3F: Data assimilation research to improve model initialization, the ability to evaluate models, and reduce model biases

Assimilating MOPITT CO data to improve the chemical state of the atmosphere

We evaluated the regional and global impact of assimilating carbon monoxide (CO) observations from the Measurement of Pollution in The Troposphere (MOPITT) on modeled CO distributions and found significant improvements in the post-assimilation model as compared to independent CO observations (Gaubert et al., 2016). The MOPITT instrument is a nadir-viewing, cross-track scanning gas correlation radiometer on board the NASA EOS Terra satellite, which launched in December, 1999. The dataset used in this study is the product of a multispectral retrieval algorithm, using both the thermal (4.7 μm) and near (2.7 μm) infrared absorption bands. Multispectral retrievals provide increased sensitivity to near-surface CO concentrations with respect to thermal infrared-only retrievals, in particular for daytime measurements over land (Worden et al., 2010, Deeter et al., 2011).

The model forecast is performed with the Community Atmospheric Model with Chemistry (or CAM-Chem), the coupled chemistry-climate model of the Community Earth System Model. The Data Assimilation Research Testbed (DART) is an open source community software facility for ensemble data assimilation (Anderson et al., 2009). We use the Ensemble Adjustment Kalman Filter (EAKF) scheme for our analysis, done every 6 hours. A 30-member ensemble of coupled (atmosphere/land/chemistry) forecasts, with perturbed emissions, is used to provide the best estimate of CO and uncertainties. The ensemble of meteorological fields provides a spread in chemistry through perturbations in the initial condition of meteorological state variables.

We compared posterior CO fields from the control run (no MOPITT assimilation) and the reanalysis run (with MOPITT assimilation) to surface and aircraft in situ CO measurements and to CO total columns observed from ground stations with up-looking FTS (Fourier Transform Spectrometers). As an illustration, we show the evaluation with the in-situ aircraft measurements from the IAGOS database (In-Service Aircraft for a Global Observing System). The assimilation of MOPITT improves the model CO distribution across all seasons with an average reduction of 50% of the root mean square error (RMSE).

Seasonal average CO profiles.
Figure 1. (a) Seasonal average CO profiles from IAGOS measurements (black), assimilation run (red), and Control Run (blue) over the U.S./Canada region from March 2002 to March 2003. The left column shows the CO observations in black along with their standard deviation (gray). The dashed lines correspond to the 5th and 95th percentiles. The right column shows the RMSE (root mean square error) and correlation coefficients as a function of altitude. Figure from Gaubert et al., (2016). Click for larger image.

References

Anderson, J. L., Hoar, T., Raeder, K., Liu, H., Collins, N., Torn, R., and Avellino, A. (2009), The Data Assimilation Research Testbed: a community facility, B. Am. Meteorol. Soc., 90, 1283-1296.

Deeter, M. N., H. M. Worden, J. C. Gille, D. P. Edwards, D. Mao, and J. R. Drummond (2011), MOPITT multispectral CO retrievals: Origins and effects of geophysical radiance errors, Journal of Geophysical Research: Atmospheres, 116(D15), n/a–n/a, doi:10.1029/2011JD015703.

Gaubert, B., A. F. Arellano Jr., J. Barré, H. M. Worden, L. K. Emmons, S. Tilmes, R. R. Buchholz, F. Vitt, K. Raeder, N. Collins, J. L. Anderson, C. Wiedinmyer, S. Martinez Alonso, D. P. Edwards, M. O. Andreae, J. W. Hannigan, C. Petri, K. Strong, N. Jones (2016), Toward a chemical reanalysis in a coupled chemistry-climate model: An evaluation of MOPITT CO assimilation and its impact on tropospheric composition, J. Geophys. Res. Atmos., 121, 7310-7343, doi:10.1002/2016JD024863.

Worden, H. M., M. N. Deeter, D. P. Edwards, J. C. Gille, J. R. Drummond, and P. Nédélec (2010), Observations of near-surface carbon monoxide from space using MOPITT multispectral retrievals, Journal of Geophysical Research (Atmospheres), 115(d14), 18314, doi:10.1029/2010JD014242.