This section identifies and describes research topic areas considered medium priorities by workshop participants. These medium-priority topics were considered medium priorities by either the whole group of workshop participants, by the MPO-DOT workshop participants, or by one of the discussion sections (if the topic was not discussed by all participants). Each recommendation is described in the order in which it was prioritized by participants of the one-day workshop (see Tables 2 and 3). A brief description of the goals of that research, its value to the transportation community, and additional background information is given for each research topic. Multiple research topics were considered medium priorities as shown in Table 4. These research priorities are summarized in Table 7. Table 7 includes three medium-priority research needs that were identified by workshop participants but not included in the PM literature assessment (Tamura et al., 2005) .
|Research issue||Monitoring||Characterization||Emissions||Modeling||Controls||Recommendation number from Literature Assessment (Tamura et al., 2005). Also available in Table 1 in this document.|
|M.1. Collect information on fugitive dust emissions||X||LA-12|
|M.2. Compile a compendium of control strategy information||X||LA-18|
|M.3. Create short-term MOBILE fixes||X||New|
|M.4. Create a data information repository for MPOs and DOTs||X||New|
|M.5. Evaluate roadway project effects on emissions||X||LA-9|
|M.6. Improve information for MOBILE users regarding default assumptions||X||LA-13|
|M.7. Improve PM measurements||X||LA-2|
|M.8. Increase the spatial extent and temporal resolution of PM measurements||X||LA-3|
|M.9. Collect exhaust emissions from gross-emitters||X||LA-8|
|M.10. Estimate the uncertainty in the planning/emissions/air quality process||X||New|
Research goals. Understand and resolve gaps between fugitive dust emissions data and ambient concentrations near roadways. Determine possible dependency of fugitive dust emissions on speed and other parameters.
Value to Transportation Community. Resolving the discrepancy between fugitive dust emission inventory estimates and ambient concentrations may reduce the assumed need for costly fugitive dust control strategies in some areas. This research may also help existing or future models predict fugitive dust emissions from mobile sources with more accuracy.
Background. Fugitive dust can be a dominant component of PM emissions inventories, especially for larger sized particles such as PM10. EPA calculation methodologies attributed essentially all PM10 and PM2.5 emissions near roadways to fugitive dust prior to 2003 (U.S. Environmental Protection Agency, 2003). However, source apportionment of ambient concentrations have shown that fugitive dust is not as large of a contributor as emissions would indicate for PM10, and fugitive dust is a small component of PM2.5 (Watson and Chow, 2000; Countess et al., 2001; Fitz, 2001) .
The EPA's paved road fugitive dust emission model contains a number of unrealistic assumptions about fugitive dust emissions. This model depends solely on vehicle weight and roadway silt loading, does not incorporate any dependence on speed, and is only applicable to vehicles of at least two tons driving at speeds of 10-55 mph. Researchers have shown that speed can significantly influence fugitive dust emissions (Langston, 2004) .
Research goals. Compile and maintain an updated compendium of mobile source PM control measures for the transportation community.
Value to Transportation Community. A comprehensive compilation of control strategies with consistent evaluation of costs and benefits would provide transportation and air quality practitioners with an excellent resource for choosing the appropriate control strategy for their region. In addition, this research topic would help to identify which control strategies need additional evaluation of costs and benefits (see Section 3.4), and would be an excellent example of information that could be placed in an information archive for DOTs and MPOs (see Section 4.4).
Background. MPOs and DOTs need reliable information regarding control strategies, including the consequences for other air quality and transportation goals. Several control measure evaluations have been conducted, but evaluations have significant discrepancies due to methodological inconsistencies, or provide limited results. A few compilations of PM control strategies are currently available. These compilations include
Research goals. Create near-term patches for MOBILE to improve mobile source emissions estimates for the first round of PM conformity and SIP planning.
