A EPA
United States
Environmental Protection
Agency
NATIONAL PORT STRATEGY
ASSESSMENT: Reducing Air
Pollution and Greenhouse Gases
at U.S. Ports
Appendices
Office of Transportation Air Quality
EPA-420-R-16-011app
September 2016
L r _
V.
-------
Note: This document contains the appendices of the
National Port Strategy Assessment: Reducing Air Pollution
and Greenhouse Gases at U.S. Ports. The full report can be
accessed at: https://www.epa.gov/ports-initiative.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
-------
List of Appendices to National Assessment of Port Strategies:
Reducing Air Pollution and Greenhouse Gases at U.S. Ports
Appendix A. Baseline Emission Inventory Methodology 1
Appendix B. Business as Usual Emission Inventory Methodology 48
Appendix C. Analysis of Emission Reduction Scenarios 72
Appendix D. Stratified Summary of Results 115
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports i
-------
Appendix A: Baseline Emission Inventory Methodology
Appendix A. Baseline Emission Inventory Methodology
A.l. Introduction
This assessment included the development of representative, national scale inventories for the baseline
and Business as Usual (BAU) cases for different pollutants and years, followed by the analysis of various
strategies to reduce port-related mobile source emissions. This appendix details the methodology used
to develop the baseline emission inventories for the calendar year 2011.
Separate inventories for various pollutants were developed for the drayage trucks, rail, cargo handling
equipment (CHE), harbor craft, and ocean going vessels (OGV) sectors. The following pollutants were
included in these inventories: nitrogen oxides (NOx), fine particulate matter (PM2.5), volatile organic
compounds (VOCs), sulfur dioxide (S02), carbon dioxide (C02), black carbon (BC), acetaldehyde,
benzene, and formaldehyde. Note that the selected air toxics (acetaldehyde, benzene, and
formaldehyde) were only analyzed for the non-OGV sectors and S02 was only analyzed for the OGV
sector. In general, inventories were developed for each port analyzed in this assessment using national
scale methodology and data, although port-specific data and adjustments were included where
available and are noted where appropriate in this appendix. This assessment does not provide specific
data for local decision-making at individual ports or specific neighborhoods.
EPA developed this national scale assessment based on estimated emissions from a representative
sample of seaports, which are listed in Table A-l. Several of the ports have publicly available emission
inventories that were used to improve the information applied in this assessment. Ports were chosen to
represent typical deep sea ports in the United States, so it does not include any inland freshwater ports.
However, it is expected that emission reduction strategies would be applicable to other seaports, Great
Lakes and inland river ports, or other freight and passenger facilities with similar mobile source profiles.
It should be noted that this project is not intended to provide port-specific results. EPA did not consult
with the 19 ports before port areas were selected for this assessment. The results from the 19 ports
were combined throughout this assessment to present a national scale picture of how emissions change
between the baseline, future BAU case, and what can be done to further reduce emissions.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
1
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-l. Ports Selected for Assessment
No.
Port
Published Inventory
Available
1
Port of New York and New Jersey
Yes
2
Port of New Orleans
No
3
Port of Miami
No
4
Port of South Louisiana
No
5
Port of Seattle
Yes
6
Port of Baton Rouge
No
7
Port Arthur
Yes
8
Port of Portland, OR
Yes
9
Port of Mobile
No
10
Port of Houston
Yes
11
Port of Baltimore
No*
12
Port of Hampton Roads (Norfolk)
Yes
13
Port of Philadelphia
Yes
14
Port of Charleston
Yes
15
Port of Corpus Christi
Yes
16
Port Tampa Bay
No
17
Port of Savannah
No
18
Port of Coos Bay, OR
No
19
Port of San Juan, PR
No
* The Port of Baltimore published a CHE inventory, but it was not available during the timeframe for
inclusion in this assessment.
While deciding which ports to incorporate in this national scale assessment, EPA selected a
representative sample of ports that was intended to be diverse. EPA considered several factors in
selecting this sample of port areas: the geographic location of a given seaport; the type and size of
different ports; the availability of port emission inventories; and whether or not a port was located in or
adjacent to a nonattainment or maintenance area for the national ambient air quality standards
(NAAQS). The geographic distribution of the ports selected for assessment is shown in Figure A-l.1 The
assessment was also based on other publically available activity data for certain port areas that have
been used for other previous EPA analyses, as noted in throughout this appendix and Appendix B.
1 EPA notes that the Port of Los Angeles and the Port of Long Beach were initially selected for inclusion in this national scale
assessment. However, EPA decided to not include emission estimates for these port areas to avoid biasing the national scale
results. Many of the strategies in this report are already being implemented at these ports, and there are also port-specific
baseline and future emission inventories in place, all of which could potentially impact the final results. EPA consulted on this
decision with the MSTRS Ports Workgroup, which included several ports, government agencies, community groups, and other
policy and technical experts.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
2
-------
Appendix A: Baseline Emission Inventory Methodology
Figure A-l. Geographic Distribution of Selected Ports2
# WA MT ND ...| Al NB __
# MN \ / I PE
SD Wl \ / ME \ NS
• 0R |D WY ™
NE IA , <^c. MA
IL OH PA •
nv United States IN Ma *
CO KS MO „v wv Ut
CA KY V/W
OK AR tn NC
AZ NM cr
MS AL S£
i-v TX i GA*
• •
• *FL
Mexico
Cuba
Puerto Rico
I
The following sections detail how the baseline inventories were developed for each sector.
A.2. Drayage Trucks
The 2011 base year drayage emissions were modeled using EPA's DrayFLEET Model. The truck activity is
based on the total tonnage and TEUs moved at a port allocated by mode to drayage using FHWA's
Freight Analysis Framework (FAF).
A.2.1. Data Sources
Two primary types of activity data were collected for the analysis: tonnage3 and twenty foot equivalent
units4 (TEUs) by port. Activity data from the US Army Corps of Engineers (USAGE) Waterborne
Commerce Statistics on TEUs and tonnage by port were collected for the base year 2011. For
containerized movements, data on the number of TEUs were used to estimate truck drayage activity.
TEUs were translated into truck movements based on estimates of TEUs per container. The USAGE
Waterborne Commerce Statistics includes data on domestic empty containers, but not foreign empty
2 Map data: Google, INEGI.
3 Available at: http://www.navigationdatacenter.us/wcsc/bv portnamesll.html.
4 Available at: http://www.navigationdatacenter.us/db/wcsc/archive/xls/manll/.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
3
-------
Appendix A: Baseline Emission Inventory Methodology
containers. Data on foreign empty containers were collected separately from ports or other industry
sources.5
For non-containerized cargo, the analysis used data on the tonnage of freight originating and
terminating in the port from the ACE Waterborne Commerce Statistics. This cargo was classified as bulk,
liquid, container, or other. The number of truckloads was determined based on the cargo densities and
payload estimates by commodity.
Data from the FAF for 2012 were used to estimate the percentage of containers and tonnage moving by
truck and other modes for each port. FAF identifies the port of export, the domestic mode of
transportation, and the foreign mode. For freight moving via water for the foreign mode of
transportation, exports and imports were combined and the percentage of tonnage moving by truck for
the domestic mode was estimated.
A.2.2. Modeling Approach
The EPA DrayFLEET Model6 was used to estimate emissions generated from all truck movements. TEUs
and TEUs per container were input on the primary inputs page. For non-containerized freight, tons and
average truck cargo weight were input on the secondary inputs page. Gate queues and average marine
terminal transaction times were entered based on data available for each port or default values when
this information was unavailable. The age distribution of drayage trucks came from the MOVES2010b
national default age distribution for combination short-haul trucks.
A 0.5 km boundary was included in this analysis. Drayage emissions for the port and within a 0.5 km
buffer outside the port were modeled separately. This was accomplished by estimating the distance
drayage vehicles travel inside the port and the distance they travel outside the port within a 0.5 km of
the port boundary. A visual inspection of port maps was made to estimate these distances. In the case
of emissions outside of the port, a distance longer than 0.5 km was selected for some ports to account
for route circuitry around the port boundary.
One model scenario was set up to include the total drayage distance, while a second scenario included
only the distance within the port. The difference in emissions estimates between these models was used
to estimate emissions outside of the port boundaries but within the 0.5 km buffer.
The DrayFLEET Model estimates emissions for PM2.5, NOx, HC, CO, and C02. Emissions for the air toxics
formaldehyde, benzene, and acetaldehyde were estimated as a fraction of HC emissions based on diesel
speciation profiles calculated from running MOVES2010b, as shown in Table A-2. These fractions vary
depending on whether or not VOC is controlled (model year 2007 and later). Weighted speciation
5 Available at: http://aapa.files.cms-
plus.com/Statistics/N0RTH%20AMERICAN%20P0RT%20C0NTAINER%20TRAFFIC%202Qll.pdf.
6 Available at: http://www.epa.eov/smartwav/forpartners/documents/dravaee/420bl2065.pdf.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
4
-------
Appendix A: Baseline Emission Inventory Methodology
factors were calculated for 2011 based on the percent of 2007 model year and greater trucks in the
fleet. Approximately 86% of drayage trucks in 2011 were model year 2006 or earlier.
Table A-2. Diesel Truck Air Toxic Speciation Profiles Based on MOVES2010b7
Pollutant
Toxic/VOC without Control
Toxic/VOC with Control
Acetaldehyde
0.035559
0.06934
Benzene
0.007835
0.01291
Formaldehyde
0.078225
0.21744
Black carbon (BC) emissions were estimated as a 77% of PM2.5 emissions, consistent with EPA's Report
to Congress.8
A.3. Rail
The 2011 baseline inventory of rail emissions is based on two primary sources: published port emission
inventories and the 2011 National Emission Inventory (NEI). The inventory for all 19 ports considered
here is drawn from one of these two sources, as described below.
A.3.1. Data Sources
A.3.1.1. Published Port Inventory Data
EPA performed a review of available port inventories and determined that for four ports, the on-
terminal estimates from published rail emission estimates are a better match for this project's scope
than values derived from the NEI. Table A-3 lists these four ports and the information included in this
analysis from each inventory.
7 U.S. Environmental Protection Agency, MOVES2010b: Additional Toxics Added to MOVES. EPA-420-B-12-029a, May 2012, Sec
3.1.1. Available at: http://www.epa.gov/otaq/models/moves/documents/420bl2029a.pdf.
8 U.S. Environmental Protection Agency, Report to Congress on Black Carbon, EPA-450/R-12-001, March 2012, p. 87.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
5
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-3. Rail Emissions from Published Port Inventories Included in Baseline Inventory
Port
Published Inventory
Rail Type Included
Charleston
South Carolina State Ports Authority 2011
Emissions Inventory Update (April 2013)
"On-Terminal" results for Line-Haul and
Switcher Locomotives
Norfolk
Port of Virginia 2011 Comprehensive Air Emissions
Inventory Update (January 2013)
"On-Terminal" results for Line-Haul and
Switcher Locomotives
Seattle
Puget Sound Maritime Air Emissions Inventory,
August 2012 (May 2013)
Port of Seattle estimates of switcher and
Line-Haul Locomotives, with the
understanding that the line-haul inventory
only includes emissions from "near
port/adjacent rail yards."
New York / New
Jersey
Port of New York and New Jersey Port Commerce
Department 2012 Multi-Facility Emissions
Inventory (August 2014)
Switcher emissions only.
