![[logo] US EPA](https://webarchive.library.unt.edu/eot2008/20081109143659im_/http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
Highlights
Past Highlights
January 2007
Community Modeling and Analysis System (CMAS) Sponsored Peer Review Report on Community Multi-Scale Air Quality (CMAQ) Modeling. The Community Modeling and Analysis System (CMAS) Center is coordinating an expert peer review of recent developments in air quality modeling by the Atmospheric Sciences Modeling Division. The particular emphases of the review are the modeling and interaction of meteorological, physical, and chemical processes of the Community Multi-Scale Air Quality (CMAQ) modeling program, as well as their applications and evaluations. Seven expert reviewers were charged with examining the quality, productivity, scientific relevance, and strengths and weaknesses of the components of the modeling program. The review took place in Research Triangle Park, North Carolina, during December 18-20, 2006. A draft report has been received for review in January, and a technical response to the points raised by the reviewers is being prepared. It is anticipated that the final report of the peer review committee will be submitted by CMAS to the Division by April 30, 2007. The final report and the technical response will be posted on the CMAS website at that time. (william.benjey@noaa.gov)
Community Multiscale Air Quality (CMAQ) Model – Bi-directional surface fluxes. Chemical surface fluxes have been represented in the Community Multiscale Air Quality (CMAQ) model as products of pre-computed dry deposition velocities and ground level concentrations. This method assumes zero concentration at the surface and thus precludes bi-directionality as well as surface saturation effects. We have developed a new surface exchange module within the CMAQ model that allows for more complex algorithms including stomatal and ground compensation points and non-linear component fluxes. For chemical species that behave according to simple linear dry deposition, the new scheme simply moves the dry deposition velocity calculation from the pre-processor (MCIP) to a subroutine within CMAQ. However, for species such as ammonia and mercury, both deposition and evasion can be represented. For ammonia, the bi-directional flux model includes compensation concentration computed as a function of surface temperature for the stomatal and ground pathways, and cuticle resistance computed as a function of surface relative humidity and the ratio of in-canopy concentrations of ammonia to SO2. This model has been tested in a ‘stand-alone’ configuration against field data provided by John Walker (EPA/ORD/NRMRL) and shown to replicate the typical nighttime and morning deposition and midday emission pattern. In CMAQ, the magnitude of the compensation point is computed according to the fractional area of agricultural land use. The CMAQ with the new bidirectional module is being run for the summer of 2002 to compare to the currently-released version of CMAQ (without the bidirectional module) and to field measurement data. (jonathan.pleim@noaa.gov)
Using Observations from a Dense Surface Mesonet to Improve Mesoscale Model Forecasts. Division staff members have been collaborating with a contractor to develop a methodology to best use observations from a dense surface mesoscale network (“mesonet”) to improve mesoscale meteorological model forecast initialization, verification, and interpretation. The focus of this work is to incorporate mesonet data from the Meteorological Assimilation Data Ingest System (MADIS) data portal and to use the Pennsylvania State University/National Center for Atmospheric Research Mesoscale Model (MM5) to generate short-term (~6-h) forecasts for the Washington, D.C. metropolitan area. This research is an extension of a broader collaboration within the NOAA Air Resources Laboratory as part of the UrbaNet research program. A Chesapeake Bay breeze case study was selected for the initial experiments, and twenty-two sensitivity experiments were conducted with various configurations of MM5 to ingest data from ~325 hourly surface mesonet sites for observation nudging during a 6-h pre-forecast dynamic initialization period followed by 6-h forecasts for a domain with 4-km horizontal grid spacing. The MM5 forecasts were then processed through the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) Model to create forward trajectories. There were important differences in the horizontal and vertical trajectories throughout the model domain between the MM5 using static initialization (no observation nudging) compared with the dynamically initialized MM5 forecasts (using the mesonet data for observation nudging). However, further analysis suggested that the differences between the forecast trajectories were related solely to the initialization time (t=0 for static initialization vs. t=-6 for dynamic initialization) and were independent of