We use the Weather Research and Forecasting Model with the chemistry/Data Assimilation Research Testbed (WRF‐Chem/DART) chemical weather forecasting/data assimilation system with multiconstituent data assimilation to investigate the improvement of air quality forecasts over eastern China. We assimilate surface in situ observations of sulfur dioxide (SO2), nitrogen dioxide (NO2), ozone (O3), carbon monoxide (CO), particulate matter with diameters less than 2.5 μm (PM2.5) and 10 μm (PM10), and satellite aerosol optical depth to adjust the related anthropogenic emissions as well as the chemical initial conditions. We validate our forecast results out to 72 hr by comparison with the in situ observations. Results show that updated emissions improve the model performance between 10% and 65% root mean square error reduction for the assimilated species except particulate matter with a diameter between 2.5 and 10 μm (PM2.5‐10), which is slightly improved due to the limited anthropogenic contribution to it. In a sensitivity experiment with a different update interval, the CO improvement is found to be sensitive to the cycling time used to update the CO emissions. In another sensitivity experiment when NO2 observations are not assimilated and nitrogen oxides (NOx) emission are adjusted by only O3, NO2 forecasts show similar root mean square error improvement but have lower spatial correlation, indicating the value and limitation of the O3‐NOx cross‐variable relationship.

Abstract. Observations of trace gases from space-based instruments offer the opportunity to constrain chemical and weather forecast and reanalysis models using the tools of data assimilation. In this study, observing system simulation experiments (OSSEs) are performed to investigate the potential of high space-and time-resolution column measurements as constraints on urban NO x emissions. The regional chemistry-meteorology assimilation system where meteorology and chemical variables are simultaneously assimilated is comprised of a chemical transport model, WRF-Chem, the Data Assimilation Research Testbed, and a geostationary observation simulator. We design OSSEs to investigate the sensitivity of emission inversions to the accuracy and uncertainty of the wind analyses and the emission updating scheme. We describe the overall model framework and some initial experiments that point out the first steps toward an optimal configuration for improving our understanding of NO x emissions by combining space-based measurements and data assimilation. Among the findings we describe is the dependence of errors in the estimated NO x emissions on the wind forecast errors, showing that wind vectors with a RMSE below 1 m s −1 allow inference of NO x emissions with a RMSE of less than 30 mol/(km 2 × h) at the 3 km scale of the model we use. We demonstrate that our inference of emissions is more accurate when we simultaneously update both NO x emissions and NO x concentrations instead of solely updating emissions. Furthermore, based on our analyses, we recommend carrying out meteorology assimilations to stabilize NO 2 transport from the initial wind errors before starting the emission assimilation. We show that wind uncertainties (calculated as a spread around a mean wind) are not important for estimating NO x emissions when the wind uncertainties are reduced below 1.5 m s −1 . Finally, we present results assessing the role of separate vs. simultaneous chemical and meteorological assimilation in a model framework without covariance between the meteorology and chemistry.

Abstract. This paper introduces the Weather Research andForecasting Model with chemistry/Data Assimilation Research Testbed (WRF-Chem/DART) chemical transport forecasting/data assimilation system together with the assimilation of compact phase space retrievals of satellite-derived atmospheric composition products. WRF-Chem is a state-ofthe-art chemical transport model. DART is a flexible software environment for researching ensemble data assimilation with different assimilation and forecast model options. DART's primary assimilation tool is the ensemble adjustment Kalman filter. WRF-Chem/DART is applied to the assimilation of Terra/Measurement of Pollution in the Troposphere (MOPITT) carbon monoxide (CO) trace gas retrieval profiles. Those CO observations are first assimilated as quasi-optimal retrievals (QORs). Our results show that assimilation of the CO retrievals (i) reduced WRF-Chem's CO bias in retrieval and state space, and (ii) improved the CO forecast skill by reducing the Root Mean Square Error (RMSE) and increasing the Coefficient of Determination (R 2 ). Those CO forecast improvements were significant at the 95 % level.Trace gas retrieval data sets contain (i) large amounts of data with limited information content per observation, (ii) error covariance cross-correlations, and (iii) contributions from the retrieval prior profile that should be removed before assimilation. Those characteristics present challenges to the assimilation of retrievals. This paper addresses those challenges by introducing the assimilation of compact phase space retrievals (CPSRs). CPSRs are obtained by preprocessing retrieval data sets with an algorithm that (i) compresses the retrieval data, (ii) diagonalizes the error covariance, and (iii) removes the retrieval prior profile contribution. Most modern ensemble assimilation algorithms can efficiently assimilate CPSRs. Our results show that assimilation of MO-PITT CO CPSRs reduced the number of observations (and assimilation computation costs) by ∼ 35 %, while providing CO forecast improvements comparable to or better than with the assimilation of MOPITT CO QORs.

No abstract