A.3.1.2. NEI Data
For the 15 ports for which published inventories were not used, the methodology for estimating the
baseline rail emissions relied on results from EPA's 2011 vl NEI.9 In the NEI, emissions are reported by
source classification codes (SCCs). Table A-4 shows those related to the rail sector (line-haul class l-lll
and yard locomotives) included in this analysis and the category in which each is classified in the NEI.
Yard locomotives are categorized in the NEI as point sources.10 It should be noted that the SCCs do not
distinguish between the types of rail activities; therefore, there is no way to explicitly differentiate the
port-related locomotive emissions from other rail emissions in the NEI.
Table A-4. Relevant SCCs for Port Rail Inventory Analysis
SCC
NEI Data Category
Description
2285002006
Nonpoint
Line-haul Locomotives: Class 1 Operations
2285002007
Nonpoint
Line-haul Locomotives: Class II / III Operations
28500201
Point
Yard Locomotives
The 2008 NEI rail emission inventory was based on a methodology developed by the Eastern Regional
Technical Advisory Committee (ERTAC), a group coordinating the efforts of 27 eastern state air quality
agencies. The approach allocates locomotive emissions into three categories: Class I line-haul, Class II
and III line-haul, and Class I switchers at rail yards.11 The emission totals were based on activity levels
9 Available at: https://www.epa.eov/air-emissions-inventories/2011-national-emissions-inventorv-nei-data.
10 There is an additional SCC in the NEI for yard locomotives (2285002010: Nonpoint Yard Locomotives). However, all EPA
estimates in the NEI for yard locomotive emissions are recorded as point sources (SCC 28500201).
11 Rail classes are defined by the Surface Transportation Board. Class I refers to rail companies with operating revenues of
$433.2 million or more in 2011. Class II typically refers to what is more commonly known as regional railroads. Class III are short
line railroads that generate less than $20 million in revenue (1991 dollars). The ERTAC methodology excludes Amtrak activity
because of the difference in activity characteristics from the other rail classes.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
6
-------
Appendix A: Baseline Emission Inventory Methodology
either reported publicly to federal agencies or provided confidentially to ERTAC by the rail companies.
ERTAC also obtained proprietary fleet mix data from the Class I railroads to weight emission factors for
rail segments on which those companies operated trains. The 2011 NEI rail emissions were estimated
from the 2008 inventory using an annual growth rate of -2.475% for Class I railroads and -8.37% for Class
II and III railroads.12
NEI rail yard emissions and locations are based on data from the Federal Railroad Administration and an
analysis of the Bureau of Transportation Statistics public rail network.13 NEI shapefiles for rail activity
include rail line geometries, identifier codes, county FIPS codes, lengths, track rights, and other
geographic and administrative attributes. Shapefiles for ports and rail lines were obtained from the NEI
website.14 Updates to these shapefiles were made using the latest available data for Houston, Miami,
Norfolk, Savannah, and New York/New Jersey.
A.3.2. Modeling Approach
For ports that had published inventories, those data were included in the aggregate inventories
reported in this assessment as described in Table A-3. However, in some cases these published
inventories did not include all the pollutants needed for this assessment (e.g., VOC, carbon dioxide,
black carbon, acetaldehyde, benzene, and formaldehyde). For example, the Charleston and Norfolk
inventories do not include VOC emissions; instead, these emissions were calculated from reported total
hydrocarbon values using EPA conversion factors.15 Seattle's C02 emissions were assumed to be 99% of
the reported C02e emissions. Charleston reports neither C02 nor C02e. In this case, C02 emissions are
estimated from CO emissions by regressing the two pollutants across all other reported inventories. This
is justified by the high correlation (r2=0.984) and the fact that most engines do not include
aftertreatment for CO emissions that would substantially alter this ratio.
None of the inventories included black carbon (BC) or any of the three air toxics, acetaldehyde, benzene,
and formaldehyde. In all cases, black carbon locomotive emissions were estimated to be 77% of PM2 5
values, as reported in EPA's Report to Congress.16 Air toxic emissions were estimated VOC inventories
using speciation factors, which were calculated based on the NEI 2011 vl total switcher or line-haul VOC
and air toxics emissions from the 19 ports. Table A-5 shows the resulting speciation factors.
12 Methodology for updating 2008 NEI locomotive inventory for 2011vl NEI is described in ERG Memorandum to EPA
"Development of 2011 Railroad Component for National Emissions Inventory," September 5, 2012. Available at:
ftp://ftp.epa.eov/Emislnventorv/2011/doc/2011nei Locomotive.pdf.
13 A description of the methodology used to assess and allocate rail yard emissions in the NEI can be found in EPA (May 2011)
report "Documentation for Locomotive Component of the National Emissions Inventory Methodology" prepared by Eastern
Research Group. Available for download at ftp://ftp.epa.gov/Emislnventorv/2011/doc/2008nei locomotive report.pdf.
14 Available at: http://www.epa.gov/ttn/chief/net/2011inventorv.html.
15 U.S. Environmental Protection Agency, Conversion Factors for Hydrocarbon Emission Components, NR-002d, July 2010.
16 U.S. Environmental Protection Agency, Report to Congress on Black Carbon, EPA-450/R-12-001, March 2012, p. 87.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
7
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-5. Calculated Air Toxic Speciation Factors for Rail
BAU Source
voc
Formaldehyde
Benzene
Acetaldehyde
Rail Line
1
0.0416
0.00249
0.0181
Rail Yard
1
0.0260
0.00155
0.0113
For ports that relied on the NEI, additional calculations were required as the NEI reports locomotive
emissions only down to the county level. The steps to calculate locomotive emissions associated with
rail lines and rail yards in and around each specific port are described below.
Using the port and rail shapefiles, the rail lines were mapped in a GIS program. The rail yards were also
mapped using latitude and longitude data associated with each rail yard in the NEI point source
database. Rail lines and rail yards located within the study area were selected. Rail lines that cross the
study area boundary were "clipped" in the GIS program so that only the portions of the rail segments
that fall within the study area were selected. The GIS program was also used to measure the lengths of
each rail line segment in the study area.17 Attribute tables associated with each rail line segment and rail
yard within the study area were exported for analysis.
The exported attribute data include unique identifiers for each rail line (FRAARCID) and rail yard
(eis_facili). The rail line attributes also include data on the length of each segment; for segments that
were clipped to the study area boundary, the GIS program recalculates the new length of the rail line.
In the 2011 NEI Documentation section of the NEI website, a table with the fractions of county
emissions for rail is available for download. This table includes a field with each rail segment's unique
identification code (FRAARCID), the SCCs for line-haul locomotives (see Table A-4) and the fraction of the
total county locomotive emissions allocated to each rail segment by SCC. Using the rail segment unique
identifier, the fractions of county emissions were linked to the matching rail segment.18 For segments
that were clipped at the study boundary, the ratio of the clipped length to the original length was used
to adjust the emissions fraction proportionally; for example, a rail line with only half of its length within
the study boundary had its associated fraction of county emissions halved.
The 2011 NEI Nonpoint database identifies counties in the United States using Federal Information
Processing Standard (FIPS) codes. Counties are also identified in the rail shapefile attribute tables by FIPS
codes. Using these codes, the total county locomotive emissions were joined to each rail segment. The
county emissions were then multiplied by the fractions (or clipped fractions) identified earlier to
calculate the portion of county emissions allocated to each rail segment.
17 Note that the attribute data already in the rail line shapefile include lengths for each rail segment (as per MILES). However,
for this analysis, the segment lengths were re-measured in the GIS program to ensure consistency with the port shapefile used
to establish the study area.
18 Note that not all rail segments are listed in the county fraction table; thus, in this inventory, they have no emissions
associated with them.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
8
-------
Appendix A: Baseline Emission Inventory Methodology
The 2011 NEI Point database includes emission levels (total_emissions) by pollutant (pollutant_cd) for
each point source facility (eis_facility_site_id). The emissions for each rail yard located within the study
area were looked up using the facility identification codes (eis_facility_site_id in the NEI Point dataset
and eis_facili in the rail shapefile attribute table).
Using these steps, the emissions for the rail lines and rail yards in the study area were estimated.
However, these emissions were limited to those pollutants included in the 2011 NEI version 1, which
includes NOx, VOCs, PM2.5, benzene, acetaldehyde, and formaldehyde.19The other pollutants assessed in
this project but not included in the NEI (black carbon and carbon dioxide) were calculated separately.
Consistent with EPA's Report to Congress, black carbon emissions are assumed to be 77% of PM2.5
emissions.20
C02 emissions must be estimated from the other species available in the 2011 NEI. Emissions factors
from the 2009 EPA report "Current Methodologies in Preparing Mobile Source Port-Related Emission
Inventories" were applied as described below.21 Table A-6 shows the locomotive emission factors from
the report.
Table A-6. Line-haul Locomotive Emission Factors (g/bhp-hr)
Calendar Year
HC
CO
NOx
PM10
CO2
N2O
CH4
2005
0.48
1.28
13
0.32
483
0.04
0.013
2006
0.47
1.28
12.79
0.32
483
0.04
0.013
2007
0.45
1.28
12.15
0.30
483
0.04
0.013
2008
0.42
1.28
11.14
0.28
483
0.04
0.013
2009
0.39
1.28
10.17
0.26
483
0.04
0.013
Based on the consistent values of the CO and C02 emission factors over the five years backcasted in the
report, CO was used as a basis for scaling emissions for C02, using the ratio between the two factors.
The C02 emissions factor is 377 times greater than the CO emission factors (483/1.28). Thus, to calculate
carbon dioxide emissions, carbon monoxide values for each rail line and rail yard were multiplied by 377.
A.4. Cargo Handling Equipment
For the cargo handling equipment sector, existing information from the NEI or other models could not
be used to estimate the 2011 baseline inventory. Both the NEI and the NONROAD model allocate
emissions to counties but do not allocate emissions to localized areas like ports well. A regression
19 This work is based on vl of the 2011 NEI, the most current published version of the dataset at the time of analysis.
20 U.S. Environmental Protection Agency, Report to Congress on Black Carbon, EPA-450/R-12-001, March 2012, p. 87.
21 U.S. Environmental Protection Agency, Current Methodologies in Preparing Mobile Source Port-Related Emission Inventories,
April 2009.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
9
-------
Appendix A: Baseline Emission Inventory Methodology
analysis was performed using published port emission inventories to predict emissions at all 19 ports of
interest. This methodology is designed to avoid issues with direct calculation models and NEI values and
instead extrapolate known CHE inventories to ports with unknown values.
A.4.1. Data Sources
A.4.1.1. NEI
Using the 2011 NEI emissions estimates was not feasible for CHE. NEI values are reported at the county
level and by source classification code (SCC). SCC codes do not match well to the types of equipment
moving goods at ports, and the SCCs corresponding to equipment at ports may also be in use within the
surrounding county for applications not associated with port activity. Further, while SMOKE22 has
allocation surrogates for these SCCs, these surrogates do not account for activity at ports. Therefore,
countywide values from the published version of EPA's 2011 vl NEI23 were instead used as a quality
assurance check on the results.
A.4.1.2. Published Port Emission Inventories
Table A-7 lists those ports with sufficiently recent and comprehensive CHE emission inventories to use in
this analysis. All are for calendar year 2011, except Oakland, which is 2012. Port Arthur (2000), Portland
(2000), Corpus Christi (1999), Philadelphia (2003), and Houston (2007) also have published inventories,
but were considered too old to use here. New York/New Jersey (2008) also has a published CHE
inventory, but it is both old and only covers container terminals, which account for only about 16% of
goods handled by weight and was considered this too limited to use for regression. Richmond, CA, also
has a published inventory (2005), but it is both too old and limited to use in this regression analysis.