the dynamic initialization and the observation selection and weighting techniques that were used. Although there were some subtle differences between the sensitivity runs during the dynamic initialization period, all MM5 forecasts that used the same physics options converged to effectively the same forecast solution shortly after the observation nudging was discontinued. The forecast near-surface time series of wind speed, wind direction, temperature, and humidity were shown to be more sensitive to the choice of the planetary boundary layer scheme in MM5 than to the observation nudging techniques. These results were presented at the American Meteorological Society’s 14th Symposium on Meteorological Observations and Instrumentation in San Antonio, Texas, on 17 January 2007. (tanya.otte@noaa.gov, thomas.pierce@noaa.gov
SCIENTIFIC ACCOMPLISHMENTS
Air Quality Forecast Model Development and Testing. Model performance in California’s central valley is being investigated through diagnostic sensitivity analysis. During 2006, persistent under-predictions in surface O3 forecasts were noted in the region. Sensitivity runs examining possible uncertainties in California mobile emission estimates were performed by (a) doubling the on-road mobile emissions, (b) doubling the off-road mobile emission estimates, and (c) combination of a and b. Initial comparisons of model results from these simulations with surface measurements, along with analysis of emission data sets, suggest that uncertainties associated with off-road emission estimates may play a role in the noted under-predictions. Emission from this sector being compared in the model runs with estimates for the 2002 National Emissions Inventory. Increasing the O3 lateral boundary condition vertical profile within the boundary layer did not show a significant impact. The role of mixing is also being investigated through model simulations. This work will support model improvements for experimental O3 forecasting over the continental U.S. during 2007. (rohit.mathur@noaa.gov, george.pouliot@noaa.gov)
Using models to select monitoring sites for the near-road urban environment. A growing number of epidemiological studies conducted in various locations throughout the world suggest an increase in occurrence of adverse health effects for populations living, working or going to school near major roads. The U.S. Environmental Protection Agency’s (EPA) Office of Research and Development (ORD) is planning a joint study with the Federal Highway Administration in Las Vegas, Nevada. This project is a part of the ORD multi-laboratory effort to assess impacts of traffic emissions on near road air quality, population exposures, and adverse health effects. The primary objectives of this field study are to measure pollutant concentrations related to traffic patterns near major roadways and to characterize the pollutant dispersion away from the roadway.
Division scientists are participating in the study design, including modeling of the field site, and identification of meteorological and monitoring data needs. The analysis of meteorological data from three airports in Las Vegas provided critical information to select the list of potential study locations. For each of the proposed sites, model simulations are conducted to aid in design decisions. Division scientists are using the fine-scale wind field and dispersion model Quick Urban and Industrial Complex (QUIC). This model can be used to simulate the dispersion of pollutants released in geometrically complex urban areas. Using the meteorological information and the basic geometry of the schools and surrounding area (including the location of the roadways), the model simulated characteristic airflow patterns around the three sites. (vlad.isakov@noaa.gov , bowker.george@epa.gov)
Briefing the Ecosystem Research National Program Director on ASMD Multimedia Research. Three Division scientists briefed Rick Linthurst, the Ecosystem Research Program National Program Director (NPD) on January 3, 2007 and January 16, 2007. The briefings included an overview of the multimedia linkage work in ASMD and lessons learned. The overview was followed by more detailed descriptions of the work to enhance the Community Multiscale Air Quality Model (CMAQ) to make it more ecosystem friendly, including mercury bi-directional air-surface exchange, the work together with Chesapeake Bay and its accomplishments, the work with Tampa Bay, and the efforts to develop the Watershed Deposition Tool to allow watershed modelers to use CMAQ deposition outputs. The new effort to collaborate with the Ecosystem Research Division in Athens, GA on air-water model linkages for multimedia modeling was discussed on January 16th. (robin.dennis@noaa.gov)
