA’s Carbon Tracker and CSU’s SiBcrop-RAMS. Observations from two tall towers of the National Oceanic and Atmospheric Administration (NOAA) and five towers of Ring2 PSU network are used for this comparison. The towers are located in an intensively agricultural region of the North American continent. Comparisons to date show that both models predict higher mid-summer concentrations at three sites located in the “corn belt.” Both models have difficulty reproducing the observed monthly-average spatial gradient across these sites. The models also underestimate the maximum observed spatial gradients in daytime, daily-averaged boundary layer concentrations. These results suggest that the rapid photosynthetic rates found in corn are not yet well-simulated in these models, and that these data, when used in inversions, will provide a valuable constraint on regional fluxes.