Value to Transportation Community. Regulatory guidelines require DOTs and MPOs to begin the PM2.5 conformity process one year following EPA designation of nonattainment areas, and require SIPs to be submitted for nonattainment areas by 2008. The new MOVES emissions model will not be available in time for DOTs and MPOs to use in the initial round of conformity and SIP development. In order to more accurately predict mobile source emissions for the initial round of PM2.5 conformity, short-term patches to the most serious problems in MOBILE should be added.
Background. MOBILE is the required emissions model most MPOs and DOTs use to calculate regional and project-level emissions, yet it has large uncertainties for its estimates of PM emissions. For example, when modeling heavy-duty diesel vehicle emissions, MOBILE does not vary PM emissions by speed. As mentioned in Section 3.3, MOVES is the new emissions model being developed by the EPA to replace MOBILE. MOVES will have a modal structure that will be more physically realistic in its prediction of PM. Since there is a regulatory requirement to use MOBILE for conformity determinations, it will be important to have any patches for MOBILE approved by the EPA for use in conformity determinations.
Research goals. Create and maintain an up-to-date information repository, such as a web site, for the transportation community.
Value to Transportation Community. A well-maintained web site with PM-specific data and information could be of benefit to transportation planners. Key information in such a repository could include
Many of the research topics directly pertinent to the transportation community could be distributed through such a central repository. For example, the compendium of control strategies (Section 4.2), MOBILE fixes (Section 4.3), and guidelines for users of tools (see Section 4.5) could all be included. On-line discussion forums could allow for exchange of insights concerning the usefulness of information archived, or could facilitate the exchange of advice from previous users of the archived information.
Research goals. Evaluate how changes in driving speed on a given facility impact emissions.
Value to Transportation Community. Available emissions modeling tools have a limited ability to vary PM emissions by speed because of the lack of adequate speed correction factors. Road grades (particularly for HDDV) and the impact of acceleration/deceleration on PM emissions are also issues for emissions modeling tools. The transportation community needs these models to accurately evaluate the effect of changes in average speed or congestion levels on emissions for conformity determinations.
Background. Emissions in MOBILE have traditionally been based on emissions data by type of trip, rather than data for different speeds on a given facility (e.g., freeway). MOBILE6.2 includes facility-specific speed correction factors for CO, VOC, and NOx, although further improvements are needed to differentiate arterial and freeway travel behavior. However, additional research is needed to evaluate how facility-specific speed changes affect PM and PM precursor emissions and how to incorporate this information into MOBILE and MOVES. These data are important to the transportation community to evaluate how congestion mitigation and facility choices impact PM emissions on the project and regional levels.
Research goals. Develop guidance for MOBILE users on which default inputs are most important to replace with local- or region-specific data.
Value to Transportation Community. Choosing local or default inputs to the MOBILE model can significantly impact PM emission predictions. The transportation community needs to know which inputs model results are most sensitive to. This research would provide a prioritized list of user inputs that have the most impact on predicted PM emissions for MOBILE users.
Background. MOBILE model users need to prioritize data collection efforts for model inputs for conformity determinations. Sensitivity analyses have been conducted (e.g., Tang et al., 2003; Granell et al., 2004) , but have not identified which of the sensitive inputs can be substantially improved or changed through data collection efforts. The importance of changing default values to local inputs depends on three factors:
Research goals. Improve measurement methods to more accurately measure PM, its components, and PM precursors.
Value to Transportation Community. Improved PM measurements will reduce uncertainty in source apportionment, and improve the ability of SIP strategies to achieve PM NAAQS, and reduce the risk that conformity emission budgets will be set arbitrarily or will need to be substantially adjusted at risk to the conformity process.
Background. Measurement methods to measure PM mass and its components are affected by positive and negative biases due to volatilization or adsorption of semi-volatile PM. Mobile sources are a major source of these semi-volatile species, which may result in PM measurements that may under or over-predict the relative proportion of individual PM components, and, thus, the importance of mobile sources to overall PM problems. Improved measurements of PM mass can help to reduce or correct for these biases.