Table A-7. Published Port Inventories Used in This CHE Emission Regression Analysis
Port
Anacortes
Tacoma
Port of Virginia
Everett
Port of Los Angeles
Charleston
Olympia
Port of Long Beach
Oakland
Seattle
A.4.1.3. US Army Corps of Engineers Waterborne Commerce
Data for the 2011 baseline year came from Waterborne Commerce of the United States (WCUS), collected
and published by the US Army Corps of Engineers. These data included cargo throughput, in terms of
tonnage24 and TEUs25 and includes both domestic and international movements.26 USACE TEU data were only
22 SMOKE is the Sparse Matrix Operator Kernel Emissions Model. It is an emission inventory processing system used in certain
photochemical modeling applications, such as the Community Multi-Scale Air Quality Model (CMAQ).
23 Available at: https://www.epa.eov/air-emissions-inventories/2011-national-emissions-inventorv-nei-data.
24 Available at: http://www.navigationdatacenter.us/db/wcsc/archive/xls/manll/.
25 Available at: http://www.navigationdatacenter.us/wcsc/bv portnamesll.html.
26 Available at: http://www.navigationdatacenter.us/data/datawcus.htm and
http://www.navieationdatacenter.us/data/datappor.htm.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
10
-------
Appendix A: Baseline Emission Inventory Methodology
available for the 84 ports USACE considers "principal." Cargo throughput for most domestic waterways is
available through the WCUS dataset. These data track cargo throughput (in thousands of tons) in 146
commodity categories. For this analysis, these commodity categories were grouped into four conveyance
methods: dry and break bulk, liquid bulk, containerized (TEUs), and other (principally roll-on/roll-off and
automobile). Passenger is not considered a "conveyance method" and not included here. Table A-8 lists the
USACE commodity types tracked in the WCUS dataset and the four conveyance methods used.
Table A-8. Commodity Types and Matching Conveyance Methods
Conveyance
Method
Product
Bulk
Aluminum, Aluminum Ore, Animal Feed, Prep., Barley & Rye, Building Stone, Cement & Concrete,
Clay & Refrac. Mat., Coal Coke, Coal Lignite, Cocoa Beans, Copper, Copper Ore, Corn, Dredged
Material, Ferro Alloys, Fert. & Mixes NEC, Flaxseed, Forest Products NEC, Fuel Wood, Grain Mill
Products, Gypsum, Hay & Fodder, l&S Bars & Shapes, l&S Pipe & Tube, l&S Plates & Sheets, l&S
Primary Forms, Iron & Steel Scrap, Iron Ore, Lime, Limestone, Lumber, Machinery (Not Elec),
Manganese Ore, Marine Shells, Metallic Salts, Misc. Mineral Prod., Molasses, Non-Ferrous Ores NEC,
Non-Ferrous Scrap, Non-Metal. Min. NEC, Non-Metal. Min. NEC, Oats, Ordnance & Access., Peanuts,
Petroleum Coke, Phosphate Rock, Pig Iron, Primary l&S NEC, Primary Wood Prod., Pulp & Waste
Paper, Radioactive Material, Sand & Gravel, Slag, Soil & Fill Dirt, Sorghum Grains, Starches, Gluten,
Glue, Sugar, Sulphur, (Dry), Unknown or NEC, Waste and Scrap NEC, Waterway Improv. Mat, Wheat,
Wheat Flour, Wood Chips, Wood in the Rough
Container
Alcoholic Beverages, Animals & Prod. NEC, Bananas & Plantains, Coffee, Coloring Mat. NEC, Cotton,
Dairy Products, Electrical Machinery, Empty Containers, Fab. Metal Products, Farm Products NEC,
Food Products NEC, Fruit & Nuts NEC, Glass & Glass Prod., Groceries, Inorg. Elem., Oxides, &
Halogen S, Manufac. Prod. NEC, Manufac. Wood Prod., Meat, Fresh, Frozen, Meat, Prepared,
Medicines, Natural Fibers NEC, Newsprint, Nitrogen Func. Comp., Paper & Paperboard, Paper
Products NEC, Perfumes & Cleansers, Pesticides, Pigments & Paints, Plastics, Rice, Rubber & Gums,
Rubber & Plastic Pr., Smelted Prod. NEC, Soybeans, Textile Products, Tobacco & Products,
Vegetables & Prod.
Liquid
Acyclic Hydrocarbons, Alcohols, Ammonia, Asphalt, Tar & Pitch, Benzene & Toluene, Carboxylic
Acids, Chem. Products NEC, Chemical Additives, Crude Petroleum, Distillate Fuel Oil, Fruit Juices,
Gasoline, Inorganic Chem. NEC, Kerosene, Liquid Natural Gas, Lube Oil & Greases, Naphtha &
Solvents, Nitrogenous Fert., Oilseeds NEC, Organic Comp. NEC, Organo-lnorganic Comp., Other
Hydrocarbons, Petro. Jelly & Waxes, Petro. Products NEC, Phosphatic Fert., Potassic Fert., Residual
Fuel Oil, Sodium Hydroxide, Sulphur (Liquid), Sulphuric Acid, Tallow, Animal Oils, Vegetable Oils,
Water & Ice, Wood & Resin Chem.
Other
Aircraft & Parts, Explosives, Fish (Not Shellfish), Fish, Prepared, Shellfish, Ships & Boats, Vehicles,
Vehicles & Parts
A.4.2. Modeling Approach
A regression model was developed based on the observed relationship between port cargo throughput
and CHE emissions. The recent CHE emission inventories for the ports listed in Table A-7 were collected,
and gaps that existed for certain pollutants in some of the inventories were filled as described in the
next section. Several different regression models were explored to determine correlations between
NOx, VOC, PM2.5, and C02 emissions in tons per year against cargo throughput:
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
11
-------
Appendix A: Baseline Emission Inventory Methodology
¦ Method 1: regressed total CHE emissions for each pollutant against the tonnage in the four
conveyance types
¦ Method 2: regressed total CHE emissions for each pollutant against the total non-container
tonnage and number of TEUs
¦ Method 3: regressed total CHE emissions for each pollutant against total tonnage throughput
only, excluding conveyance type
¦ Method 4: represents an unweighted average of the predictions from the above three methods
Method 4 was used in this analysis as the best option based on engineering judgement. This regression
model has similar limitations as with other sectors that rely on NEI values. As with those sectors, it does
not allow characterization of emissions by equipment age, fuel type, terminal type, existing use of
control technology, or other discriminators. In this analysis, all regressions were performed against total
CHE emissions rather than against individual equipment types due to data limitations.27
In cases where different definitions of hydrocarbons were included in the published inventories, all
species were converted to VOC using NONROAD factors.28 For published inventories where only C02e
was included, C02 was estimated as 99% of C02e. BC emissions were taken as 77% of the regressed
PM2.5 emissions, consistent with EPA's Report to Congress.29
For all modeled ports, VOC, PM2.5, NOx, and C02 were determined from the regression model. Benzene,
acetaldehyde, formaldehyde, and BC emissions were then determined from the VOC or PM2.5, as
appropriate, either from the regressed estimates (for modeled ports) or from the published values
(where available). Speciation factors for benzene, acetaldehyde, and formaldehyde relative to VOC were
derived from values in version 1 of the 2011 NEI. National total emissions of the air toxics and VOCs
were summed for the relevant CHE SCCs (see Table A-9) and ratios were calculated for each. These
speciation factors are very similar to the national averages considered for each SCC individually, which
showed little variation, but accommodated the different fuel types in use in CHE nationally.
Four calculated port inventories were compared against countywide NEI values for certain SCCs to
confirm the results as reasonable. Only the emissions from SCCs that potentially may operate on ports
were included in this comparison. See Table A-9 for a complete listing of SCCs included. The calculated
inventories for the four ports were compared to the countywide NEI values for the associated areas as
listed in Table A-10.
27 Note that this regression includes Los Angeles and Long Beach as inputs. These ports were later removed from the
assessment's results but not until after this analysis had been conducted. It is possible that the CHE emissions profile at these
ports differs from other, non-California ports and may influence the results.
28 NONROAD HC Conversion Factors. Available at http://www.epa.gov/otaq/models/nonrdmdl/nr-002.pdf.
29 U.S. Environmental Protection Agency, Report to Congress on Black Carbon, EPA-450/R-12-001, March 2012, p. 87.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
12
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-9. 2011 NEI CHE Types by SCC
SCC
SCC Level Four
SCC Level Three
2270002015
Rollers
Construction and Mining Equipment
2270002027
Signal Boards/Light Plants
Construction and Mining Equipment
2270002036
Excavators
Construction and Mining Equipment
2270002045
Cranes
Construction and Mining Equipment
2270002051
Off-highway Trucks
Construction and Mining Equipment
2270002060
Rubber Tire Loaders
Construction and Mining Equipment
2270002063
Rubber Tire Tractor/Dozers
Construction and Mining Equipment
2270002066
Tractors/Loaders/Backhoes
Construction and Mining Equipment
2270002069
Crawler Tractor/Dozers
Construction and Mining Equipment
2270002072
Skid Steer Loaders
Construction and Mining Equipment
2270002075
Off-highway Tractors
Construction and Mining Equipment
2270003010
Aerial Lifts
Industrial Equipment
2270003020
Forklifts
Industrial Equipment
2270003030
Sweepers/Scrubbers
Industrial Equipment
2270003050
Other Material Handling Equipment
Industrial Equipment
2270003070
Terminal Tractors
Industrial Equipment
2270006005
Generator Sets
Commercial Equipment
2270006010
Pumps
Commercial Equipment
2270006015
Air Compressors
Commercial Equipment
2270006025
Welders
Commercial Equipment
Table A-10. NEI Areas Used in Regression Comparison
Port
NEI Areas of Comparison
Port of Houston
Harris County, TX
Port of New Orleans
St. Charles, Jefferson, and Orleans Parishes
Port of Portland
Multnomah County, OR
Port of Savannah
Chatham County, GA
The comparison showed that the calculated port inventories were between 2% and 54% of the
corresponding NEI inventories for C02, NOx, PM2.5, and VOCs, with percentages varying by port and
pollutant. For the air toxics studied, three of the port inventories were between 3% and 32% of their
corresponding NEI inventories, but one port was calculated to be approximately 185% of its NEI
inventory for air toxics. However, without additional information on other construction, mining,
industrial, and commercial activity present in each of these areas, it is difficult to comprehensively
assess the reasonableness of these values. Based on engineering judgement, the regression model was
determined to be adequate for the purposes of this assessment.
A.5. Harbor Craft
This section discusses the methodology used to develop the 2011 baseline harbor craft sector emission
inventory. The term "harbor craft" is used synonymously for all Category 1-Category 2 vessels (C1/C2).
For this sector, all emissions are determined based on the 2011 NEI. Existing port inventories were not
used to assess harbor craft emissions because most port inventories only included harbor craft
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
13
-------
Appendix A: Baseline Emission Inventory Methodology
emissions related to their own port operations, and did not include, for example, harbor craft activity
that could be related to private terminals.
A.5.1. Data Sources
The primary data source for this sector was a draft of 2011 NEI v2. In the NEI, emissions are reported by
source classification codes (SCCs). Table A-ll shows those related to the harbor craft emission sectors
included in this analysis.
Table A-ll. Relevant SCCs for Port Inventory Analysis
see
Emission Type Code*
Description
2280002100
M
Harbor Craft (C1/C2) at Port
2280002200
C
Harbor Craft (C1/C2) underway
* Emission type codes for C1/C2 vessels are defined as M for maneuvering (in port) and C for
cruise (out of port).
