Regional model simulations of meteorology and chemistry for the period covering the NASA DISCOVER-AQ field experiment
C. P. Loughner
K. E. Pickering
October 26, 2012 15:26:17
Description of Problem
It is difficult to interpret what satellite observations of air pollutants throughout the entire vertical column of the atmosphere means in terms of what pollutants people breathe at Earth’s surface. DISCOVER-AQ aims to close this gap by forming relationships between column content and surface concentrations. In addition, DISCOVER-AQ aspires to further improve how satellite observations are interpreted when large spatial and/or temporal variability of pollution is present. These goals will be achieved by using observations of air pollution and meteorological variables from satellite, aircraft, sondes, tethered balloons, ground, and ship based observations as well as meteorological and air quality model simulations to determine how current and future satellite observations can more effectively diagnose ground level air pollution. Observations were collected during the first of four field campaigns during July 2011 in the Washington, DC and Baltimore, MD metropolitan areas. Model simulations covering this field campaign are currently underway to help interpret the observations and achieve the above mentioned goals. In addition, the observations and model simulations will be used to evaluate the models and determine the spatial and temporal variability of air pollution deposition.
Scientific Objectives and Approach
Meteorological and air quality model simulations are being performed to achieve the above mentioned objectives by following these steps:
Simulate the meteorology for the DISCOVER-AQ field campaign with the Weather Research and Forecasting (WRF) model. Four modeling domains are used at horizontal resolutions of 36, 12, 4, and 1.3 km. The finest domain is centered over the Washington, DC and Baltimore, MD metropolitan areas and the Chesapeake Bay.
Prepare emissions input files for the Community Multi-scale Air Quality (CMAQ) model for the 36, 12, 4, and 1.3 km horizontal resolution modeling domains and the WRF model coupled with Chemistry (WRF-Chem) for the 36, 12, and 4 km modeling domains. Meteorological model output from the WRF model simulation is required as input for creating emissions input files for the CMAQ and WRF-Chem models. Anthropogenic emissions input files are created by processing projected 2012 emissions from the 2005 National Emissions Inventory with the Sparse Matrix Operator Kernel Emissions (SMOKE) model, biogenic emissions are created with the Model of Emissions of Gases and Aerosols from Nature (MEGAN), and biomass burning emissions are from the Fire Inventory from NCAR (FINN). WRF-Chem emissions input files are passed on to other scientists to perform WRF-Chem and NASA Unified-WRF (NU-WRF) model simulations.
De-bug the WRF-Chem model and pass on the model fixes to other scientists for them to incorporate into their versions of the WRF-Chem and NU-WRF model codes.
Prepare chemical initial and boundary conditions for the CMAQ model with output from the Model for Ozone and Related Chemical Tracers chemical transport model (MOZART CTM).
Run the CMAQ model for the 36, 12, 4, and 1.3 km horizontal resolution domains and make the model output available for other scientists to perform their own analyses.
Analyze the CMAQ model output alongside observations to evaluate the model simulation and investigate the role of the Chesapeake Bay breeze on the spatial and temporal variability of air pollution concentrations, column content, and deposition over and near the Chesapeake Bay.
Steps 1-4 described above have been completed. WRF-Chem model setup and code bug fixes have been applied by other scientists at the NASA GSFC following my instructions. CMAQ Version 5.0 was released in late February 2011. I am currently testing the CMAQ code and a simulation covering the 2011 DISCOVER-AQ field campaign will be underway in the near future.
A preliminary analysis of the vertical variability of air pollution was conducted over the Chesapeake Bay using in-situ ship and aircraft based observations alongside the NOAA-NMM-CMAQ modeling system, which is run at a horizontal resolution of 12 km. It was found that this model does not have a large enough vertical gradient compared with observations in the stable marine boundary layer. In addition, the model does not simulate elevated ozone aloft at around 950 mb, which the observations show. This suggests the model is not transporting pollution vertically associated with a bay breeze circulation. Model simulations at 12 km or coarser are not expected to capture local scale circulations, like a bay breeze, but they should be captured with resolutions at 1.3 and 4 km.
Other Publications and Conferences
Loughner, C.P., D. Goldberg, K.E. Pickering, M. Tzortziou, A. Weinheimer, R.A. Ferrare, C. Hostetler, P. Lee, J.H. Crawford, A. Mannino, D.J. Knapp, D.D. Montzka, L.T. Marufu, R.R. Dickerson, J. Hair, R. Rogers, and M. Obland, 2011: Evaluation of CMAQ boundary layer processes and air quality over the Chesapeake Bay and Maryland, 3rd International Workshop on Air Quality Forecasting Research, Potomac, MD.
Loughner, C.P., D. Goldberg, M. Tzortziou, A. Cede, N. Abuhassan, C. Retscher, A.J. Weinheimer, R.A. Ferrare, C.A. Hostetler, P. Lee, K.E. Pickering, J.H. Crawford, A. Mannino, J.R. Herman, D.J. Knapp, D. Montzka, T.L. Marufu, R.R. Dickerson, J.W. Hair, R.Rogers, M.D. Obland, 2011: Horizontal and vertical distribution of air pollution over Maryland and the Chesapeake Bay, AGU Fall Meeting, San Francisco, CA.