Chesapeake Bay Meeting Atmospheric Deposition Meeting. A Division scientist participated in a meeting held January 4, 2007 in Annapolis, Maryland at the Chesapeake Bay Program Office to review the Chesapeake Bay Program’s use of air deposition inputs to model atmospheric deposition-associated loads to the Bay with the watershed model. The meeting helped to clarify the steps taken to convert modeling results into information requested by the managers of the Bay Program. The steps reviewed were deposition inputs to the watershed, derivation of nitrogen loading to the Bay, and development of future projections. Several questions about the air deposition modeling were clarified. It was decided that this type of exchange was useful and that it could continue on an ad hoc basis. (robin.dennis@noaa.gov)
North American Research Strategy for Tropospheric Ozone (NARSTO) Assessment Kick-off Meeting. NARSTO held a workshop with many of the authors that are expected to contribute to its Assessment of Technical Challenges of Applying Accountability-Based Air Quality Management with a Multi-Pollutant Framework. A division scientist has been invited to be one of the authors. The workshop was held January 9-10, 2007 in Durham, NC. Extensive discussion centered on clarifying the meaning of multi-pollutant risk assessments and the distinction between a multi-pollutant assessment and an accountability assessment. Risk paradigms, health effects and ecosystem effects perspectives were explored. A division scientist presented a description of one-atmosphere regional air quality modeling capabilities with regard to assessment needs for health and ecosystem effects. The co-chairs will continue to refine the focus of the assessment, given the discussion at the workshop, and revise the outline for the overall assessment. Workshops centered on health effects and ecosystem effects will occur this spring. A follow up meeting of the ecosystem group will take place in April in coordination with the National Atmospheric Deposition Program (NADP) meeting in Burlington, VT. (robin.dennis@noaa.gov)
Community Multiscale Air Quality (CMAQ) Model – Multipollutant Version. The CMAQ Multipollutant Model is being prepared for delivery to EPA’s Office of Air Quality Planning and Standards. Multiple sensitivity simulations with the multipollutant model, the air toxics model, the mercury model, and the base model (with chlorine chemistry) have been performed. The purpose of these tests was to normalize results between all of these model versions. To get consistent results for ozone between these models, some modifications were required in both the air toxics and mercury models. Results were presented to the model development team. In addition, model results with different compiler optimization levels on the SGI Altix machine are being tested and analyzed. (shawn.roselle@noaa.gov)
Community Multiscale Air Quality (CMAQ) Model –Assessing the Impact of Point-Source NOx Emission Controls. The CMAQ model has been applied using modeling scenarios designed to examine the effects of point-source emission reductions of nitrogen oxides (NOx) on hourly and maximum ozone concentrations in the eastern United States. These simulations have focused on model process analysis results to investigate the relative changes in different dynamic and chemical processes contributing to hourly ozone concentrations between the base case and NOx emission reduction scenarios. Model results for days with southwesterly wind flows across the Ohio River Valley during summer 2002 are of particular interest since relatively higher ozone concentrations in the eastern United States occurred during days with this flow pattern. Decreases in ozone and total nitrogen species concentrations were found downwind of point sources in the emission reduction scenario. Spatial patterns of the chemical production of ozone, vertical mixing, and dry deposition over the daytime period also revealed differences between the base and emission reduction scenarios. Areas downwind of major point sources that experienced large NOx emission reductions exhibited somewhat lower ozone dry deposition and less vertical mixing of ozone. Significant decreases in the photochemical production of ozone were found downwind in the emission reduction scenario. These results corroborate earlier analysis results which revealed that lower maximum 8-hour ozone values occurred in areas downwind of major emission regions, such as the Ohio River Valley, after substantial NOx emission reductions. However, the emission reductions increased ozone production efficiency (OPE) - the net number of ozone molecules produced per molecule of NOx consumed - over large areas of the eastern United States. Different processes are being investigated along trajectory paths emanating from grid cells containing major point sources exhibiting substantial NOx emission reductions to assess the differences in chemical production and ozone concentrations with distance downwind of point sources. Analysis of vertical profiles of modeled NOx concentrations downwind during the early morning hours indicated concentration decreases in the post-control scenario at all heights; however, the largest NOx concentration reductions, as expected, were found in the elevated plumes. (james.godowitch@noaa.gov)
