Section 4.0 Emission Inventories
This chapter begins with a review of the annual 2002 emission inventory, even though for purposes of an ozone implementation plan the more appropriate measure is an emission rate based on a "typical" ozone season day (OSD). Ozone season emissions are presented in the second section. A third section is devoted to future year projections. Both OSD and future projections use these 2002 annual estimates as the baseline. OSD emissions are adjusted for the various types of emission source sectors, based on their activity level during the summer ozone season. The source sectors enumerated in this chapter are divided into point, EGU, area, non-road mobile, on-road mobile, and biogenic sources.
4.2 Summary of 2002 Baseline Annual Emissions
The fundamental unit for the inventory of each source sector and contaminant is an annual tons per year emissions level reported on a "by-county" basis. The by-county and total statewide inventory for CO, NOx, and VOCs are detailed in Appendix B. The statewide totals are summarized in Table 4.1. Tons per year are reported to the nearest ton, except where there is less than one (1) ton. Those instances are reported in tenths to distinguish them from categories where there are no (or zero) emissions.
|2002 Annual||Tons per Year||Percent of All Sectors|
For the Poughkeepsie area, the summary is tabulated below as Table 4.2.
|2002 Annual||Tons per Year||Percent of All Sectors|
For both the Statewide (Table 4.1) and Poughkeepsie (Table 4.2) annual inventories, the percent share of each sector for each of the contaminants is shown in the right-hand portion of the tables above.
The by-county and by-sector details (presented as Appendix B) are also available in spreadsheets (MS Excel).
4.2.1 Point Inventory Methodology
New York State has an integrated emissions, permitting, compliance, and fee billing computer system identified as New York's Air Facility System (AFS). The emissions module of AFS is a database which contains detailed facility and emissions information for all of the major (Title V) sources within New York State. This database is used to generate annual emission statement forms which are sent out to the State's major facilities each year. Emission statements survey the type and amount of fuel consumed (combustion sources), throughput rates (non-combustion processes), average hours of operation, percent operation by season, control descriptions/efficiencies, and estimates of actual emissions for each regulated contaminant. The 2002 emissions from point sources were obtained directly from Title V major sources via the required emission statement surveys. These data from the major sources were further subdivided into EGU and (other) point source sectors.
All of this data was submitted to MARAMA / MANE-VU for additional quality assurance (QA) and for their use in preparing the projection inventories. MARAMA, the Mid-Atlantic Regional Air Management Association, Inc. is a voluntary, non-profit association of ten state and local air pollution control agencies. MARAMA is cooperating with the Northeast States for Coordinated Air Use Management (NESCAUM) and the Ozone Transport Commission (OTC) to provide staff support to the Mid-Atlantic and Northeast Visibility Union (MANE-VU). The inventory summary work described in this chapter was prepared by MANE-VU as a coordinated effort among the states to develop a consistent inventory throughout the region with the most efficient process. The MANE-VU methodology and results can be found in the document "Development of Emission Projections for 2009, 2012, and 2018 for NonEGU Point, Area, and Nonroad Sources in the MANE-VU Region," February 2007.
4.2.2 Area Inventory Methodology
Area sources are defined and calculated in accordance with the descriptions and methodologies in the EPA Emissions Inventory Improvement Program (EIIP) Volume III - Area Source series, and the Air Toxic Emission Protocol for the Great Lakes States. Area sources collectively represent individual stationary sources that have not been inventoried as specific point sources. These individual sources treated collectively as area sources are typically too small, numerous, or difficult to inventory using the methods for the other classes of sources. Area sources represent a collection of emission points for a specific geographic area, most commonly at the county level; however, any geographic area can be used to present area sources. Facilities and emission points are grouped together with other like sources into area source categories. These area source categories are combined in such a way that emissions can be estimated for an entire category using one methodology. This methodology normally requires a step to exclude the emissions from sources that have already been accounted for as point sources. The area source categories must be defined in such a way to avoid overlap or duplication with point, mobile or biogenic emissions sources.
