7.5A Evaluation of the New Dust Treatment in the Community Multiscale Air Quality (CMAQ) Model Version 5.0

Thursday, 10 January 2013: 2:30 PM
Room 16A (Austin Convention Center)
K. Wyat Appel, EPA, Research Triangle Park, NC; and G. Pouliot, H. Simon, H. Pye, S. Napelenok, J. Young, and S. Roselle

Previous operational performance evaluations of the Community Multiscale Air Quality (CMAQ) model indentified dust as a large source of PM2.5 bias in several seasons. The latest release of CMAQ (v5.0) includes updated treatment of dust (often referred to as soil). These updates include a mechanism for a calculating wind-blown dust emissions within the model and explicit speciation of trace metals (i.e. Fe, Al, Ca, Si, K, Mg, Mn and Ti), which are major constituents of dust and were previously lumped into a catch-all species called PMother. In addition, important updates have been made to emissions including adjustments for the effects of snow cover and soil moisture on fugitive dust, updates to the “transportable fraction”, and explicit speciation of dust and other anthropogenic PM2.5¬ emissions sources to include the newly tracked trace metals. In addition to the goal of improving the model performance of PMother, the explicit speciation of the trace metals allows for improvements in the model chemistry as well, as several of new trace metal species can catalyze aqueous sulfate formation and participate in other inorganic chemistry processes. The earlier treatment of this chemistry relied on a background concentrations for these metals which did not vary in space and time, whereas CMAQ now integrates model-predicted Fe and Mn concentrations into the sulfate reactions. An annual 2006 CMAQ model simulation is used to analyze the impact of these model updates on dust and sulfate estimates across the United States. The model estimated concentrations of dust and trace metals are compared to surface observations across the continental United States. In addition to the operational evaluation, several sensitivity simulations have been performed to diagnostically evaluate the sources of model error. This work will report on the results of these operational and diagnostic evaluations.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner