Input Guidelines for Motor Vehicle Emissions Simulator Model, Volume 3: Final Report (2015)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4.0 Practitioners’ Survey 4.1 OVERVIEW As part of the Phase 1 research for this project, a survey was conducted of agencies using or transitioning to the use of MOVES. The survey was conducted in July through September 2012. The objectives of the survey were to characterize how agencies are using the MOVES model, current practices in developing MOVES inputs (including data sources and processing methods), and needs for additional guidance or sample data inputs. This information was used to help identify and prioritize Phase 2 activities, and also to identify tools, data sources, and methods that could be highlighted in the Practitioners’ Handbook. This section provides a summary of the findings of a web-based survey of data sources and current practice for MOVES. Appendix C presents the survey responses in detail, and Appendix D contains the survey instrument. The survey sample was recruited by distributing the web-based survey to as many transportation and air quality agencies as possible within areas that are subject to, or expected to be subject to, transportation conformity. Organizations that represent MOVES stakeholders in these communities assisted with distributing the survey, including the Association of Metropolitan Planning Organizations, American Association of State Highway and Transportation Officials (AASHTO) Standing Committee on the Environment, AASHTO Air Quality Subcommittee, and the National Association of Clean Air Agencies. Survey information was also distributed at the Northern Transportation Air Quality Summit August 6-8, 2012 by members of the NCHRP 25-38 panel. Efforts were made to obtain survey responses from a cross-section of agencies by geographic region of the country and size of population served (both metropolitan and state). While the survey was primarily distributed by email, some phone calls were made to attempt to recruit agencies from categories underrepresented in initial responses, such as smaller MPOs. After presenting respondents and uses of MOVES, the discussion of survey results is categorized into fleet data, regional activity inputs, and project activity inputs to mirror the organization of the literature review. For most survey questions there were about 70 responders for each question. Because most MPOs do not do project-level studies, the response rate dropped to about 50 for many of the questions related to project-level inputs. 4.2 RESPONDENTS AND USES OF MOVES Seventy-nine responses were received representing up to 38 MPOs, 14 state DOTs, 22 state air agencies, and 5 agencies of other types, including city DOTs, 4-1

metro air districts, and county public health agencies. (Not all respondents identified the name of their agency, and in a few cases, two people at a single agency responded.) As anticipated in the design of the sampling plan, the sample is biased toward regions or agencies that are subject to the requirements of transportation conformity, with enough staff and resources to have developed some familiarity with MOVES. • About two-thirds of the respondents represented either MPOs with populations exceeding one million, or state or multistate jurisdictions. However, representation was also obtained from areas of other sizes, and 12 percent of respondents represented agencies serving a population under 200,000. • Only three respondents indicated that their area was in full attainment of ambient standards for carbon monoxide (CO), ozone, and particulate matter (PM10, PM2.5). The remainder represented areas that were fully or partially classified as nonattainment or maintenance areas. • More than half of the respondents indicated that they already had used MOVES, and most of the rest were transitioning to or planning to use MOVES. Only 4 percent indicated that they had no intent to use MOVES. • The most common applications of MOVES were for SIP development and regional transportation conformity, with the majority of respondents using MOVES for both purposes. About 45 percent were using or planned to use MOVES for greenhouse gas analysis or policy/legislative evaluation. About 20 percent planned to use MOVES for mobile source air toxics (MSAT) analysis, either at a regional or project level. • Very few agencies (4 percent of respondents) had used MOVES for project- level analysis, although 40 percent planned to use it at this level for conformity analysis, and about 30 percent planned to use it for environmental assessment/documentation or GHG analysis. Fleet Data Inputs Locally derived data, primarily state vehicle registration databases, were the most common source for age distribution inputs. For light-duty vehicles, 75 percent of respondents used either local data for age distributions, or a combination of locally derived data and MOVES defaults. For heavy-duty vehicles, 64 percent made some use of local data. Similarly, the majority of agencies were using state registration data for vehicle source type distributions. MOVES defaults were often used with the local data to fill in gaps where needed. For example, MOVES default distributions may be used to proportion the light-duty truck population into the subset used as passenger vehicles and the subset used as commercial vehicles. 4-2