New York has applied the methodologies as identified in EIIP and/or the Air Toxic Emission Protocol for the Great Lakes States, including appropriate 2002 actual activity data to develop the 2002 periodic area source inventory. The area sources are broken down according to Area Source Codes (ASC). Details of area source methodologies are provided as Appendix D.
All of the area source data was submitted to MANE-VU for additional QA and for its use in preparing the projection inventories. The MANE-VU methodology and results can be found in the document "Development of Emission Projections for 2009, 2012, and 2018 for Non-EGU Point, Area, and Nonroad Sources in the MANE-VU Region," February 2007.
4.2.3 On-Road Inventory Methodology
The on-road component of the 2002 base year inventory includes an estimate of emissions from all motorized vehicles operated on public roadways. All on-road mobile emissions were estimated using EPA's MOBILE6 emission model and individual inputs for each of the 62 counties in the state. These inputs include varying temperature, traffic, and/or air quality programs. "Base-year" inventory inputs were derived from 2002 data, where applicable, and reflect the programs and controls that were in effect in 2002. In order to yield more accurate annual inventories the modeling was done using specific inputs for each month. Brief descriptions of these input types are provided below.
A new 2002 Daily Vehicle Miles Traveled (DVMT) inventory was constructed by the New York State Department of Transportation (NYSDOT) to provide DVMT estimates by county, geographic component (urban, small urban, and rural) and functional class. This resulting VMT by county and by functional class is then multiplied by a seasonal adjustment factor to account for seasonal differences. This seasonal adjustment factor is also supplied by the NYSDOT. For ozone season day, the seasonal adjustment factor is a 10 year average of "summer" seasonal adjustment factors supplied by NYSDOT.
The vehicle mix for each of the 11 NYSDOT regions in New York State is used to produce VMT by vehicle type. There are 28 fuel and weight categories employed by MOBILE6. The main objective is to create a separate, distinct (where justified) vehicle mix for each of the twelve roadway types in the Federal Highway Administration (FHWA) classification scheme.
The vehicle age distributions used in MOBILE6 are obtained from the New York State Department of Motor Vehicles (NYSDMV) registration data for the current year at the beginning of each July. Each record is sorted into the 28 vehicle types by county. The 2002 registration distribution was used for 2002 inventories. Diesel fractions are obtained at the same time as the registration distributions.
EPA default Mileage Accumulation Rates for all vehicle types were taken from EPA's Fleet Characterization Data for Mobile6.1.
NYSDOT created vehicle use profiles similar to those used as inputs to California's EMFAC model. One of these inputs is the percent of vehicle trips in each hour; these values also equate to the number of starts per hour.
Hourly temperatures were obtained from the National Oceanic and Atmospheric Administration for New York and vicinity. Each area of the State was then matched to a NWS station. The Department uses hourly values to more accurately model hourly emissions. Monthly average hourly temperatures were created from recorded hourly temperature data for all of 2002 for each of the weather stations used for ozone temperatures.
The relative humidity data for modeling of ozone exceedance days were calculated from hourly airport observations that the Department obtained from the National Climatic Data Center. Dewpoint observations for the same dates and locations that were used in temperature calculations were also used to determine hourly relative humidity values. The Department uses actual recorded hourly values to more accurately model hourly emissions. In modeling annual emissions an average daily absolute humidity value was calculated for each month of the year.
The Planning Division of NYSDOT developed speed estimates for air quality modeling in 1994. Speeds were developed for 15 areas, some as small as a single county, throughout the state along with each of the 12 possible functional classes and four time periods. When modeling these speeds in MOBILE6, the AVERAGE SPEED command was not used because it can only model a single speed for the entire day. The SPEED VMT command allows the modeling of different hourly speeds and was therefore chosen as the input format for New York State speeds.
The Stage II Refueling program began for the NYMA area in 1989. However, refueling emissions are not included as part of the mobile source inventory; rather, they are calculated separately and included in the area source component of the inventory.