Incorporation of Pleim-Xiu Land Surface Model into the Weather, Research and Forecasting Model (WRF) Model. The Weather, Research and Forecasting (WRF) model, a new generation mesoscale weather model, will soon be used to produce meteorology for our air quality simulations. To make this new model consistent with the current Mesoscale Model version 5 (MM5), the Pleim-Xiu Land Surface Model (P-X LSM) is being incorporated within WRF. The long awaited new version of the Weather, Research and Forecasting model (WRFV2.2) was released in late December 2006. In January 2007, we focused on porting the Pleim-Xiu Land Surface Model, the Asymmetric Convective Model version 2 (ACM2) boundary layer model and surface layer scheme into this new version of WRF. All of these new physics options have been tested and compared to the equivalent Mesoscale Meteorological model Version 5 (MM5) and default physics setup of the WRF model. Tests indicate reasonable model performance for a July 1 - 8, 2002 simulation period; however, this model configuration simulates warmer than observed temperatures during the daytime. The method of soil temperature initialization in the WRF model, in particular the deep soil temperature, is different than that of the Mesoscale Meteorological model Version 5. Additionally, the surface analyses are not currently available for the soil moisture nudging scheme; these should improve the simulations when they are available. Our current efforts are to improve the soil temperature initialization and implement the soil moisture nudging. (robert.gilliam@noaa.gov, jonathan.pleim@noaa.gov)
Considering Change of Support when Evaluating an Air Quality Model. Using summer 2001 ozone predictions and observations, a variety of ways are being tested to address the Achange of support@ issue. Change of support occurs when comparing a point-based monitor observation to a grid cell average value produced by CMAQ. The first method develops a statistical prediction of ozone values from observations at a set number of points within each grid cell, and then these predictions are averaged for each grid cell. Block kriging, an additional method being tested, is similar to the first method but computationally more efficient. Cross validation is being used to compare what complexity of spatial model is necessary. This approach is also being applied to summers 2002 and 2004 to test the block kriging procedure for dynamic evaluation of the NOx SIP Call (EPA’s requirement for 22 eastern states to address the regional transport of ground-level ozone). We are computing the observation-based gridded spatial fields for June, July, and August of these years to then compute the change in 8hr max O3 for the CB4 and SAPRC runs. Once this is completed we can compare the results to the scatter plots using the Amatched@ observed and model values. (Foley.Kristen@epa.gov)
Analysis of Health Data for North Carolina (NC). Some simple poisson regression models have been applied for the Mecklenburg County, NC mortality data using observed ozone, PM2.5 and meteorological variables. Currently, code is being developed to estimate a distributed lag model that is typically done with these types of data. (Foley.Kristen@epa.gov)
Space-time modeling: ASMD developed a statistical space-time model based on principles of the dynamic linear model (DLM) and process convolution/kernel smoothing approaches. Such models are not part of standard statistical software packages (such as SAS or R), so customized C code had to be developed. Several iterations were needed to improve the efficiency of this code, since the model fitting procedure uses Markov chain Monte Carlo sampling procedures and includes extensive linear algebra and other calculations. (jenise.swall@noaa.gov)
Testing a New version of R software: ASMD installed the newest version of R (2.4.1) – a system for statistical computation and graphics - for use with HPCC computers running Red Hat Linux 9.0 (or more current versions). R packages commonly used in the division were also updated as necessary, re-compiled, and re-installed. Installations included the new version of the RMET (R Model Evaluation Toolkit) packages developed under contract by Battelle. The new version of R, with its related packages, is being tested by a core group of R users in the division. If all continues to go well, this new version will soon become the division=s standard. (jenise.wwall@noaa.gov)