The NEI contains some measure of both state, local, and tribal (SLT) estimates and EPA estimates of
emissions. Consistent with the national focus of this analysis, and to insure that a consistent
methodology was used for each area, only the EPA derived emission estimates were used instead of the
SLT submitted estimates.
A.5.2. Modeling Approach
The NEI includes an allocation of marine vessel emissions to GIS shapefiles for near-port operating
modes (maneuvering) and out-of-port (cruise) modes for harbor craft. These NEI-defined polygons
allocate harbor craft emissions to either port-encompassing polygons30 or open water shipping lane
polygons,31 respectively. The polygons of port boundaries are based on maps provided directly from
ports, from local port authorities and port districts, and from satellite imagery and GIS tools. Polygons
were created on land, bordering waterways, and coastal areas, and were split by county boundary such
that no shape file crosses county lines and county total emission can be easily summed.32 These are
referred to in this report as the "NEI shapes."
This analysis identified all NEI shipping lane shapes within 5 km of each port facility,33 where 5 km
represents the outermost edge of cruising activity for the purposes of this assessment. All NEI shipping
lane shapes were clipped at this 5 km buffer edge and the portion of each shape within the buffer was
recorded. There were two exceptions to this exercise for the Port of Miami and Port of Hampton Roads.
30 Available at: http://www.epa.eov/ttn/chief/eis/2011nei/2011 ports shapefile.zip.
31 Available at: http://www.epa.gov/ttn/chief/eis/2011nei/shippinelanes 112812 shapefile.zip.
32 U.S. EPA, 2011 NEI Technical Support Document, November 2013, available at: https://www.epa.gov/air-emissions-
inventories/2011-nei-technical-support-document.
33 In cases where ORD shapes are taken representing the port, the 5 km buffer for shipping activity extends from the boundary
of these shapes, not the original NEI shapes.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
14
-------
Appendix A: Baseline Emission Inventory Methodology
For the Port of Miami, the cruising activity in the Port of Miami River was not included. Additionally, NEI
shapes for Newport News were excluded from the Port of Hampton Roads.
To calculate the harbor craft inventory, at port maneuvering emissions for NOx, PM25, VOC, and CO
reported in the draft 2011 v2 NEI were summed by harbor craft type for each port location. Cruising
emissions for the same pollutants were determined by summing the NEI emissions for each port's
associated shipping lane shapes. In cases where a shipping lane shape extends beyond the 5 km buffer, a
proportional fraction of the shape's emissions were attributed to the port. For example, if 10% of a
port's shipping lane shape lies within the 5 km buffer, 10% of the total cruising mode emissions
associated with that shape are included in the inventory for that port.
The draft version 2 of the 2011 NEI that was used for this analysis did not contain estimates for C02, air
toxics, or black carbon. However, a previous study34 indicated that C02 and VOC show a direct
correlation, more so than for CO or other pollutant. Therefore, a scaling factor for VOC to C02 emissions
was calculated based on a previous iteration of the draft 2011 version 2 NEI. The scaling factor was
determined to be 3,247.9 tons C02 per ton VOC, which did not vary by vessel type.
Similarly to C02, the air toxics were speciated from VOC using factors calculated from a previous
iteration of the draft 2011 version 2 NEI. Table A-12 shows these speciation factors.
Table A-12. Select Air Toxic Speciation Factors from VOC Emissions for Harbor Craft
see
Pollutant
Speciation Factor
2280002100
Acetaldehyde
0.0557235
2280002200
Acetaldehyde
0.046436
2280002100
Benzene
0.015258
2280002200
Benzene
0.012715
2280002100
Formaldehyde
0.1122
2280002200
Formaldehyde
0.0935
Consistent with EPA's Report to Congress, black carbon emissions are assumed to be 77% of PM2 5
emissions.35
A.6. Ocean Going Vessels
This section details the methodology used for developing the ocean going vessel (OGV) 2011 baseline
emission inventories. It is based primarily upon the methodology used for the Category 3 Marine Engine
Rulemaking36 (C3 RIA). Using the C3 RIA modeling approach, the OGV emission inventories were
34 U.S. Environmental Protection Agency, Current Methodologies in Preparing Mobile Source Port-Related Emission Inventories,
April 2009.
35 U.S. Environmental Protection Agency, Report to Congress on Black Carbon, EPA-450/R-12-001, March 2012, p. 87.
36 U.S. Environmental Protection Agency, Regulatory Impact Analysis: Control of Emissions of Air Pollution from Category 3
Marine Diesel Engines, EPA Report EPA-420-R-09-019, December 2009. Available at:
http://www.epa.eov/otaa/rees/nonroad/marine/ci/420r09019.pdf.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
15
-------
Appendix A: Baseline Emission Inventory Methodology
calculated using energy-based emission factors combined with activity profiles for vessels calling at each
port.
A.6.1. Data Sources
Consistent with the C3 RIA, the three primary C3 activity data sources used in this assessment were the
US Army Corps of Engineers Entrances and Clearances data, Lloyd's Register of Ships (Lloyd's data), and
Marine Exchange/Port Authority (MEPA) data. Each is described below.
A.6.1.1. US Army Corps of Engineers Entrances and Clearances
The USACE 2011 Entrances and Clearances data37 were used to determine ship calls (trips or visits to a
port). The Maritime Administration (MARAD) maintains the Foreign Traffic Vessel Entrances and
Clearances database, which contains statistics on U.S. foreign maritime trade. USACE compiles these
data to build a database which contains information on the vessel International Maritime Organization
(IMO) number, type of vessel, commodities, weight, customs districts and ports, and origins and
destinations of goods.
There are several limitations to using USACE Entrances and Clearances data. First, they do not contain
any average time in mode or speed information, which is important when estimating emissions. Second,
they only represent foreign cargo movements. Domestic cargo traffic (U.S. ships delivering cargo from
one U.S. port to another U.S. port covered under the Jones Act38) are not always accounted for in the
database. However, some trips made by Jones Act vessels are accounted for if they are carrying cargo
from a foreign port to a U.S. port or from a U.S. port to a foreign port, since these are considered foreign
cargo movements. At most ports, domestic commerce is carried out by Category 2 ships, although there
are a few exceptions, particularly on the West Coast.39 While Automatic Identification System (AIS) data
can be used to determine domestic cargo trips, the processing of these data is time consuming and not
within the scope of this assessment. Third, the Entrances and Clearances data do not always match
MEPA data because the USACE Entrances and Clearances data include cargo movements from both
public and private terminals at a port while the MEPAs usually only cover calls at public terminals. Port
Authorities generally do not have jurisdiction over private terminals. Since the USACE Entrances and
Clearances data account for over 90% of the emissions from Category 3 ships calling on U.S. ports,40
these limitations should not have a substantial impact on the calculation of C3 emissions for this report.
A.6.1.2. Lloyd's Register of Ships
Lloyd's Register of Ships offers the largest database of commercially available maritime data in the world
and is produced by IHS Global Limited.41 The 2014 version used in this assessment has details on 180,000
37 USACE, Vessel Entrances and Clearances. Available at: http://www.navigationdatacenter.us/data/dataclen.htm.
38 Merchant Seaman Protection and Relief 46 USCS Appx § 688 (2002) Title 46. Appendix. Shipping Chapter 18.
39 ICF International, Inventory Contribution of U.S. Flagged Vessels, June 2008.
40 Merchant Seaman Protection and Relief 46 USCS Appx § 688 (2002) Title 46. Appendix. Shipping Chapter 18.
41 Available at: http://www.sea-web.com.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
16
-------
Appendix A: Baseline Emission Inventory Methodology
vessels and 200,000 companies that own, operate, and manage them. Lloyd's data contain the following
information on ship characteristics that are important for preparing detailed marine vessel inventories:
All data are referenced to both ship name (current or former) and IMO number. Only IMO number is a
unique identifier for each ship. Lloyd's insures many of the OGVs on an international basis, and for these
vessels, the data are quite complete. For other ships using a different insurance certification authority,
some of the data are missing such as main engine power, size and configuration, and vessel service
speed.
A.6.1.3. Marine Exchange/Port Authority Data
For this analysis, Marine Exchange/Port Authorities data were used to estimate hoteling and
maneuvering times. As with the C3 RIA, much of the MEPA data came from a 1999 report43 that
described how to calculate marine vessel activity at deep sea ports and contained detailed port activities
of eight deep sea ports. The detailed inventories were developed by obtaining ship call data from MEPA
at the various ports for 1996 and matching the various ship calls to data from Lloyd's Maritime
Information Services to provide ship characteristics. A 2002 port emission inventory for the Port of
Houston44 was also added to the list of "Typical Ports" in the 1999 report. The ports for which detailed
inventories were developed are shown in Table A-13 along with the level of detail of shifts (movements
within ports between berth and anchorage or between different berths) for each port. Most ports
provided the ship name, IMO number, the vessel type, the date and time the vessel entered and left the
port, and the vessel flag.
In addition to the detailed port inventories of ship activity, the 1999 report (which was also used in the
C3 RIA) laid out a methodology to determine which of the typical ports can be used as a surrogate for
other "like" or "modeled" ports that were being modeled in the C3 RIA. Consistent with the C3 RIA,
hoteling and maneuvering times from the typical ports were used in this analysis to determine emissions
at berth at the modeled ports. If the typical port data included shifts and anchorages, these data were
also used in the modeled port. Anchorage times were only broken out for the Patapsco River Ports
(Baltimore) in the typical port data.
42 Vessel service speed is the average speed maintained by a ship under normal load and weather conditions.
43 ARCADIS Geraghty & Miller, Commercial Marine Activity for Deep Sea Ports in the United States, EPA Report EPA420-R-99-
020, September 1999. Available at http://www.epa.gov/otaq/models/nonrdmdl/c-marine/r99020.pdf .
44 Starcrest Consulting Group, Houston Galveston Area Vessel Emissions Inventory, November 2002.
¦ Name (current and former)
¦ Ship Type
¦ Build Date
¦ Flag
¦ Deadweight tonnage (DWT)
¦ Vessel service speed42
¦ Main engine power, size, and configuration
¦ Limited data on Auxiliary engines
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
17
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-13. Deep Sea MEPA Vessel Movement and Shifting Details Used in the C3 RIA
MEPA Area and Ports3
Data Year
MEPA Data Include
Lower Mississippi River including
the ports of New Orleans, South
Louisiana, Plaquemines, and Baton
Rouge
1996
Information on the first and last pier/wharf/dock (PWD) for
the vessel (gives information for at most one shift per
vessel). No information on intermediate PWDs, the time of
arrival at the first destination PWD, or the time of departure
from the River.
Consolidated Port of New York
and New Jersey and other ports on
the Hudson and Elizabeth Rivers
1996
All PWDs or anchorages for shifting are named. Shifting
arrival and departure times are not given. Maneuvering and
hoteling times are estimated from average speed and
distance rather than calculated from date and time fields.
Delaware River Ports including the
ports of Philadelphia, Camden,
Wilmington and others
1996
All PWDs or anchorages for shifting are named. Shifting
arrival and departure times are not given. Maneuvering and
hoteling times are estimated from average speed and
distance rather than calculated from date and time fields.
Puget Sound Area Ports including
the ports of Seattle, Tacoma,
Olympia, Bellingham, Anacortes,
and Grays Harbor
1996
All PWDs or anchorages for shifting are named. Arrival and
departure dates and times are noted for all movements,
allowing calculation of maneuvering and hoteling both for
individual shifts and the overall call on port.