Improved measurement methods to measure PM components may also help to quantify unique organic chemical tracers. The majority of chemical species emitted by mobile sources are organic compounds, which are difficult to identify and measure individually. These chemical tracers can be used to identify individual emissions sources such as diesel vehicles, gasoline vehicles, woodsmoke, and others. Therefore, improved chemical speciation of PM can be used to more accurately measure mobile source contributions to local or regional PM problems. Findings can also be used to support goals of roadside monitoring (Section 3.1) and emissions and hot-spot model evaluation (Sections 3.2 and 3.3).
Research goals. Provide more spatially and temporally resolved monitoring data for air quality models, source apportionment, and roadside monitoring data.
Value to Transportation Community. Increased resolution and spatial distribution of PM monitors and instruments will improve our ability to understand the influence of pollutant transport compared to localized sources and to assess exposures.
Background. Air quality models are used to characterize the contribution of mobile sources to ambient PM, set transportation and conformity emissions budgets, and evaluate control strategy efficacy. Ambient monitoring data from a large number of monitors with sub-daily resolution is needed to evaluate air quality models. These models cannot be rigorously tested using 24-hr average measurements from a few monitors. A large number of high quality measurements are needed to test whether an air quality model is predicting the concentration at the right time for the right reason. In addition, higher resolution data will be needed if comparisons are to be drawn to time periods with varying traffic activity. Also, there is increasing interest in weekend versus weekday pollution episodes, and greater time-resolution of PM data will be needed to distinguish weekend from weekday PM episodes.
Research goals. Evaluate the frequency and emissions of gross-emitting vehicles in the fleet.
Value to Transportation Community. Accurately assessing and understanding the total emissions from gross-emitters will help improve emissions modeling tools, thus assisting the transportation and air quality communities with creation of conformity emission budgets and selection of appropriate control strategies for reducing emissions.
Background. A small fraction of light-duty vehicles and heavy-duty diesel vehicles may be contributing a very large fraction of the total PM emissions. These vehicles, referred to as gross-emitters, may be the primary targets for control strategies. However, it is difficult to accurately determine the number of gross-emitting vehicles in the fleet or to assess their typical emissions profiles (i.e., a variety of vehicle attributes contribute to determining whether any one vehicle is a gross-emitter; thus, it is difficult to develop a "typical" profile for gross-emitters). Moreover, a gross-emitter of some chemical species such as NOx may not be a gross-emitter of PM. Gross-emitters are not included in current PM emissions models, but are currently treated in the post-processing stage. Additional evaluation of the fleet composition to determine the percentage of gross-emitting vehicles and their emissions profiles is needed to understand the importance of gross-emitters and to implement control strategies to reduce their numbers. Research is already underway to better evaluate the proportion of on-road PM emissions originating from light-duty versus heavy-duty vehicles, but definitive results have yet to be published, and additional gross-emitter evaluations will help to elucidate the main on-road PM problems.
Research goals. Diagnose and reduce the largest sources of uncertainty in the planning, emissions assessment, and air quality management processes.
Value to Transportation Community. Identifying and reducing the largest sources of uncertainty in the planning, emissions, and air quality process will help transportation and air quality practitioners to better estimate mobile source emissions.
Background. The tools and analysis methodologies used to demonstrate transportation conformity include large uncertainties. Notwithstanding these uncertainties, final conformity approval decisions are made based on a threshold value analysis that may estimate emissions to within hundredths of a ton. The implied precision of these determinations is out of scale with the uncertainties inherent in the tools used to complete the analyses. While the existing approach simplifies the conformity process, it does not adequately reflect the uncertain data and tools used to make the final decision. The unrealistic precision in conformity determinations has left many air quality and transportation planning professionals with the sense that conformity is a "paperwork exercise" contributing little value to transportation planning decisions. Analyzing the uncertainties in the planning, emissions, and air quality modeling process could identify those areas where the largest uncertainties exist. These highly uncertain areas could then be targeted for further research to reduce their uncertainties. The results may help to improve the quality of conformity assessments and enhance the ability of the conformity process to influence transportation planning decisions.