Guide to Truck Activity Data for Emissions Modeling (2019)

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.

1 1.1 Purpose and Motivation This document is a guide for transportation practitioners on methods, procedures, and data sets needed to capture truck activity, vehicle characteristics, and operations to assist in estimating and forecasting criteria pollutants, air toxics, and greenhouse gas (GHG) emissions from goods and services movement. The guide was prepared under NCHRP Project 08-101. The guide is primarily intended to support practitioners preparing inputs to the EPA’s Motor Vehicle Emis- sions Simulator (MOVES) model, the accepted emissions model for most of the United States. However, the information in the guide may be useful for users of EMFAC in California and for those using other methods to estimate truck emissions. The guide simply presents options for input preparation and should not be construed as a required or recommended practice by either the EPA or the U.S. DOT. A separate final report prepared for this project describes the research project and includes additional information, including a literature review, emissions sensitivity analysis, and considerations for future research and model development (available at http:// apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=3860). This guide enumerates various sources of truck data and how they can be obtained and used to support emissions modeling. Case studies highlight new and emerging data sources, or new uses of existing data, to develop information that is not yet widely available to most practitio- ners. Material from the case studies forms much of the main body of the guidance document; the complete case studies are available on the TRB website as NCHRP Web-Only Document 261 at http://www.trb.org/Main/Blurbs/178906.aspx. The guide serves as a supplement to Volumes 1 and 2 of NCHRP Web-Only Document 210: Input Guidelines for Motor Vehicle Emissions Simulator Model: (Porter et al., 2014a, 2014b), which provide guidance on developing local inputs to the MOVES model. Volume 1 of NCHRP Web-Only Document 210 (Porter et al., 2014a) focuses on inputs at a regional (or county) level, and Volume 2 (Porter et al., 2014b) focuses on inputs for a project level. Volumes 1 and 2 of NCHRP Web-Only Document 210 are supplemented with several data sets and online tools. The guidance in these volumes of NCHRP Web-Only Document 210 covers all vehicle types, but is not specific to trucks. The current guide supplements that handbook by describing the use of various data sources to obtain truck-specific inputs. The MOVES model is an emission factor model that requires inputs using a variety of local data, such as vehicle populations, age distributions of the vehicle fleet, vehicle miles of travel (VMT), distributions of VMT by time period and road type, vehicle speeds, and idling activity. MOVES provides considerable opportunity for detail on inputs by vehicle type. Such detail can greatly enhance the accuracy of emissions estimates produced by MOVES. However, truck- specific local data are rarely developed for most inputs, due in part to a lack of readily available data, as well as limited resources to mine available data sources. S E C T I O N 1 Introduction

2 Guide to Truck Activity Data for Emissions Modeling Although, in the past, truck-specific data often have been hard to come by, this is chang- ing with the advent of emerging data sources—such as mobile communications devices and on-board telematics. This guide explores a variety of emerging data sources and provides exam- ples of how they can be used to improve truck data inputs to MOVES. The guide also considers some sources that have been in existence, but could be more broadly applied. The guidebook and case studies also highlight some limitations of current data sources. The landscape of transporta- tion data is changing rapidly, and some of the applications investigated here are still exploratory in nature. Within a few years of this guide’s publication, it may be possible to do new things not considered herein. The primary audience for this guide is practitioners using MOVES. However, research con- ducted for the development of this guide also is informing the evolution of the MOVES model. MOVES inputs may change in the future to take advantage of the information that is becoming available, while also recognizing the limitations of what we know. This guide was prepared at a time when MOVES2014 was the latest model release; changes to the next release of MOVES were still under discussion at the time of publication. Users of future MOVES releases should review the model documentation carefully to determine whether input requirements have changed. This guidebook contains data that in some cases differ from the default data embedded in MOVES. The findings are exploratory in nature, and further research is needed to verify the rep- resentativeness of the data. The user is cautioned not to use any data contained in this guidebook or accompanying files for State Implementation Plan (SIP) or conformity analysis without first consulting with the EPA. Users also should consult with the EPA if they are using their own local data obtained from any of the data sources described in this guidebook to develop local inputs for SIP or conformity analysis. 1.2 MOVES Inputs Considered Table 1.1 indicates which MOVES inputs can be provided by vehicle or source type;1 and of those, which are covered in this guide. The activities undertaken for this guide were prioritized to be those that added the greatest value compared to the information and methods already presented in Volumes 1 and 2 of NCHRP Web-Only Document 210 (Porter et al., 2014a, 2014b); and as a result, some inputs are not covered here. The reader is referred to Volumes 1 and 2 of NCHRP Web-Only Document 210 for data sources and procedures for developing these other inputs. Note that the starts and hotelling importer options for project-scale analysis were new to MOVES2014 and were not discussed in Volumes 1 and 2 of NCHRP Web-Only Document 210, which were written in reference to MOVES2010. Table 1.2 provides suggestions for when practitioners might want to focus their resources on improving local inputs for trucks versus when existing MOVES default data, or data that are likely to be available from the state, may be adequate. These suggestions consider both the sensi- tivity of emissions to the input and the cost/level of effort involved in gathering and processing local data (see Footnote 1 of Table 1.2 for benchmarks). In general, truck-specific data gathering may be most warranted under the following circumstances: • For project-scale analysis, when the project being analyzed includes a significant amount of truck operation. For example: – New construction, expansion, access improvements, or operational improvements at a port, warehousing/distribution center, intermodal facility, or border crossing. – A road with a high volume and/or percentage of truck traffic. 1 “Vehicle type” refers to five vehicle types (as defined in MOVES2014) that correspond with FHWA Performance Monitoring System vehicle classes. “Source type” refers to 13 vehicle/use types defined in MOVES that are subsets of these vehicle type classifications.