The Port of Corpus Christi, TX
1996
Only has information on destination PWD and date and time
in and out of the port area. No shifting details.
The Port of Coos Bay, OR
1996
Only has information on destination PWD and date and time
in and out of the port area. No shifting details.
Patapsco River Ports including the
port of Baltimore Harbor, MD
1996
All PWDs or anchorages for shifting are named. Shifting
arrival and departure times are not given. Maneuvering and
hoteling times are estimated from average speed and
distance rather than calculated from date and time fields.
The Port of Tampa, FL
1996
All PWDs or anchorages for shifting are named. Arrival and
departure dates and times are noted for all movements,
allowing calculation of maneuvering and hoteling both for
individual shifts and the overall call.
Port of Houston
2000
PWD and shifts provided. RSZ, maneuvering and hoteling
times were calculated from available data by ship type.
a All marine exchange/port authority data listed above were for 1996 and were taken directly from the 1999 EPA Report. Port
of Houston data were for 2000 and were taken directly for the Starcrest 2002 inventory report.45
Several emission inventories have been published since the original C3 RIA. They provided additional
data on hoteling times, anchorage times, and maneuvering times and were used in this analysis in place
of the data used in the C3 RIA, where appropriate. The updated data sources are listed in Table A-14.
Some older inventories were also used to determine maneuvering and hoteling times, where newer
data were not available. In general, maneuvering, hoteling and anchorage times were obtained from the
45 Starcrest Consulting Group, Houston Galveston Area Vessel Emissions Inventory, November 2002.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
18
-------
Appendix A: Baseline Emission Inventory Methodology
1999 report unless detailed in Table A-14. The C3 RIA was used to determine RSZ speeds unless noted
below.
Table A-14. More Recently Published Port Emission Inventories Since C3 RIA
Port Inventory
Data Year
Data Include
Port of New York & New Jersey
Starcrest Consulting Group, The Port of New York and New
Jersey Port Commerce Department 2012 Multi-Facility
Emissions Inventory, August 2014.
https://www. panvni.gov/about/pdf/panvni-multi-facilitv-ei-
reoort 2012.odf
2012
Information to calculate maneuvering
times and hoteling times by ship type for
all ship types
Puget Sound Area Ports
Starcrest Consulting Group, 2011 Puget Sound Maritime
Emission Inventory, September 2012.
http://www.pugetsoundmaritimeairforum.org/uploads/PV
2011
Information to calculate maneuvering
times and hoteling times by ship type for
all ship types
FINAL POT 2011 PSEI Report Update 23 Mav 13 scg.
£df
Port of Charleston
Moffatt & Nichol, South Carolina State Ports Authority 2011
Emissions Inventory Update, April 2013
http://www.scspa.com/SCSPA Emission Inventory 2011 Fi
2011
Maneuvering and hoteling times by ship
type for all ship types
nal Reoort lAoril2013.Ddf
Port of Virginia
Moffatt & Nichol, The Port of Virginia 2011 Comprehensive
Emissions Inventory Update, January 2013
2011
RSZ times by area and ship type for all
ship types
Port of Baltimore
Maryland Port Authority et al., Emission Reductions from
Port of Baltimore Maritime Vessels and Cargo Handling
Equipment, September 2010
2010
RSZ times by ship type for all ship types
Port of Houston
Starcrest Consulting Group, 2007 Goods Movement Air
Emissions Inventory at the Port of Houston, January 2009
htto://www.Dortofhouston.com/static/sen/inside-the-
port/Environment/PH A-GM-AirEmissions-07.pdf
2007
RSZ speeds by area and ship type for all
ship types
Port of Corpus Christi
2005 hoteling data obtained directly from Port of Corpus
Christi
Email from Sarah Kowalski, July 2007
2005
Hoteling time by ship type for all ships
types. (These data replaced the typical
port hoteling times for Port of Corpus
Christi and were used for hoteling times
for Port Arthur, Corpus Christi, and for
tanker ships at Port of Houston.)
A.6.1.4. Port Matching
Both the original 1996 ports data shown in Table A-13 and more recently published data shown in Table
A-14 were used as "typical ports". The inventory data from these typical ports were used as surrogates
to estimate maneuvering time, hoteling time, and anchorage time (if available) for each of the 19
modeled ports unless noted in Table A-15 below. The typical port used in this analysis and those used in
the C3 RIA are shown in the table. The typical ports that have changed since the C3 RIA are italicized and
further explanation about why this change has been made is provided in Table A-16 below.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
19
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-15. OGV Published Inventories and Data
Port Name
Published Inventory Data Used
Port of New York and New Jersey
2012 Port of New York/New Jersey Inventory
Port of New Orleans
1996 Lower Mississippi River Inventory
Port of Miami
1996 Port of Tampa for passenger ship hoteling time
1996 Delaware River Inventory for other ship types
Port of South Louisiana
1996 Lower Mississippi River Inventory
Port of Seattle
2011 Puget Sound Inventory
Port of Baton Rouge
1996 Lower Mississippi River Inventory
Port Arthur
1996 Port of Corpus Christi Inventory for maneuvering
2005 Port of Corpus Christi Inventory for hoteling
Port of Portland
2012 New York/New Jersey Inventory
Port of Mobile
1996 Delaware River Inventory
2007 Houston Inventory for RSZ speeds
Port of Houston
2000 Houston Inventory used for maneuvering times
2005 Corpus Christi data for Tanker hoteling time
2012 New York/New Jersey Inventory for other ship hoteling times
2010 Port of Baltimore Report for RSZ speeds
Port of Baltimore
1996 Patapsco River Inventory for maneuvering, hoteling and
anchorage
2011 Port of Virginia Inventory for RSZ speeds
Port of Hampton Roads (Norfolk)
1996 Patapsco River Inventory for maneuvering, hoteling and
anchorage
Port of Philadelphia
1996 Delaware River Inventory
Port of Charleston
2011 Port of Charleston Inventory
Port of Corpus Christi
1996 Port of Corpus Christi Inventory for maneuvering
2005 Port of Corpus Christi Inventory for hoteling
Port Tampa Bay
1996 Port of Tampa for passenger ship hoteling time
1996 Delaware River Inventory for other ship types
Port of Savannah
1996 Patapsco River Inventory
Port of Coos Bay, OR
1996 Port of Coos Bay Inventory
Port of San Juan, PR
1996 Delaware River Typical Port Data
Notes: Italics denote ports that use a different matching typical port than what was used in the C3 RIA.
Year given reflects data year, not report year.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
20
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-16. Typical Port Matching
Port Name
Current Matching Typical Port
C3 RIA Matching Typical Port
Port of New York and
New Jersey
2012 Port of New York/New Jersey Inventory
1996 Port of New York/New Jersey
Inventory
Port of New Orleans
1996 Lower Mississippi River Inventory
1996 Lower Mississippi River Inventory
Port of Miami
1996 Port of Tampa for passenger ship
hoteling time
1996 Delaware River Inventory for other ship
types
1996 Delaware River Inventory
Port of South Louisiana
1996 Lower Mississippi River Inventory
1996 Lower Mississippi River Inventory
Port of Seattle
2011 Puget Sound Inventory
1996 Puget Sound Inventory
Port of Baton Rouge
1996 Lower Mississippi River Inventory
1996 Lower Mississippi River Inventory
Port Arthur
1996 Port of Corpus Christi Inventory for
maneuvering
2005 Port of Corpus Christi Inventory for
hoteling
1996 Port of Corpus Christi Inventory
Port of Portland
2012 New York/New Jersey Inventory
1996 Puget Sound Inventory
Port of Mobile
1996 Delaware River Inventory
1996 Port of Corpus Christi Inventory
Port of Houston
2007 Houston Inventory for RSZ speeds
2000 Houston Inventory used for
maneuvering times
2005 Corpus Christi data for Tanker hoteling
time
2012 New York/New Jersey Inventory for
other ship hoteling times
2000 Port of Houston Inventory
Port of Baltimore
2010 Port of Baltimore Report for RSZ speeds
1996 Patapsco River Inventory for
maneuvering, hoteling and anchorage
1996 Patapsco River Inventory
Port of Hampton Roads
(Norfolk)
2011 Port of Virginia Inventory for RSZ speeds
1996 Patapsco River Inventory for
maneuvering, hoteling and anchorage
Not in C3 RIA
Port of Philadelphia
1996 Delaware River Inventory
1996 Delaware River Inventory
Port of Charleston
2011 Port of Charleston Inventory
1996 Delaware River Inventory
Port of Corpus Christi
1996 Port of Corpus Christi Inventory for
maneuvering
2005 Port of Corpus Christi Inventory for
hoteling
1996 Port of Corpus Christi Inventory
Port Tampa Bay
1996 Port of Tampa for passenger ship
hoteling time
1996 Delaware River Inventory for other ship
types
1996 Port of Tampa Inventory
Port of Savannah
1996 Patapsco River Inventory
1996 Patapsco River Inventory
Port of Coos Bay, OR
1996 Port of Coos Bay Inventory
1996 Port of Coos Bay Inventory
Port of San Juan, PR
1996 Delaware River Typical Port Data
Not in C3 RIA
Notes: Italics denote ports that use a different matching typical port than what was used in the C3 RIA.
Year given reflects data year, not report year.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports 21
-------
Appendix A: Baseline Emission Inventory Methodology
For many of the ports listed in Table A-16, the data were from 1996. However, these data were used to
determine operating characteristics, and it is expected that ship operations at the ports should not have
significantly changed since 1996 unless there was a significant change in the physical layout of the port.
While the Port of Hampton Roads (Norfolk) was not part of the C3 RIA, Port of Newport News, which is
nearby, used the 1996 Patapsco River Inventory as the matching typical port. Because the Port of Newport
News was included in the C3 RIA and is in the same waterway, the 1996 Patapsco River Inventory was used
as the matching typical port for Hampton Roads.
Other ports were matched to typical ports based on the type of ships that frequent the port and port cargo
throughput. For example, a majority of the ship calls at the Port of Portland and San Juan are container ships
while at Port Arthur the majority is tankers. Thus, Portland and San Juan were matched with other container
ports and Port Arthur with other tanker ports. The Port of Mobile is served by a variety of vessel types more
consistent with the Delaware River ports, than with Corpus Christi, which only covered tanker vessels. The
Port of Houston has both container and tanker traffic, so two different typical ports were used for hoteling
times. Newer 2005 Corpus Christi data were used for tanker hoteling for the Port of Houston as well as for all
ships at Port Arthur. These newer port data provide updated information over the 1996 data. The 1996
Delaware River Inventory was also used for Port Tampa Bay because the original 1996 Port of Tampa
Inventory no longer reflected the type of vessels operating at Port Tampa Bay. For example, in 1996, there
were only four containerships visiting the port and in 2011, there were 79 containerships. As a result, the
Delaware Inventory was believed to be a better match. However, the 1996 Port of Tampa inventory provided
much better hoteling time information for passenger ships for both Tampa and Miami than the Delaware
inventory since both those ports have high passenger ship volumes as opposed to the Delaware River ports.
The remaining container ports that use different matching ports in this analysis from the C3 RIA (as shown in
Table A-16) were matched based upon TEUs unloaded and loaded per call. Amount of TEUs being unloaded
are one factor that affects hoteling time, another important factor is how efficient unloading operations are
at a port. Evaluating unloading operations and other factors that may affect hoteling time were beyond the
scope of this analysis, and TEUs were used as a proxy for hoteling time. TEUs by port and calls were used to
match maneuvering and hotelling times to typical ports. The 2012 Port of New York/New Jersey was used as
a matching typical port for maneuvering and hoteling times for the Port of Portland and for non-tanker ships
for Houston. TEUs per call for the Ports of Mobile, Miami, Tampa Bay,46 and San Juan were similar to the Port
of Philadelphia (a Delaware River Typical Port), so the 1996 Delaware River Inventory data were used as the
typical port for those ports.
