You are here:
Attributing differences in the fate of lateral boundary ozone in AQMEII3 models to physical process representations
Liu, P., C. Hogrefe, U. Im, J. Christensen, J. Bieser, U. Nopmongcol, G. Yarwood, R. Mathur, S. Roselle, AND T. Spero. Attributing differences in the fate of lateral boundary ozone in AQMEII3 models to physical process representations. Atmospheric Chemistry and Physics. Copernicus Publications, Katlenburg-Lindau, Germany, 18(23):17157-17175, (2018). https://doi.org/10.5194/acp-18-17157-2018
Quantifying the contributions of background ozone to surface concentrations has become increasingly important as ozone standards have tightened. This requires a detailed understanding of the relevant processes (vertical mixing in source and receptor regions, multi-day transport and chemistry in the free troposphere, and stratosphere/troposphere exchange processes) and how they are represented in current-generation global and regional-scale modeling tools. This study, performed in the context of the coordinated HTAP-AQMEII activity, focuses on a detailed comparison of the fate of inert tracers for lateral boundary conditions (representing background ozone) in four regional-scale air quality modeling systems applied over the U.S. for the year 2010. Utilizing inert tracers, the study found that differences in the representation of vertical turbulent mixing between the planetary boundary layer and the free troposphere is the main contributor to the model differences in the simulated inert tracer concentrations at the surface in most seasons and regions of the U.S. This highlights the importance of focusing model development, model evaluation, and measurement efforts on better characterizing aloft ozone concentrations and its interaction with the planetary boundary layer.
Increasing emphasis has been placed on characterizing the contributions and the uncertainties of ozone imported from outside the U.S. In chemical transport models (CTMs), the ozone transported through lateral boundaries (referred to as LB ozone hereafter) undergoes a series of physical and chemical processes in CTMs, which are important sources of the uncertainty in estimating the impact of LB ozone on ozone levels at the surface. By implementing inert tracers for LB ozone, the study seeks to better understand how differing representations of physical processes in regional CTMs may lead to differences in the simulated LB ozone that eventually reaches the surface across the U.S. For all the simulations in this study (including WRF/CMAQ, WRF/CAMx, COSMO-CLM/CMAQ, and WRF/DEHM), three chemically inert tracers which generally represent the altitude ranges of the planetary boundary layer (BC1), free troposphere (BC2), and upper troposphere-lower stratosphere (BC3), are tracked to assess the simulated impact of LB specification. Comparing WRF/CAMx with WRF/CMAQ, their differences in vertical grid structure explain 10%-60% of their seasonally averaged differences in inert tracers at the surface. Vertical turbulent mixing is the primary contributor to the remaining differences in inert tracers across the U.S. in all seasons. Stronger vertical mixing in WRF/CAMx brings more BC2 downward, leading to higher BCT (BCT=BC1+BC2+BC3) and BC2/BCT at surface in WRF/CAMx. Meanwhile, the differences in inert tracers due to vertical mixing is partially counteracted by their difference in sub-grid cloud mixing over the southeastern U.S. and the Gulf coast region during summer. The process of dry deposition adds extra gradients to the spatial distribution of the differences in DM8A BCT by 5 - 10 ppb during winter and summer. COSMO-CLM/CMAQ and WRF/CMAQ show similar performance in inert tracers both at the surface and aloft through most seasons, which suggests the similarity between the two models at process level. The largest difference is found in summer. Sub-grid cloud mixing plays a primary role in their differences in inert tracers over the southeastern U.S. and the oceans in summer. Our analysis on the vertical profiles of inert tracers also suggests that the model differences in dry deposition over certain regions are offset by the model differences in vertical turbulent mixing, leading to small differences in inert tracers at surface in these regions.
Record Details:Record Type: DOCUMENT (JOURNAL/PEER REVIEWED JOURNAL)
Organization:U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
NATIONAL EXPOSURE RESEARCH LABORATORY
COMPUTATIONAL EXPOSURE DIVISION
ATMOSPHERIC MODEL APPLICATION & ANALYSIS BRANCH