Model-Data Comparison of Mid-Continental Intensive Field Campaign Atmospheric CO2 Mixing Ratios

Liza Ivelisse Diaz, Penn State University, lzd120@psu.edu (Presenter)
Kenneth James Davis, Penn State, kjd10@psu.edu
Thomas Lauvaux, Pennsylvania State University, tul5@meteo.psu.edu
Natasha Miles, Penn State University, nmiles@met.psu.edu
Scott Richardson, Penn State University, srichardson@psu.edu
Andy Jacobson, NOAA, andy.jacobson@noaa.gov
Arlyn Andrews, NOAA ESRL, arlyn.andrews@noaa.gov

Atmospheric inversions are used to assess biosphere-atmosphere CO2 fluxes. However, large uncertainty and variability exists among inverse flux estimates, one of the main contributors being transport model errors. Because of these large discrepancies among flux estimates, the North American Carbon Program devised the Mid-Continental Intensive field campaign (MCI), which aims at comparing carbon flux estimates from atmospheric inversions using a dense network of in-situ atmospheric CO2 measurements with agricultural inventories. Our study shows a model-data comparison using CO2 mixing ratio observations from the mesonet of surface atmospheric instruments deployed during the MCI and modeled mixing ratios from two atmospheric transport models: the global model TM5 at 1 degree resolution from NOAA’s Carbon Tracker system and the mesoscale model WRF-CASA at 10km resolution, both models coupled to identical CO2 fluxes. Temporal and spatial statistical analyses were performed for two periods of 2007 using daily daytime average mixing ratios, to test the ability of these models to reproduce observed mixing ratios, and infer the possible causes for the discrepancies. Carbon Tracker tends to overestimate and WRF-CASA underestimate midsummer mixing ratios for sites located in the “corn belt” by 10-15ppm. In time, the models tend to be highly correlated with the observed seasonality (≥0.7), but less correlated to growing season (≤0.7), where weather-related changes in CO2 dominate observed variability. In space, Carbon tracker’s residuals are the most correlated between sites that are closer and surrounded by similar vegetation. However, WRF-CASA residuals show high correlations related to the spatial patterns of CO2 surface fluxes. Our study suggests that WRF-CASA CO2 residuals are strongly influenced by surface fluxes, whereas, Carbon Tracker CO2 residuals are influenced by an unrealistic vertical mixing and the coarse resolution of the model. Ongoing analyses are evaluating the impact of different parameterizations of WRF on the same regional CO2 data set.

Presentation Type:  Poster

Session:  Other   (Tue 11:30 AM)

Associated Project(s): 

• Davis, Ken: Regional Atmospheric Inversions to Determine Land-Atmosphere Carbon Fluxes in the Southeastern Forests of the United States ...details

Poster Location ID: 153