The Mobile6 Anti-Tampering Program command is used to specify the programs in effect in New York State. The Anti-Tampering Program is applicable statewide to all gasoline-powered vehicles during the annual safety/emissions inspections. An additional gas cap pressure check was added in 1999 The Mobile 6 I/M Program command is used to specify the Inspection/Maintenance (I/M) programs in effect in New York State.
The LEV 2 phase-in schedules were created using a spreadsheet to solve for the NMOG standard for each model year using the various motor vehicle certification standards, or "bins." The LEV 2 program is based on each vehicle meeting an NMOG standard for each model year. This standard can be met using any combination of LEV 2 bins the manufacturer desires.
4.2.4 Non-Road Inventory Methodology
The non-road component of the 2002 base year inventory includes an estimate of emissions from motorized vehicles and equipment that are not typically operated on public roadways. Emissions estimates for non-road mobile sources were estimated using four separate methodologies. EPA's Non-Road Model is used for a number of non-road emission categories while airport, commercial marine vessel and locomotive emissions are calculated separately outside of the model. In addition, all 62 counties are modeled separately and the state is separated into two areas to account for the federally mandated RFG program in place in the 10-county NYMA.
Emissions from 2-stroke gasoline, 4-stroke gasoline, liquefied petroleum gas, compressed natural gas and diesel fueled non-road vehicles as well as emissions from recreational marine vessels were estimated using the U.S. EPA Non-Road Model Version 2005. The software was finalized for use in SIP development on June 12, 2006. Using the Non-Road Model, emissions from New York were estimated for each individual county for each month of the year. To account for temperature and fuels differences across the state, county-specific temperature and fuels blend data for each month of the year were input into the model.
For 2-stroke gasoline, 4-stroke gasoline, liquefied petroleum gas, compressed natural gas and diesel fueled nonroad vehicles as well as emissions from recreational marine vessels; the U.S. EPA Nonroad Model was run on a monthly county-by-county basis. To develop emissions for a typical ozone season day, the emissions for June - August were added together and then divided by 92.
The temperature data for 2002 was acquired from the National Oceanic and Atmospheric Administration which included historical weather data from 33 airport locations across the State of New York as well as surrounding locations. This information was used to develop average high and low temperatures for each month of the year on a county by county basis.
Fuels blend data for 2002 was acquired from the New York State Department of Agriculture and Markets. This data is based on thousands of samples collected across the state from fueling stations and retention areas. These samples are analyzed for many profiles including oxygen content, RVP, and sulfur content. This analysis provides average monthly fuels profiles on a county -by-county basis for use in the model.
Aircraft emissions for New York State are estimated using the Federal Aviation Administration's (FAA) Emission Dispersion Modeling System (EDMS) Version 4. 4. Airport specific landing and take-off data by aircraft type acquired from FAA are used as inputs to the model. EDMS uses this information to estimate emissions from both aircraft and ground service equipment.
Commercial Marine Vessel (CMV) emissions are based upon the CMV emissions report prepared by the Starcrest Consulting Group in conjunction with their work on the New York Harbor Deepening Project. The emissions from Bronx, Kings, Nassau1, New York, Queens, Richmond, Rockland, Suffolk and Westchester counties are based on actual 2002 operational data from an intensive survey performed by Starcrest. The CMV inventory for the rest of the state is based on Radian Corporation's report entitled "1990 Base Year Ozone Precursor Emissions Inventory for New York State: Volume 4: Non-Road Mobile Sources," March 1993 (revised July 1993).
The Starcrest inventory includes a detailed survey of all CMV types, activity and fuel consumption and took several months to complete. This project was undertaken as part of the NYC Harbor Deepening Project to update the baseline inventory and to optimize the offsets that would be utilzed by the Army Corps of Engineers. This updated inventory was performed by Starcrest Consulting under contract to the Port Authority. While the Department would like to use the Starcrest methodology to update the CMV inventory for the rest of the state it would require an intensive effort to survey all of the counties bordering Lake Erie, Niagara River, Lake Ontario, the St. Lawrence Seaway, Lake Champlain, Hudson River, Mohawk River, Erie Canal and both the Long Island Sound and Atlantic Ocean since Suffolk County was not included in the Starcrest inventory. Other bodies of water that may have CMV traffic are the Finger Lakes, Oneida Lake, Lake George and possibly some other rivers.