Generally, the typical port data show different hoteling and maneuvering times based upon ship size. In some
cases, the specific vessel deadweight tonnage (DWT) range bin at the modeled port was not in the typical
port data. In those cases, the next nearest DWT range bin was used for the calculations, which was done in
the C3 RIA. In a few cases, the engine type for a given ship type at the modeled port might not be in the
46 While Tampa was one of the original typical ports used in the C3 RIA, it provides minimal container ship data and thus the
1996 Delaware River typical port data were substituted.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
22
-------
Appendix A: Baseline Emission Inventory Methodology
typical port data. In these cases, the closest engine type at the typical port was used. Also in a few cases, a
specific ship type in the modeled port data was not in the typical port data. In this case, data from the most
similar ship type at the typical port were used at the modeled port. Section A.6.7 provides more information
on bin matching.
A.6.2. Modeling Approach
Using the C3 RIA modeling approach, the OGV emission inventories were calculated using energy-based
emission factors combined with activity profiles for each vessel. Ships calling on each port were binned
by ship type, engine type, and DWT, and emissions for each mode were calculated for auxiliary and main
propulsion engines using the following general equation:
E = PxLFxAxEF Eq. A-l
Where
E = Emissions (grams [g]),
P = Maximum Continuous Rating Power (kilowatts [kW]),
LF = Load Factor (percent of vessel's total power),
A = Activity (hours [h]) (hours/call * # of calls), and
EF = Emission Factor (grams per kilowatt-hour [g/kWh]).
The emission factor is in terms of emissions per unit of energy from the engine. It is multiplied by the energy
needed to move the ship or perform other activities like hoteling. Only four modes of activity were
considered in this assessment: reduced speed zone, maneuvering, hoteling, and anchorage (if available).
Cruising emissions were omitted since the scope of this assessment is limited to near-port emissions.
OGVs vary greatly in design, operating speeds, and engine sizes based on vessel type. For this analysis, vessel
types were broken out by the cargo they carry. Table A-17 lists the OGV types that are used in this analysis.
Table A-17. Oceangoing Vessel Ship Types
Ship Type
Description
Auto Carrier
Self-propelled dry-cargo vessels that carry containerized automobiles
Bulk Carrier*
Self-propelled dry-cargo ship that carries loose cargo
Container Ship
Self-propelled dry-cargo vessel that carries containerized cargo
General Cargo
Self-propelled cargo vessel that carries a variety of dry cargo
Miscellaneous
Category for those vessels that do not fit into one of the other categories or are unidentified
Passenger
Self-propelled cruise ships
Reefer
Self-propelled dry-cargo vessels that often carry perishable items
Roll-on/Roll-off
(RORO)
Self-propelled vessel that handles cargo that is rolled on and off the ship, including ferries
Tanker
Self-propelled liquid-cargo vessels including chemical tankers, petroleum product tankers,
liquid food product tankers, etc.
Tug
Self-propelled tugboats and towboats that tow/push cargo or barges in the open ocean
*ln the 2002 C3 analysis, barge carriers were considered as a separate vessel type. Due to the small number of barge
carriers in the dataset, barge carriers are considered as bulk carriers.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
23
-------
Appendix A: Baseline Emission Inventory Methodology
Other characteristics that are determined from Lloyd's data are the build year, propulsion engine power,
vessel service speed, engine type (i.e., diesel, gas or steam turbines), and vessel DWT. Vessel service
speed is the average speed maintained by a ship under normal load and weather conditions in the open
ocean. Design speed is the maximum speed a vessel can travel but is generally not used due to high fuel
consumption requirements.
Both propulsion and auxiliary marine vessel engines are defined by the categories shown in Table A-18.
Note that only vessels with Category 3 propulsion engines are considered in the OGV sector. Emissions
from Category 1 and 2 vessels are addressed in the harbor craft sector. Most ships have diesel engines,
although some ships have steam turbines (ST) and others have gas turbines (GT). Some ships are electric
drive (ED). For the purpose of this analysis, Category 3 slow speed diesel (SSD) are considered 2-stroke
engines while Category 3 medium speed diesel (MSD) engines are considered 4-stroke engines. Engine
speed designations for diesel powered ships are shown in Table A-19.
Table A-18. EPA Marine Compression Ignition Engine Categories
Category
Specification
Use
1
Gross Engine Power > 37 kW*
Displacement < 7 liters per cylinder
Small harbor craft and
recreational propulsion
2
Displacement > 7 and < 30 liters
per cylinder
OGV auxiliary engines,
harbor craft, and
smaller OGV propulsion
3
Displacement > 30 liters per cylinder
OGV propulsion
* EPA assumes that all engines with a gross power below 37 kW are used for recreational applications
and are treated separately from the commercial marine category.
Table A-19. Marine Diesel Engine Speed Designations
Soeed Cateeorv
Engine Stroke Tvoe
Cateeorv
SSD
2
3
MSD
4
2.3
HSD
4
1
Information on auxiliary engine power is generally not complete in the Lloyd's data. As was done in the
C3 RIA, auxiliary engine power was estimated using a survey performed by California Air Resources
Board (ARB) in 2005.47 The survey provided average propulsion and auxiliary engine power by ship type
and is shown in Table A-20. Auxiliary to propulsion power ratios were calculated for each ship type from
the survey and used to calculate auxiliary power for all ships. All auxiliary power is assumed to be
provided by Category 2 medium speed diesel engines, consistent with the C3 RIA methodology.
47 California Air Resources Board, 2005 Oceangoing Ship Survey, Summary of Results, September 2005
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
24
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-20. Auxiliary Engine Power Ratios (ARB Survey, Except as Noted)
Ship Type
Average
Propulsion
Engine (kW)
Average Auxiliary Engines
Auxiliary to
Propulsion
Ratio
Number
Power
Each (kW)
Total Power
(kW)
Engine
Speed
Auto Carrier
10,700
2.9
983
2,850
MSD
0.266
Bulk Carrier
8,000
2.9
612
1,776
MSD
0.222
Container Ship
30,900
3.6
1,889
6,800
MSD
0.220
Passenger Ship3
39,600
4.7
2,340
11,000
MSD
0.278
General Cargo
9,300
2.9
612
1,776
MSD
0.191
Miscellaneous15
6,250
2.9
580
1,680
MSD
0.269
RORO
11,000
2.9
983
2,850
MSD
0.259
Reefer
9,600
4.0
975
3,900
MSD
0.406
Tanker
9,400
2.7
735
1,985
MSD
0.211
a Passenger ships typically use a different engine configuration known as diesel-electric. These vessels use large generator sets
for both propulsion and ship-board electricity. The figures for passenger ships above are estimates taken from the Starcrest
Vessel Boarding Program and were used for modeling purposes to split emissions between propulsion and auxiliary use.
b Miscellaneous ship types were not provided in the ARB methodology, so values from the Starcrest Vessel Boarding Program
were used. All tugs are considered as miscellaneous.
A.6.3. Emission Boundaries
The emission boundaries for the OGV sector were defined as any activity occurring within 5 km of the
National Emissions Inventory port shape files/8 with the following exceptions:
• New Orleans. Baton Rouge, and South Louisiana: The boundaries for these ports were extended
by 0.5 km beyond the port shape files because of the expected impact RSZ emissions would
have on the individual ports and to account for the RSZ emissions at the port boundaries.
• Port of Norfolk: The boundary for this port includes all activity (except that going to Newport
News) within 10 km of the Norfolk port shape files. The boundary was extended to 10 km to
capture the RSZ emissions near residential areas that were identified as important for this
analysis. One RSZ length was used for all ships.
• Port of Savannah: The boundary for this port is 10 km towards the ocean from the port shape.
The boundary was extended to 10 km to capture the RSZ emissions near residential areas. It is
truncated at the north end of the port.
• Port of Philadelphia: While the 5 km boundary includes the Port of Camden, hoteling and
maneuvering emissions from Camden were beyond the scope of this assessment and are not
considered here.
48 U.S. EPA, 2011 National Emissions Inventory Port Shape Files, August 2014. Available at:
http://www.epa.eov/ttn/chief/eis/2011nei/ports 20140729.zip
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
25
-------
Appendix A: Baseline Emission Inventory Methodology
A.6.4. Activity Modes
The following activity measurements and modes were analyzed in this assessment: number of calls,
reduced speed zones, maneuvering, hoteling, and anchorage.
A.6.4.1. Calls
The number of calls by C3 vessels at each port was determined from the USACE Entrances and
Clearances data. As was done in the C3 RIA, barges were removed from the data as these are non-
propelled vessels generally moved by a Category 1 or 2 tug. It is important to note that Entrances and
Clearances only record stops where foreign cargo was either loaded or discharged (i.e., where a ship
discharges goods but doesn't load goods, only the entrance was recorded). Thus, to get a better
estimate of calls, the maximum of either entrances or clearances for a given ship type, engine type, and
DWT bin is used in this analysis as a surrogate for calls. This approach differs slightly from the C3 RIA
methodology, which defined calls as the average of entrances and clearances for a given ship type,
engine type, and DWT bin.
A.6.4.2. Reduced Speed Zone
The reduced speed zone activity in this analysis included movements from the port entrance to the
boundary of the analysis. Port entrances were determined from the National Emission Inventory (N El)
shapefiles for the port. RSZ distances were measured using Google Earth. An example of this is shown in
Figure A-2, where the port shape in orange defines the port. The red line depicts the port boundary and
the yellow line defines the RSZ shipping lane within the boundary.
Figure A-2, Savannah RSZ Measurement
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
26
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-21 lists the RSZ distances and speeds for each port. For the Ports of Houston and Philadelphia,
"fast" ships are defined as auto carriers, container ships, passenger ships and ROROs. In addition, the
RSZ speed for the Port of Savannah is 13 knots; however, the cruise speed for some ships entering that
port is less than 13. The analysis for these ships used their cruise speed instead of the RSZ speed. The
Port of Houston RSZ is a special case as has two reduced speed zones to accommodate different speeds
within the Houston Ship Channel. Details for these calculations may be found in section A.6.9.