Regional Activity Inputs About half of all respondents (51 percent) were using both travel demand forecasting model (TDFM) and HPMS data to estimate regional VMT by source type, while most of the rest (41 percent) were using one of these sources alone. Source type distributions were estimated from data on vehicle class groups, which were then disaggregated into the 13 MOVES vehicle classes using a surrogate data source. The most common surrogates used were the MOVES default source type distributions, source type distributions adopted from MOBILE, or state vehicle registration data. Table 4.1 summarizes responses on surrogate data sources. Based on comments provided, “other” responses typically indicated that data from another agency or MOBILE6.2 data were used.1 Table 4.1 Surrogate Data Used to Map VMT to MOVES Source Types Frequency Percentage MOVES defaults 17 29% Registration data 11 19% Other 17 29% No surrogates used 3 5% Don’t know 11 19% The source for VMT-by-month fractions and day-of-week fractions was fairly evenly split among HPMS data, MOVES defaults, and “other.” The comments supplied with the “other” responses indicated that many MPOs were receiving these data from state agencies, and that often the data were from permanent traffic count stations which may or may not be part of the HPMS system. The source for VMT-by-hour fractions followed this same pattern with the difference that 11 percent of the responders indicated that VMT-by-hour fractions were derived from the TDFM or from traffic studies. Many agencies used a mix of data sources for the road type distribution and ramp fraction inputs. The most common were the TDFM and HPMS data, or some combination of the two. Fifteen percent of respondents reported using MOVES defaults for these data. The source for VMT by speed bins was also dominated by TDFM and HPMS data, with 20 percent of respondents reporting using MOVES defaults for these data. Of those using TDFM output to estimate 1 Throughout the survey, respondents who answered “other” often mentioned that they got the data from another agency, so use of the “other” category does not necessarily mean that a different underlying data source was used. 4-3

speed data, about 30 percent reported postprocessing the model results to improve the results for use with MOVES. Project-Level Inputs Link activity estimates are derived from a variety of sources. The most common source of traffic volume data reported was TDFM output (one-third of responders). Based on the comments, contractor data was also used, which suggests that volume data is probably being taking directly from traffic impact studies. For source type data, about 15 percent used local classified traffic counts; the remainder took data from state or regional analysis, MOVES defaults, or another source. Age distribution data was almost always from state-level or default sources. For operating mode and vehicle speed data, 46 percent of respondents reported using TDFM speeds and 30 percent reported using either Highway Capacity Manual (HCM) and/or microsimulation to determine project-level speeds. Only 4 percent of respondents had developed vehicle-specific power (VSP) profiles for use in project-level analysis. Responders had little experience with off-network activity estimates (including start fraction, parked fraction, and extended idle fraction). Eighty-one percent reported that their agency either had not estimated off-network data or did not know if their agency had done this. Other MOVES Inputs For temperature and humidity inputs, about one-third of respondents reported using National Climatic Data Center data and 44 percent reported using locally obtained data. For areas with an I/M program, about three-quarters developed program-specific inputs rather than using MOVES defaults. For fuel data, most respondents reported using data from local and/or state sources; only 17 percent reported using MOVES defaults. About one-quarter of respondents reported that they had performed some type of sensitivity testing on MOVES inputs. Input on NCHRP 25-38 Project Outcomes Respondents were asked what datasets or methods/guidance would be the most important to their agency. Over 60 percent identified age distributions by vehicle class, source type population, and regional vehicle class splits as being “very important.” In contrast, just under one-third identified project-level vehicle class splits or project-level operating mode distributions as “very important.” This appears to reflect the more widespread use of MOVES for regional-level as compared to project-level analysis; however, guidance on these inputs was still important for those who plan to use MOVES at a project level. Meteorology data was also identified as “very important” by one-third of respondents. A variety of other free-response answers were provided to this question. 4-4

About one-third of respondents said they would prefer data sets to be provided as part of MOVES, with about one-half preferring another publication or on-line data center. Comments provided suggested that this might depend upon the specific dataset, timeliness of updates, etc. Observations and Trends There was no established best practice for estimating MOVES inputs identified in the survey. The majority of agencies appeared to be using the best data that is readily accessible. Where that information is insufficient, it is often augmented by MOVES defaults. Commonly used sources are the same as were often used for developing MOBILE inputs, e.g., HPMS, travel demand forecasting models, and state vehicle registration data, which rarely provide the full level of detail that MOVES is capable of accepting. New, locality-specific data collection efforts focused specifically on MOVES requirements have not been widely undertaken. The bulk of experience to date with MOVES is in performing regional analysis by agencies with a history of doing similar work with MOBILE. That experience is also predominantly with criteria pollutants to meet SIP and regional conformity requirements. There is less interest in expanding the use of the model to MSAT or GHG analysis where there is not a clear requirement to do so, although a number of agencies are still planning to use MOVES for these purposes. Project-level analysis is likely to be conducted by project sponsors (such as state or local DOTs) when required, rather than by MPOs. Very few agencies have experience with developing locally specific inputs for project-level analysis. The Practitioners’ Handbook is intended to help agencies fill in some gaps by showing how available data sources can be used to develop more detailed locality-specific MOVES inputs. It also provides agencies that are less familiar with common practice in developing emissions model inputs with information on how such inputs can be properly developed from commonly available data sources. Finally, it provides some sample data sets that agencies may be able to use to more closely represent local conditions than the data embedded in MOVES, with minimal additional effort on their part. 4-5