The Department is aware that there is more recent EPA guidance regarding CMV inventory development but the methodology is very different than the work completed by Starcrest. The Department also believes that this guidance is far less detailed and would not result in an improvement to the methodology established in the development of the 1990 inventory.
The locomotive emissions inventory is based upon a report developed under contract to the New York State Energy Research and Development Authority (NYSERDA). The report is entitled "NYSERDA CLEAN DIESEL TECHNOLOGY: NON-ROAD FIELD DEMONSTRATION PROGRAM; Development of the 2002 Locomotive Survey & Inventory for New York State." The locomotive inventory is based upon a survey conducted of the national, regional, and local freight railroads, as well as passenger and commuter rail lines operating in New York State. Information collected in the survey was used in development of the emissions inventory.
For aircraft, commercial marine and locomotives the 2002 annual base year inventories were first grown to annual projection year inventories by the method described above. For all three categories, there is no documentation that supports using any seasonal adjustment factors to develop daily emissions. Therefore, ozone season day emissions were calculated by dividing the annual emissions by 365.
4.2.5 Biogenic Inventory Methodology
Biogenic emissions were calculated using Biogenic Emissions Inventory System (BEIS) v3.1.2. Daily values were totaled for each county to yield annual numbers.
4.3 Summary of 2002 Ozone Season Day (OSD) Emissions
For this portion of the inventory, the fundamental unit for the inventory of each source sector is tons per OSD. Similar to the annual inventory, the OSD inventory is reported on a "by-county" basis for the various source sectors. The by-county and total statewide inventory for CO, NOx, and VOCs are detailed in Appendix C. The statewide numbers are summarized in Table 4.3. They are reported to the nearest hundredth ton. In some cases, where there is less than one-hundredth (0.01) ton, emissions are reported in thousandths to distinguish them from categories where there are no (or zero) emissions.
|2002 OSD||Tons per Day||Percent of All Sectors|
|Point (non- EGU)||227.27||140.85||59.46||1.8%||7.8%||1.0%|
For the Poughkeepsie area, the summary is tabulated as Table 4.4.
|2002 OSD||Tons per Day||Percent of All Sectors|
|Point (non- EGU)||2.50||2.00||3.49||0.4%||2.1%||1.1%|
The percent share of each sector for each of the contaminants on a Statewide basis is shown in the rightmost columns in Table 4.3 above, and for the Poughkeepsie area, on the right side in Table 4.4 above.
4.3.1 Methodological Details Used to Compute Ozone Season Day from the Annual Estimates
OSD emission inventories are derived from annual inventories and are estimated by adjustments to reflect the relative difference of emission patterns during the ozone season when compared to cooler months. Depending upon source sector activity levels, some source categories are more or less likely to have emissions during an OSD. For example, an OSD is less likely to have emissions related to space heating and more likely to have emissions related to air conditioning or painting. Many categories have relatively constant emissions throughout the year (e.g., consumer products - deodorant, house cleaning products, etc.) OSD estimates attempt to characterize those seasonal differences to more accurately reflect emissions during the summer season.
The ORMS (on-road mobile source) sector uses a seasonal adjustment factor to adjust DOT's annual average daily vehicle miles traveled (AADVMT). This seasonal adjustment factor is an average of the June, July, and August monthly factors used by NYSDOT.
Hourly temperatures were obtained from the National Oceanic and Atmospheric Administration for New York and vicinity. Each area of the State was then matched to a NWS station. The Department uses hourly values to more accurately model hourly emissions.
The relative humidity data for modeling of ozone exceedance days were calculated from hourly airport observations that the Department obtained from the National Climatic Data Center. Dewpoint observations for the same dates and locations that were used in temperature calculations were also used to determine hourly relative humidity values. The Department uses actual recorded hourly values to more accurately model hourly emissions.
These inputs are then used in MOBILE6.2 to produce an emission factor for each vehicle and road type combination for all 62 counties. The resultant emission factor in grams/mile is multiplied by daily VMT, including seasonal adjustment, to determine daily ozone season emissions.