Table A-21. RSZ One-Way Distances and Speeds
Port Name
Distance (km)
Speed (knots)
RSZ Speed Source
Port of New York and New
Jersey
5.0
Containers -10.0
Tankers - 6.2
Others-8.0
2012 Port of New
York/New Jersey
Inventory
Port of New Orleans
28.2
10.0
C3 RIA
Port of Miami
5.0
12.0
C3 RIA
Port of South Louisiana
44.0
10.0
C3 RIA
Port of Seattle
5.6
Container - 14.0
Passenger - 15.0
Others-8.0
2011 Puget Sound
Inventory
Port of Baton Rouge
71.3
10.0
C3 RIA
Port Arthur
5.6
7.0
C3 RIA
Port of Portland
4.6
8.4
C3 RIA
Port of Mobile
5.0
11.0
C3 RIA
Port of Houston
14.1/10.4
Fast Ships -12.0
Slow Ships - 9.5
2007 Port of Houston
Inventory
16.7
Fast Ships -10.5
Slow Ships - 7.5
Port of Baltimore
7.9
Auto Carrier - 16.5
Bulk Carrier - 12.9
Container Ship -17.5
General Cargo -13.3
Passenger Ship -17.0
RoRo -14.9
Tanker - 13.0
2010 Port of Baltimore
Report
Port of Norfolk (Hampton Roads)
18.1
Container - 11.4
Other ships - 10.0
2011 Port of Virginia
Inventory
Port of Philadelphia
5.5
Fast Ships -11.0
Slow Ships - 9.0
C3 RIA
Port of Charleston
7.4
12.0
C3 RIA
Port of Corpus Christi
5.0
< 90,000 DWT - 12.0
> 90,000 DWT - 9.0
C3 RIA
Port Tampa Bay
5.0
9.0
C3 RIA
Port of Savannah
12.0
13.0
C3 RIA
Port of Coos Bay, OR
7.6
6.5
C3 RIA
Port of San Juan, PR
5.0
10.0
PR&VI ECA49
49 U.S. Environmental Protection Agency, Proposal to Designate an Emission Control Area for Nitrogen Oxides, Sulfur Oxides and
Particulate Matter, Technical Support Document, Report EPA-420-R-10-013, August 2010. Available at
http://www.epa.eov/otaq/rees/nonroad/marine/ci/420rl0013.pdf.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
27
-------
Appendix A: Baseline Emission Inventory Methodology
A.6.4.3. Maneuvering
Ships typically transition from RSZ to maneuvering speed as they begin to approach their berth. The
maneuvering distance is different for each port but is generally defined as slow speed activity within a
port. This can include movement from the port entrance to the berth, docking at the berth, and any
intra-port shifts that are shown in the matching typical port inventory. For purposes of computing
propulsion load factors and applying low speed adjustment factors, maneuvering is considered to occur
at stall speed (5.8 knots), which is consistent with the C3 RIA. Actual speeds are less due to starting,
stopping, and reversing that occur during maneuvering. In addition, most ships use assist tugs to push
them into and out of a berth.
Consistent with the C3 RIA, maneuvering times were taken from the matching typical port data. For Port
of New York/New Jersey, Port of Seattle, and Port of Charleston, maneuvering information from their
newer port inventories listed in Table A-15 were used. Details for these calculations may be found in
section A.6.9.
A.6.4.4. Hoteling
Hoteling is the time at berth when the vessel is operating auxiliary engines only or is using shore power.
Except when the vessel is using shore power, auxiliary engines are operating under load the entire time
the vessel is manned. Peak loads occur after the propulsion engines are shut down, when the auxiliary
engines are responsible for all onboard power and/or are used to power off-loading equipment.
Hoteling activity needs to be divided into auxiliary engine use and shore power to accurately account for
reduced emissions from shore power. No shore power was considered in the C3 RIA.
As with the C3 RIA, average hoteling times by ship type and DWT range were taken directly from the
matching typical port data. For Port of New York/New Jersey, Port of Seattle, and Port of Charleston,
hoteling information from their newer port inventories listed in Table A-15 were used. Additionally, Port
of Corpus Christi was updated with newer hoteling information provided by the port. Details for these
calculations may be found in section A.6.9.
A.6.4.5. Anchorage
Anchorage occurs when the ship has arrived at a port but no berth is available. Anchorage is only
considered if it occurs within the port boundaries and anchorage data are available. Shore power and
advanced marine emission control systems cannot be applied while at anchorage. While the C3 RIA
assumed all calls involve anchorage, it was assumed in this assessment that not all ship calls involve
anchorage. To estimate the percentage of calls in which anchorage occurs in the 1996 Patapsco River
Inventory, the data were reanalyzed to provide that percentage by ship type/engine type/DWT bin. This
represents the number of calls for which a ship type/engine type/DWT bin anchors within the port (but
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
28
-------
Appendix A: Baseline Emission Inventory Methodology
not at a berth) divided by the total calls for that bin. Anchorage time was calculated accordingly. This is
consistent with the newer Starcrest methodology used for Port of Los Angeles50 and Long Beach.51
A.6.4.6. Summary
The various activity modes for Category 3 vessels analyzed in this assessment are summarized in Table A-22.
Table A-22. Vessel Movements and Time-ln-Mode Descriptions
Summary Table Field
Description
Call
A call is one entrance and one clearance. Since the USACE Entrances and Clearances
data do not provide a record for an entrance where no foreign cargo discharged or a
record of a clearance where no foreign cargo is loaded at a port, the number of
entrances and clearances may not be the same. Therefore, the number of calls were
taken as the maximum of the entrances or clearances at a port as grouped by ship type,
engine type, and deadweight tonnage bin.
Reduced Speed Zone
(RSZ) (hr/call)
Time when a ship reduces speed before entering a port. This can be a long distance
down a river or channel and generally ends at the port entrance.
Maneuver (hr/call)
Time when a ship is being berthed or de-berthed, traveling to an anchorage or moving
between berths. Maneuvering is assumed to occur within the port area, generally
beginning and ending at the entrance of the port. This will include shifts within a port
area moving from one berth to another. For purposes of calculating load factors,
maneuvering was assumed to occur at an average speed of 5.8 knots. Maneuvering
times were taken from the typical port data or calculated from published inventories.
Hoteling (hr/call)
Hoteling is the time at berth when the vessel is operating auxiliary engines only or is
using shore power. Peak loads occur after the propulsion engines are shut down, when
the auxiliary engines or shore power is responsible for all onboard power and/or are
used to power off-loading equipment.
Anchorage (hr/call)
If the port data included anchorage, it is broken out separately for this analysis. Some
emission reduction techniques cannot be applied while at anchorage. This mode was
ignored if not specifically identified.
A.6.5. Load Factors
As in the C3 RIA, load factors are expressed as a percent of the vessel's total propulsion or auxiliary
power. At service or cruise speed, the propulsion load factor is assumed to be 83%. At lower
speeds, the Propeller Law is used to estimate ship propulsion loads, based on the theory that
propulsion power varies by the cube of speed as shown in the equation below.
50 Starcrest Consulting Group, Port of Los Angeles Air Emissions Inventory - 2011, July 2012.
51 Starcrest Consulting Group, Port of Long Beach Air Emissions Inventory-2011, July 2012.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
29
-------
Appendix A: Baseline Emission Inventory Methodology
LF = (AS/MS)3 Eq. A-2
Where
LF = Load Factor (percent),
AS = Actual Speed (knots), and
MS = Maximum Speed (knots).
Maximum speed is calculated from service speed, which is available in the Lloyd's data, as 1.00/0.94
multiplied by service speed as was done in the C3 RIA. While load factors were calculated using the
above propeller law, load factors below 2% were set to 2% as a minimum.52
Load factors for auxiliary engines vary by ship type and operating mode. Table A-23 shows the auxiliary
engine load factors determined by Starcrest, through interviews conducted with ship captains, chief
engineers, and pilots during its vessel boarding programs.53 These were used in the C3 RIA. Auxiliary load
factors listed in Table A-23 are used together with the total auxiliary engine power (determined from
total propulsion power and the ratios from Table A-20) to calculate auxiliary engine emissions.
Consistent with the C3 RIA, emissions from auxiliary boilers are not explicitly calculated here, but the
load factors presented below are large enough to include emissions from both auxiliary engines and
auxiliary boilers.
Table A-23. Auxiliary Engine Load Factor Assumptions
Ship Type
RSZ
Maneuver
Hotel Anchor
Auto Carrier
0.30
0.67
0.24
Bulk Carrier
0.27
0.45
0.22
Container Ship
0.25
0.50
0.17
Passenger Ship
0.80
0.80
0.64
General Cargo
0.27
0.45
0.22
Miscellaneous
0.27
0.45
0.22
RORO
0.30
0.45
0.30
Reefer
0.34
0.67
0.34
Tanker
0.27
0.45
0.67
Tug
0.27
0.45
0.22
52 Starcrest Consulting Group LLC, Update to the Commercial Marine Inventory for Texas to Review Emission Factors, Consider a
Ton-Mile El Method, and Revised Emissions for the Beaumont-Port Arthur Non-Attainment Area, prepared for the Houston
Advanced Research Center, January 2004.
53 Starcrest Consulting Group, Port-Wide Baseline Air Emissions Inventory, prepared for the Port of Los Angeles, June 2004.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
30
-------
Appendix A: Baseline Emission Inventory Methodology
A.6.6. Emission Factors
Emission factors vary by engine type, engine tier, fuel type, fuel sulfur levels, and load factor, for both
propulsion and auxiliary engines.
A.6.6.1. Propulsion Engine Emission Factors
Propulsion engine emission factors used in this analysis are shown in Table A-24. They come from a
European Commission study (referred to here as Entec)54 and are similar to the ones used in the C3 RIA.
For example, the brake specific fuel consumption (BSFC) for MSD in this analysis was 213 g/kWh for
residual oil (RO) and 203 g/kWh for marine distillate oil/marine gas oil (MDO/MGO) compared to the
210 g/kWh and 200 g/kWh, respectively, used in the C3 RIA. PMio emission factors for MSD engines
were also recalculated using the equations listed below that were determined based on existing engine
data in consultation with ARB.
RO PMio EF = 1.35 + BSFC x 7 x 0.02247 x (Fuel Sulfur Fraction - 0.0246) Eq. A-3
MDO & MGO PMio EF = 0.23 + BSFC x 7 x 0.02247 x (Fuel Sulfur Fraction - 0.0024) Eq. A-4
PMz.s EF = 0.92 x PMio EF Eq. A-5
The above equations are based upon the fact that the sulfate component in PMio has a molecular
weight seven times that of sulfur and the assumption that 2.247% of the fuel sulfur is converted to PMio
sulfate. PM2.5 was assumed to be 92% of PMio. These assumptions and formulas were used in the C3
RIA.
For S02, the emission factors were based upon a fuel sulfur to S02 conversion factor from ENVIRON,
assuming that 97.753% of the fuel sulfur was converted to S02 and taking into account the molecular
weight difference between S02 and sulfur.55
SO2 EF = BSFC x 2 x 0.97753 x Fuel Sulfur Fraction Eq. A-6
C02 emission factors were calculated from the BSFC assuming a fuel carbon content of 86.7% by
weight56 and a ratio of molecular weights of C02 and C at 3.667.
C02 EF = BSFC x 0.867x3.667 Eq. A-7
54 Entec UK Limited, Quantification of Emissions from Ships Associated with Ship Movements between Ports in the European
Community, prepared for the European Commission, July 2002.