Development of a Biomass Burning Emissions Inventory Using Satellite and Ground-Based Information. The National Emissions Inventory (NEI) for biomass burning (wild land fires, prescribed burning, agricultural burning), until recently, has been extremely limited about the specific location and timing of major fire events. For years 2001 and earlier, biomass-burning related emissions are spatially resolved for the conterminous 48 states and temporally resolved to monthly time-scales. For the 2002 NEI, a detailed fire inventory was prepared by the Regional Planning Organizations (RPOs) and combined into a single national inventory by the EPA. However, this process has been time-consuming and expensive. For regional air quality applications and modeling, spatially and temporally resolved fire emissions are needed. The 2006 NARSTO report indicated that emissions from open biomass burning, including agricultural and forest prescribed burning, wildfires, and residential backyard burning need to be characterized with greater accuracy. It will be prohibitively expensive to create a detailed fire inventory at the national level as was done for the 2002 NEI; therefore, an approach to estimating wildland fire emissions that is better than the pre-2002 method but not as time consuming as the 2002 NEI is needed. A method has been devised for combining satellite-based biomass burning activity information with existing ground-based fire activity information. Our assumption is that neither ground-based nor satellite-based information is complete, so tour goal is to combine the information from both sources developing some relationship between satellite-derived pixel counts and fire size (burned area). (george.pouliot@noaa.gov, roy.dev@epa.gov, thomas.pierce@noaa.gov, and mobley.david@epa.gov)
RESEARCH ACTIVITIES
Urban Canopy Parameterization. ASMD began a new project to examine the simulation of air toxics concentrations in urban areas using parameterization of the urban landscape. Both the near-field and Eulerian grid models (e.g., AERMOD and CMAQ) lack a rigorous treatment for urban surfaces. This project will examine several areas to provide a practical operational treatment of urban areas and lead to improved predictions of air toxic concentrations. First, available urban canopy parameterizations will be reviewed. Next, the viability of working algorithms with urban meteorological field measurements will be evaluated. Ultimately, the algorithm will be transferred into the CMAQ modeling system for operational use. The use of urban canopy parameterization will also be evaluated in the AERMOD modeling system to provide a consistent treatment of flow and turbulence in both AERMOD and CMAQ. (alan.huber@noaa.gov)
Weather Research and Forecasting Model with Chemistry (WRF/Chem). The Division has acquired the January 2007 release of Weather Research and Forecasting model with Chemistry (WRF/Chem) Version 2.2 source code to further develop in-house expertise with prognostic models using on-line coupled dynamics and chemistry,. This version of WRF/Chem comes from Dr. Jerome Fast’s group at the Pacific Northwest National Laboratory and includes features such as the Fast-J photolysis routine, the CBM-Z chemical mechanism, and the MOSAIC sectional aerosol module with direct feedback to the radiative forcing and microphysics. Version 2.2 of the Advanced Research WRF (ARW) meteorological dynamic core of this WRF/Chem model now includes four dimensional data assimilation (FDDA) analysis and observation nudging, and positive-definite advection for turbulent kinetic energy, moisture, scalars, and chemical species. WRF/Chem V2.2 was successfully installed and tested on the Division’s computer clusters. The new WRF Preprocessing System (WPS) is being applied to a domain of 12 km grid cells within the Eastern United States for an early August 2006 case study to simulate bidirectional feedbacks between meteorological and chemical processes during an eastern U.S. ozone episode. Results from the case study will likely be compared with the operational National Air Quality Forecasting guidance. (jerry.herwehe@noaa.gov)
Enhancements to Aqueous Chemistry in CMAQ. The Community Multiscale Air Quality (CMAQ) Aqueous Chemistry (AQCHEM) subroutine has been changed to calculate cloud-produced secondary organic aerosol (SOA) from aqueous reactions of glyoxal and methylglyoxal. The new aerosol variable is called AORGC. CMAQ is being applied to August 2004, using SAPRC99 and AE4 to compare with a base run (using the original AQCHEM) for the same time period. (ann.carlton@noaa.gov)
CONFERENCES & WORKSHOPS
EPA Workshop on Integrated Modeling. Five Division scientists attended a workshop on Integrated Modeling for Integrated Environmental Decision Making, held January 30, 2007 to February 1, 2007 at the Environmental Protection Agency (EPA) in Research Triangle Park, North Carolina. The workshop was sponsored by the EPA Council on Regulatory Environmental Modeling (CREM) and hosted, in part, by scientists from the EPA’s Ecosystem Research Division (ERD). The goals were to build a vision for the role of integrated modeling in informing EPA’s decision-making and share experiences and perspectives on past modeling successes, lessons learned, and emerging priorities.