Introduction 3 (continued on next page) MOVES Input MOVES Input Table Description Reference Project-Scale Inputs Age distribution by vehicle class sourceTypeAgeDistribution Fraction of vehicle population by age (years) for 13 source types. Section 3.0 Source (vehicle) type fraction sourceTypeHourFraction Fraction of vehicle hours of travel by 13 source types, for each link. Not covered—see NCHRP Web-Only Document 210, Volumes 1 and 2a Traffic volume Link Total traffic volume for analysis period, for each link. Not covered—see NCHRP Web-Only Document 210, Volumes 1 and 2 Average speed Link Average speed, by link. Section 4.0 Operating mode distributions opModeDistribution Percent of time spent in each operating mode bin, by link (applies to starts as well). Section 5.0 Drive cycles driveScheduleSecondLink Speed and acceleration by 1- second time slice, by link. Section 5.0 Vehicle population offNetwork Vehicle population using facility during the analysis period by 13 source types. Not covered—see NCHRP Web-Only Document 210, Volumes 1 and 2 Start fraction Starts importer Fraction of vehicles that start during the analysis period. Section 6.0 Starts per vehicle Data importer (generic) Starts per vehicle, by hour of day and day of week. An alternative input to those in starts importer, but data may support this input more readily than start importer inputs. Section 6.0 Parked fraction offNetwork Fraction of time vehicles spend parked in an hour. Not covered—see NCHRP Web-Only Document 210, Volumes 1 and 2 Extended idle fraction offNetwork Fraction of vehicles idling during project period (MOVES only supports combination long-haul truck idling). Section 7.0 Hotelling operating mode Hotelling importer Fraction of long-haul combination trucks (by model year) that are in extended idle, auxiliary power unit (APU), battery, or engine-off mode. Section 7.0 County-Scale Inputs Age distribution by vehicle class sourceTypeAgeDistribution Fraction of vehicle population by age (years) for 13 source types. Section 3.0 Source (vehicle) type population sourceTypeYear Fraction of vehicle population by 13 source types. Not covered—see NCHRP Web-Only Document 210, Volumes 1 and 2 VMT by vehicle class HPMSVTypeYear Fraction of total VMT for 5 vehicle types. Not covered—see NCHRP Web-Only Document 210, Volumes 1 and 2 Table 1.1. MOVES inputs that can be provided for trucks.

4 Guide to Truck Activity Data for Emissions Modeling Road type distribution roadTypeDistribution Fraction of VMT occurring by 5 road types, for 13 source types. Not covered—see NCHRP Web-Only Document 210, Volumes 1 and 2 Ramp fractionb roadType Fraction of vehicle hours of travel occurring on freeway ramps (restricted-access roads). Not covered—see NCHRP Web-Only Document 210, Volumes 1 and 2 Average speed distribution avgSpeedDistribution Fraction of VMT by 16 speed bins, for 13 source types, 4 road types, 24 hours of day, and weekday versus weekend. Section 4.0 Starts per day Starts importer Number of total starts per day in a given zone (county). Section 6.0 Start allocations Starts importer Fractions that allocate total starts to source type, day and hour (three separate tables). Section 6.0 Starts per vehicle Data Importer (generic) Starts per vehicle, by hour of day and day of week. Data may support this input more readily than start importer inputs. Section 6.0 Start operating mode distributions (OMDs) Starts importer Distribution of soak periods preceding starts by hour of day and day of week. Section 6.0 Hotelling hours Hotelling importer Total hours spent in each hotelling operating mode by source type (long-haul combo trucks), hour, day, month, and vehicle age. Section 7.0 Hotelling activity rate (hours per rural VMT) HotellingCalendarYear As alternative to above, specifies the rate of hotelling hours per rural VMT to allow another means of alternate input for total hours. Section 7.0 Hotelling operating mode Hotelling importer Fraction of long-haul combination trucks (by model year) that are in extended idle mode, APU mode, or engine-off mode. Section 7.0 aPorter et al., 2014a, 2014b. bRamp fraction is expected to be removed as an input option in the next release of MOVES. Temporal adjustments monthVMTFraction dayVMTFraction hourVMTFraction Fraction of VMT occurring by month of year or 13 source types; by hour of day and weekday versus weekend for 5 road types and 13 source types. Not covered—see NCHRP Web-Only Document 210, Volumes 1 and 2 MOVES Input MOVES Input Table Description Reference Table 1.1. (Continued). • For county-scale analysis, to generally improve truck data inputs for inventory development, regional conformity, mobile source air toxics, or GHG analysis, especially if the county includes a high volume of truck traffic and/or truck patterns that may differ substantially from typical patterns (e.g., a unique industry mix or border zone). Note that many of the inputs show a wide range of sensitivity, depending upon the pollutant being analyzed and the assumptions made regarding a shift in input values. The reader is referred to Section 6.0 of the contractor’s final report for NCHRP Project 08-101 (available at https://apps. trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=3860) for more details on sensitivity analysis findings from a variety of studies.