For NRMS (non-road mobile sources) the following methodologies were used:
1. For 2-stroke gasoline, 4-stroke gasoline, liquefied petroleum gas, compressed natural gas and diesel fueled non-road vehicles as well as emissions from recreational marine vessels; the EPA Non-Road Model was run on a monthly county-by-county basis. To develop emissions for a typical OSD, the emissions for June through August were added together and then divided by 92.
2. For aircraft, commercial marine vessels and locomotives the 2002 annual base year inventories were first grown to annual projection year inventories by the method described above. For all three categories, there is no documentation that supports using any seasonal adjustment factors to develop daily emissions. Therefore, OSD emissions were calculated by dividing the annual emissions by 365.
Area source (non-point) sector OSD emissions are calculated based upon the area source category. For example, for consumer products, the annual emissions are simply divided by 365 because consumer products are generally used uniformly throughout the year. For dry cleaning, the emissions are assumed to be consistent throughout the year, but are assumed to be five day per week emissions, so annual emissions for this category are divided by 260 (5*52) to estimate OSD emissions. For AIM coatings, activity is higher during the summer, so based upon EPA guidance, an adjustment factor of 1.3 is applied during the summer - annual emissions are multiplied by 1.3 and divided by 365 to estimate OSD emissions.
Point source sector OSD emissions are calculated from the operational information provided in the emission statement forms. This information includes the process throughput and a breakdown of operation by season, including the number of days the process was in operation during that season.
For biogenics, technical staff used statewide annual by-day BEIS output to calculate what they have come to call "biogenic OSD expansion factors". These are subsequently applied to the "annual by-county" estimates to generate OSD tonnages. The expansion factors use BEIS model statewide daily values from June, July, and August to compute a representative ton per OSD. Details of this series of computations are available upon request.
4.4 Summary of Future Year Emissions
For the Poughkeepsie area, the future years of interest are 2008, and 2009. The inventories for those years are presented in Tables 4.5, and 4.6.
|2008 OSD||Tons per Day||Percent of All Sectors|
|Point (non- EGU)||2.78||2.51||4.46||0.6%||3.2%||1.6%|
|2009 OSD||Tons per Day||Percent of All Sectors|
4.4.1 Projection Methodologies for Point, EGU, and Area Sources
The 2002 non-EGU point and area source emissions inventories were projected using the growth factors in tables provided by MANE-VU. The emissions used for projections were interpolated for the years 2005, 2008, and 2011. The MANE-VU methodology and results can be found in the document "Development of Emission Projections for 2009, 2012, and 2018 for Non-EGU Point, Area, and Non-Road Sources in the MANE-VU Region," February 2007.
For EGU point sources, EPA has recommended the use of the IPM model to project EGU emissions. MANE-VU followed this recommendation, so the MANE-VU projections for point sources used IPM to estimate EGU emissions. When the IPM modeled emissions were compared to the actual 2005 emissions for New York, or when IPM modeled emissions were compared to the permit applications that the Department has received for new EGUs, it became obvious that, for New York, the IPM projected emissions were not realistic (for example, in NYCMA, IPM projected more than a 70 percent reduction in NOx emissions from EGUs by 2009 with the generation - and associated emissions - moving further upstate). This re-siting of facilities by IPM and hence the movement of emissions does not accurately reflect the reality of the constraints of the electrical grid in New York State, nor does it reflect the realities of siting new power plants in New York. In order to present a more realistic projection of EGU emissions for New York, it is assumed that the 2005 actual EGU emissions will represent the EGU emissions for the future years. 2005 is the most recent data available. The trend in recent years for EGUs has been decreasing emissions statewide (25 percent NOx reduction between 2002 and 2005). Although it is forecasted that generation will increase in New York in future years, emissions are not expected to increase due to the Clean Air Interstate Rule (CAIR) which establishes NOx and SO2 emission caps. The only exception to assuming that the 2005 actual EGU emissions represent future year EGU emissions is where there is a consent agreement that limits future year emissions for a facility. In that case, the future year emissions for that specific facility have been reduced to meet the limits contained in the agreement.