Integrated modeling was broadly defined as a model that includes sub-models for multiple media, pollutants, or cross multiple disciplines. Integrated modeling was agreed to be a powerful decision-making tool when pollution models were integrated with risk, ecological, or exposure models. The ever-increasing computer computational power is expected to be used to further integrate models. The increasing use of this technique indicated the need for robust model evaluation techniques to address the uncertainties of the growing complexity of integrated models. Standards in model input and output data, robust model and data evaluation, and model sensitivity to target monitoring and field experiments were recommended.
There were lively discussions in the breakout groups held throughout the workshop. Two Division scientists were co-chairs of breakout sessions. The organizing committee will develop a white paper based on the workshop feedback to further develop a vision for integrated modeling. The CMAQ community can play a role here as it continues to forecast, link global and local scales, and link air and water models. Another objective of the workshop was to build bridges among the different modeling communities. ASMD scientists presented a poster on the Watershed Deposition Tool that sparked great interest by several participants, producing contacts for further interactions. The extent of CMAQ modeling and interactions in several arenas of multi-media modeling was new to many of the participants. This was a good opportunity to let people know about modeling work in ASMD and build bridges for future collaborations. (robin.dennis@noaa.gov. , jesse.bash@noaa.gov)
87th Annual Meeting of the American Meteorological Society, 14th Symposium on Meteorological Observations and Instrumentation, San Antonio, Texas, January 14-18, 2007. Tanya Otte attended this meeting and presented her research in the UrbaNet program related to using a dense network of surface observations to improve short-term, high-resolution meteorological model forecasts. Ms. Otte also co-authored a presentation that described the vertical coupling between the National Weather Service’s operational Weather Research and Forecasting Non-Hydrostatic Mesoscale Model (WRF-NMM), and the Community Multiscale Air Quality (CMAQ) Model for the air-quality forecasting system.
Evelyn Poole-Kober also attended this meeting, and presented the new model for library services that is used at the Environmental Protection Agency. Ms. Poole-Kober co-authored a presentation that described the ten-year history of Atmospheric Science Librarians International (ASLI).
Atmospheric Science Librarians International Conference. At the ASLI conference, Ms. Pool-Kober chaired a session on “Publishing and Product Updates.” Ms. Poole-Kober’s Division summer intern, Kari Kozak (a student at University of North Carolina), also attended the ASLI conference and presented her internship work on digitizing documents to make them available via the internet. Ms. Poole-Kober received the ASLI Spirit Award, January 17, 2007, for her sustained leadership in membership recruitment, offices held, and many other ASLI activities.
Presentations included:
Lin, H.-M., R. Mathur, T.L. Otte, P.C. Lee, and J. Pleim. On the development of linkage in the vertical coordinate between the NAM and CMAQ models.
Needles, M., M. Latyszewskyj, and E. Poole-Kober. Celebrating ten years of learning and networking – The Atmospheric Science Librarians International.
Otte, T.L., L.J. Reynolds, A. Huffman, T.E. Pierce, and R.R. Draxler. On the use of observations from a dense mesonet to improve short-term, high-resolution mesoscale model forecasts.
Poole-Kober, E. The new model for library services at the Environmental Protection Agency.
Kozak, K. From the warehouse to the information super highway: Digitizing NOAA documents.