Introduction 5 MOVES Input Sensitivitya Cost/Level of Effortb When Might Local Truck-Specific Data Collection Be Warranted Project-Scale Inputs Age distribution by vehicle class Moderate to Very High Moderate to High Local truck population that is likely to differ greatly in age versus state or national fleets (e.g., drayage fleets serving a port). Source (vehicle) type fraction Moderate to Very High Modest to Moderate Classification data generally advised, unless the project location appears to be representative of average state traffic mix. Traffic volume (VMT fraction by class) Modest to Very High Modest Traffic volumes by vehicle class are a necessity. Average speed Moderate to Very High Modest to Moderate Conditions where speeds might differ significantly for trucks (e.g., hilly/mountainous topography). OMDs or drive cycles Modest to Very High High Major project where a simulation model is already being developed. Could apply representative drive cycle or OMD from library at low level of effort. Source type population (off-network) High Moderate Evaluation of major truck-generating off-network facility. Start fractions, start OMDs, and parked fraction Modest Moderate to High Atypical truck activity patterns (e.g., warehousing center of port); high fractions of gasoline-powered trucks; if emissions during a specific time period are of concern. Hotelling (total daily hours, operating mode, distribution) High to Very High High Hotelling activity expected at or in proximity to project. County-Scale Inputs Age distribution by vehicle class Moderate to High Moderate to High Local truck population that is likely to differ greatly in age versus state or national fleets. Source (vehicle) type population Moderate to Very High Modest To Moderate Population by source type is generally a necessity. VMT by vehicle class Modest to Very High Modest VMT by vehicle class is generally a necessity. Temporal adjustments Modest Modest Mainly if emissions during a specific time period are of concern (e.g., for hot-spot analysis or regional air quality modeling). Road type distribution Modest to Moderate Modest Generally if already available (e.g., from a travel demand model or state traffic count system). Average speed distribution Modest to Very High Modest to Moderate Conditions where speeds might differ significantly for trucks (e.g., hilly/mountainous topography). Start activity (number of starts, soak distribution) Modest Moderate to High Atypical truck activity patterns due to industry mix or presence of major port or border crossing; high fractions of gasoline-powered trucks; if emissions during a specific time period are of concern. Hotelling activity (total hours, temporal distribution) High to Very High High High levels of truck hotelling expected due to major truck routes, truck stop/rest area facilities, etc. aSensitivity benchmarks: Modest = <5 percent; Moderate = 5 to 15 percent; High = 15 to 50 percent; Very High = >50 percent. Source data are presented in the contractor’s final report for NCHRP Project 08-101. bCost/level of effort benchmarks: Modest = existing/readily available data source; Moderate = data purchase or collection + analysis in range of ~$5,000 to $15,000; High = data purchase or collection + analysis in range of >$15,000. Table 1.2. Suggestions on focusing effort to improve local truck data inputs.

6 Guide to Truck Activity Data for Emissions Modeling 1.3 Outline of Guide Section 2.0 of this guide presents an overview of potential data sources on truck activity and fleet characteristics, including sources that were used to develop case studies, and those that were investigated, but not used for various reasons. Sections 3.0 through 7.0 provide information on data sources and options by MOVES input. 1.4 Related Resources Appendices A through G of this guide are provided as NCHRP Web-Only Document 261 (available at http://www.trb.org/Main/Blurbs/178906.aspx) and include seven case studies that document analysis of data and serve as the basis for much of the guide. The contractor’s final report for NCHRP Project 08-101 provides more information on how the research to develop this guide was conducted (available at https://apps.trb.org/cmsfeed/ TRBNetProjectDisplay.asp?ProjectID=3860). Data from the case studies are available in three MS Excel files that are available on the NCHRP Research Report 909 web page (http://www.trb.org/Main/Blurbs/178921.aspx): • NCHRP08-101_Data_CS1+7_DriveCyc-OpMode.xlsx (Case Studies #1 and #7). • NCHRP08-101_Data_CS2_AgeDist.xlsx (Case Study #2). • NCHRP08-101_Data_CS3+4+6_Starts-Idle-Hotelling.xlsx (Case Studies #3, #4, and #6).