Sample calculations for point and area source growth and control are provided in Appendix J.
4.4.2 On-Road Projection Methodology
New York State is modeled by using individual inputs for each of the 62 counties. Each county receives varying temperature, traffic, and/or air quality programs. The mobile source projection inventory was developed by using Mobile6 emission factors and vehicle miles traveled (VMT) projections for each future inventory year prepared by the New York State Department of Transportation (NYSDOT). This projection uses linear regression of Highway Performance Monitoring System (HPMS) historical data for forecasting VMT. These projections employed HPMS data from 1981 to 2002.
Mobile 6.2 is then run to produce emission factors for each vehicle and road type combination for all 62 counties. The resultant emission factor in grams/mile is multiplied by daily VMT, including seasonal adjustment, to determine daily emissions.
4.4.3 Non-Road Projection Methodology
The U.S. EPA Nonroad Model Version 2005, was used to develop future year nonroad emissions projections for 2-stroke gasoline, 4-stroke gasoline, liquefied petroleum gas, compressed natural gas and diesel fueled nonroad vehicles as well as emissions from recreational marine vessels. When completing future year projections, the model incorporates emissions effects that result from both anticipated changes in equipment activity as well as deterioration of equipment. The model also accounts for expected turnover of old equipment. In addition, the following EPA nonroad emission control programs are built into the model:
- New Phase 2 Standards for Small Spark-Ignition Non-handheld Engines (March 1999) which covers NOx and HC reductions from mowers, edgers, lawn tractors, and other non-hand held gasoline equipment.
- Final Phase 2 Standards for Small Spark-Ignition Handheld Engines (March 2000) which covers NOx and HC reductions from trimmers, leaf blowers, chain saws, and other handheld gasoline equipment.
- Emission Standards for New Nonroad Engines (September 2002) which covers NOx, HC and CO from the following new engines and vehicles:
- Large Industrial Spark-Ignition Engines (forklifts, electric generators, airport baggage, etc.)
- Recreational Vehicles (snowmobiles, dirt-bikes, ATVs)
- Recreational Diesel Marine Engines (for use in yachts and cruisers)
- Clean Air Nonroad Diesel Rule (May 2004) which covers NOx, PM and SOx emissions from diesel engines used in most construction, agricultural, industrial and airport equipment
In addition, this rule includes and requires a 99 percent reduction in diesel sulfur by 2010.
Future year nonroad emissions projections for the aircraft, commercial marine and locomotives categories were calculated using the growth factors developed for the MANE-VU Emissions Projections Technical Support Document. These projections were developed using combined growth and control factors developed from emission projections for U.S. EPA's Clean Air Interstate Rule (CAIR). The control programs in place that were used to develop the growth factors were:
- Adopted Aircraft Engine Emissions Standards (April 1997) which reduces NOx and CO from new aircraft engines
- Final Emissions Standards for Locomotives (December 1997) which reduces NOx, HC, CO and PM from new and remanufactured diesel powered locomotive engines. This rule requires a reduction in diesel sulfur which will result in a reduction in SOx.
- Emission Standards for New Commerical Marine Diesel Engines (November 1999) which reduces NOx and PM from diesel marine engines over 37 kW. This rule requires a reduction in diesel sulfur which will result in a reduction in SOx.
The following regulations were not built into the growth factors.
November 2005 - New Emission Standards for New Commercial Aircraft Engines
March 2007 - EPA Proposal for More Stringent Emissions Standards for Locomotives and Marine Compression-Ignition Engines
4.4.4 Biogenic Future Year Emissions
Biogenic emissions levels were maintained at the 2002 levels for all future years.
1 The update to the Nassau and Suffolk CMV inventory only pertains to that portion included in the New York Harbor study completed by Starcrest. Only portions of these counties affected by the New York City Harbor Deepening Project are included. The remaining portions of Nassau and Suffolk counties are based upon emissions from the 1990 Radian Report grown to reflect 2002 vessel activity.