VISITORS
Derek Rice, NOAA Undergraduate Hollings Scholar
Chemical Engineering
Bucknell University
Lewisburg, Pennsylvania
Derek Rice visited the Division, January 4, 2007, for an orientation prior to his summer research internship that will begin May 29, 2007. (prakash.bhave@noaa.gov)
PUBLICATIONS
G.E. Bowker and H.C. Crenshaw. Electrostatic forces in wind-pollination: Part 1, Measurement of the electrostatic charge on pollen. Atmospheric Environment, 41, 1587-1595 (2007). Doi:10.1016/j.atmosenv.2006.10.047.
G.E. Bowker and H.C. Crenshaw. Electrostatic forces in wind-pollination: Part 2, simulations of pollen capture. Atmospheric Environment, 41, 1596-1603 (2007). DOI:10.1015/j.atmosenv.2006.10.048.
V. Isakov, A. Venkatram. J. Touma. D. Koracin, and T.L. Otte. Evaluating the use of outputs from comprehensive meteorological models in air quality modeling applications. Atmospheric Environment, 41, 1352-2310 (2007). DOI:10.1016/j.atmosenv.2006.10.043.
Bhave, P.V., G.A. Pouliot, and M. Zheng. Diagnostic model evaluation for carbonaceous PM2.5 using organic markers measured in the southeastern U.S. Environmental Science & Technology, Web publication date: January 26, 2007.
Reff, A., S.I. Eberly, and P.V. Bhave. Receptor modeling of ambient particulate matter data using positive matrix factorization. Journal of Air and Waste Management Association 57:146-154 (2007)
Hanna, S.R., M.J. Brown, F.E. Camelli, S.T. Chan, W.J. Coirier, O.R. Hansen, A.H. Huber, S. Kim, and R.M. Reynolds, 2006: Detailed simulations of atmospheric flow and dispersion in downtown Manhattan. Bulletin Am. Meteorol. Soc. , 87 :12, 1713-1726.
Singh, R, A. Huber, and J Braddock. Sensitivity Analysis and Evaluation of MicroFacPM: A Microscale Motor Vehicle Emission Factor Model for Particulate Matter Emissions. JOURNAL of Air & Waste Assoc (in press).
Roy, R. Mathur, and A. Gilliland, A comparison of CMAQ‑based aerosol properties with IMPROVE, MODIS and AERONET data, Journal of Geophysical Research-Atmospheres, in press.
Yu, S., R. Mathur, K. Schere, D. Kang, J. Pleim, and T. L. Otte, 2007. A comprehensive evaluation of the Eta-CMAQ forecast model performance for O3, its related precursors, and meteorological parameters during the 2004 ICARTT study. Journal of Geophysical Research, in press
Yu, S., P.V. Bhave, R.L. Dennis, and R. Mathur, 2007. Seasonal and regional variations of primary and secondary organic aerosols over the continental United States: Observation-based estimates and Model evaluation. Environmental Science & Technology, revised version submitted.
T.J. Sullivan, B.J. Cosby, J.R. Webb, R.L. Dennis, A.J. Bulger, and F.A. Deviney. Streamwater Acid-Base Chemistry and Critical Loads of Atmospheric Sulfur Deposition in Shenandoah National Park, Virginia, submitted to Journal of Environmental Monitoring and Assessment (submitted).
G.E. Bowker, R. Baldauf, V. Isakov, A. Khylstov, and W. Petersen. Modeling the effects of sound barriers and vegetation on the transport and dispersion of air pollutants from roadways (in review).
C.G. Nolte, A.B. Gilliland, and C. Hogrefe.Linking global to regional models to assess future climate impacts on air quality in the United States: Surface ozone concentrations. Journal of Geophysical Research (In review).
K.W. Appel, A.B. Gilliland, G. Sarwar, and R.C. Gilliam. Evaluation of the Community Multiscale Air Quality (CMAQ) model version 4.5: Uncertainties and sensitivities impacting . model performance; Part I - ozone.Atmospheric Environment (In review).