55 Memo from Chris Lindhjem of ENVIRON, PM Emission Factors, December 15, 2005.
56 Entec UK Limited, Quantification of Emissions from Ships Associated with Ship Movements between Ports in the European
Community, prepared for the European Commission, July 2002.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
31
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-24. Emission Factors for Tier 0 OGV Propulsion Engines, g/kWh
Engine
Type
Fuel Type
Fuel
Sulfur
Emission Factors (g/kWh)
NOx
PMio
PM2.5
HC
CO
SO2
CO2
BSFC
SSD
RO
2.70%
18.1
1.42
1.31
0.6
1.4
10.29
621
195
MGO/MDO
0.10%
17.0
0.19
0.17
0.6
1.4
0.36
589
185
MSD
RO
2.70%
14.0
1.43
1.32
0.5
1.1
11.24
678
213
MGO/MDO
0.10%
13.2
0.19
0.17
0.5
1.1
0.40
646
203
GT
RO
2.70%
6.1
1.47
1.35
0.1
0.2
16.10
971
305
MGO/MDO
0.10%
5.7
0.17
0.15
0.1
0.2
0.57
923
290
ST
RO
2.70%
2.1
1.47
1.35
0.1
0.2
16.10
971
305
MGO/MDO
0.10%
2.0
0.17
0.15
0.1
0.2
0.57
923
290
The IMO adopted mandatory Tier I NOx emission limits in Annex VI to the International Convention for
Prevention of Pollution from Ships in 1997. These NOx limits apply for all marine engines over 130 kW for
engines built on or after January 1, 2000, including those engines that underwent a major rebuild after
January 1, 2000. For the C3 RIA, the effect of the IMO standard was determined to be a reduction in NOx
emission rate of 11% below that for engines built before 2000. Therefore, for engines built between 2000
and 2010 (Tier I), a factor of 0.89 was applied to the calculation of NOx emissions for both propulsion and
auxiliary engines.57
IMO Tier II NOx emission standards started in 2011. For the C3 RIA, the effect of Tier II was determined to be
a NOx reduction of 2.5 g/kWh reduction over Tier I engines. Tier III took effect in 2016 and EPA determined in
the C3 RIA that the effect of Tier III to be an 80% reduction from Tier I. Thus Tier III emission factors are 20%
of Tier I emission factors. All emission factors used here are consistent with the C3 RIA.
In addition to the MARPOL Annex VI emission limits that apply to all ships engaged in international
transportation, U.S. vessels must also comply with EPA's Clean Air Act requirements for engines and fuels.
The NOx emission limits for Category 3 engines are equivalent to the MARPOL Annex VI NOx limits. EPA's
sulfur limit for distillate locomotive or marine (LM) diesel fuel sold in the United States is more stringent than
the ECAfuel sulfur limit; the sulfur limit for ECAfuel for use on Category 3 marine vessels is equivalent to the
MARPOL Annex VI SOx limits. EPA also has standards for C3 engines,58 which are generally the same or more
stringent but almost all C3 engines used in international shipping fall under IMO regulations.
57 In addition, as part of the IMO standards, marine diesel engines built between 1990 and 1999 that are >90 liters per cylinder
need to be retrofit to meet Tier I emission standards upon engine rebuild if a retrofit kit is available. Consistent with the C3
RIA, it was assumed that 80% of these ships will have retrofit kits available and that this phase-in will happen over five years,
with 20% of eligible ships each year starting in 2011. Since the 2011 phase-in represents less than 0.4% of NOx emissions by
ships at the 19 ports, no engines were assumed to be rebuilt in the 2011 inventory for calculation purposes.
58 U.S. Environmental Protection Agency, Control of Emissions from New Marine Compression-Ignition Engines at or Above 30
Liters per Cylinder, Federal Register, Vol 75, No 83, April 30, 2010.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
32
-------
Appendix A: Baseline Emission Inventory Methodology
As with the C3 RIA, emission factors are considered to be constant down to about 20% load. Below that
threshold, emission factors tend to increase as the load decreases. This is because diesel engines are less
efficient at low loads and the BSFC tends to increase. Thus, while mass emissions (grams per hour) decrease
with low loads, the engine power tends to decrease more quickly, thereby increasing the emission factor
(grams per engine power) as load decreases. Energy and Environmental Analysis Inc. demonstrated this
effect in a study prepared for EPA in 2000.59
Low-load multiplicative adjustment factors used in the C3 RIA are presented in Table A-25. As these
adjustment factors were derived for diesel engines, the low load adjustment factors should only be applied
to MSD and SSD engines. This is a modification of the C3 RIA methodology where low load adjustment
factors were also applied to steam turbine engines. However, since the boiler that drives the steam turbines
also drives the auxiliary engines, the total load on the boiler (propulsion and auxiliary load) is higher than that
the propulsion steam turbine load only. Thus, it is assumed that they are always higher than 20% and
therefore no low load factor should be applied.
Table A-25. Calculated Propulsion Engine Low Load Multiplicative Adjustment Factors
Load
NOx
HC
CO
PM
S02
CO2
2%
4.63
21.18
9.68
7.29
3.36
3.28
3%
2.92
11.68
6.46
4.33
2.49
2.44
4%
2.21
7.71
4.86
3.09
2.05
2.01
5%
1.83
5.61
3.89
2.44
1.79
1.76
6%
1.60
4.35
3.25
2.04
1.61
1.59
7%
1.45
3.52
2.79
1.79
1.49
1.47
8%
1.35
2.95
2.45
1.61
1.39
1.38
9%
1.27
2.52
2.18
1.48
1.32
1.31
10%
1.22
2.20
1.96
1.38
1.26
1.25
11%
1.17
1.96
1.79
1.30
1.21
1.21
12%
1.14
1.76
1.64
1.24
1.18
1.17
13%
1.11
1.60
1.52
1.19
1.14
1.14
14%
1.08
1.47
1.41
1.15
1.11
1.11
15%
1.06
1.36
1.32
1.11
1.09
1.08
16%
1.05
1.26
1.24
1.08
1.07
1.06
17%
1.03
1.18
1.17
1.06
1.05
1.04
18%
1.02
1.11
1.11
1.04
1.03
1.03
19%
1.01
1.05
1.05
1.02
1.01
1.01
20%
1.00
1.00
1.00
1.00
1.00
1.00
Low load adjustment factors were not applied to diesel electric drive systems for loads below 20%
because in these systems, multiple engines are used to generate power, and some can be shut down to
allow others to operate at a more efficient setting.
59 Energy and Environmental Analysis Inc., Analysis of Commercial Marine Vessels Emissions and Fuel Consumption Data,
EPA420-R-00-002, February 2000.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
33
-------
Appendix A: Baseline Emission Inventory Methodology
A.6.6.2. Auxiliary Engine Emission Factors
As with propulsion engines, the auxiliary engine emission factors used in this analysis comes from
Entec60 and was used in the C3 RIA. However, the BSFC used in this analysis differs from what was used
in the C3 RIA. The BSFC was updated in this analysis to match the 227 g/kWh and 217 g/kWh for RO and
MGO/MDO, respectively, to match that listed by Entec for auxiliary engines. This is instead of the 210
g/kWh and 200 g/kWh, respectively, that was used in the C3 RIA.
An ARB survey published in 200561 found that almost all ships used RO in their main propulsion engines.
Only 29% of all ships (except passenger ships) used distillate (MGO/MDO) in their auxiliary engines, with
the remaining 71% using RO. Only 8% of passenger ships used distillate in their auxiliary engines, while
the other 92% used RO.
Distillate fuels discussed in the ARB survey ranged from 0.03% to 1.5% sulfur in those ships that used
distillate in their auxiliaries. For the purposes of this analysis, the use of 1.0% sulfur distillate fuel in
these engines was assumed for the 2011 baseline estimate. The C3 RIA assumed 1.5% sulfur distillate
but calculated that the difference between using 1.0% versus 1.5% did not result in significant
overestimation of emissions. The value used in this analysis is consistent with that used by Entec for
global distillate sulfur levels. Note that no distinction was made here between MGO and MDO, and they
are referred to here as "MGO/MDO."
The equations listed above in section A.6.6.1 were used to calculate the PMio, PM2.5, S02, and C02
emission factors for this analysis based upon the different fuel sulfur levels and BSFCs as mentioned
above. All other factors match the C3 RIA values. Table A-26 provides these auxiliary engine emission
factors. Consistent with the C3 RIA assumptions, there is no need for a low load adjustment factor for
auxiliary engines, because they are generally operated in banks. When only low loads are needed, one
or more engines are shut off, allowing the remaining engines to operate at a more efficient level.
Table A-26. Tier 0 Auxiliary Engine Emission Factors by Fuel Type, g/kWh
Fuel Type
Fuel Sulfur
Emission Factors (g/kWh)
NOx
PMio
PM2.5
HC
CO
S02
CO2
BSFC
RO
2.70%
14.7
1.44
1.32
0.4
1.10
11.98
723
227
MGO/MDO
1.00%
13.9
0.49
0.45
0.4
1.10
4.24
668
217
MGO/MDO
0.10%
13.9
0.18
0.17
0.4
1.10
0.42
691
217
Using the percentages of RO and Distillate found in the ARB survey for passenger ships and other ships,
Table A-27 provides weighted emission factors for the two ship types for use in the analyses.
60 Entec UK Limited, Quantification of Emissions from Ships Associated with Ship Movements between Ports in the European
Community, Table 2.10, prepared for the European Commission, July 2002.
61 California Air Resources Board, 2005 Oceangoing Ship Survey, Summary of Results, September 2005.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
34
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-27. Tier 0 Auxiliary Engine Emission Factors by Ship Type, g/kWh
Ship Type
Fuel Sulfur
Emission Factors (g/kWh)
NOx
PM10
PM2.5
HC
CO
S02
CO2
BSFC
Passenger
2.56%
14.6
1.36
1.25
0.4
1.1
11.36
718
226
0.10%
13.9
0.18
0.17
0.4
1.1
0.42
691
217
Other
2.21%
14.5
1.16
1.07
0.4
1.1
9.74
707
224
0.10%
13.9
0.18
0.17
0.4
1.1
0.42
691
217
A.6.6.3. Fuel Sulfur Levels
Where RO is used in propulsion or auxiliary engines, it is assumed to be 2.7% sulfur for all ports for the
2011 baseline analysis. In the C3 RIA, West Coast ports were assumed to use 2.5% sulfur RO instead of
2.7% sulfur. Using 2.7% sulfur RO in Washington and Oregon ports is consistent with the 2011 Puget
Sound inventory62 where 2.7% sulfur RO was assumed.
A.6.6.4. Black Carbon
BC is the light-absorbing component of particulate matter and is formed by the incomplete combustion
of carbon-based fuels. Like C02, BC is a global warming pollutant. EPA's Report to Congress on Black
Carbon63 lists 0.03 as the BC/PM2.5 factor to use for C3 commercial marine vessels for all engine types
and fuels for 2011.
A.6.6.5. Treatment of Electric-Drive Ships
Many passenger ships and tankers have either diesel-electric or gas turbine-electric engines that are
used for both propulsion and auxiliary purposes. Lloyd's identifies these types of engines in its database
and that information was used to distinguish them from direct and geared drive systems for this
analysis. Generally, the power Lloyd's lists is the total power for electric drive vessels. To separate out
propulsion from auxiliary power for purposes of calculating emissions, the total power listed in the
Lloyd's data was divided by one plus the ratio of auxiliary to propulsion power given in Table A-20 to
give the propulsion power portion and the remaining portion was considered auxiliary engine power.64
In addition, no low load adjustment factor was applied to diesel electric engines for loads below 20%
because several engines are used to generate power, and some can be shut down to allow others to
operate at a more efficient setting. This same methodology was used for the calculations in the C3 RIA.
62 Starcrest Consulting Group, 2011 Puget Sound Maritime Emission Inventory, September 2012.
http://www.pugetsoundmaritimeairforum.org/uploads/PV FINAL POT 2011 PSEI Report Update 23 May 13 sce.pdf
63 U.S. Environmental Protection Agency, Report to Congress on Black Carbon, EPA-450/R-12-001, March 2012, p. 87.
64 ICF International, Commercial Marine Port Development - 2002 and 2005lnventories, September 2007.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
35
-------
Appendix A: Baseline Emission Inventory Methodology
A.6.7. Bin Mismatches
In some cases, the ship type/engine type/DWT range bin in the modeled port was not provided in the
typical port inventory. In that case, the nearest match to the bin at the modeled port was used from the
typical port inventory. This same methodology of using near bins was used in development of the C3
RIA. Table A-28 shows the typical port bins used when there was no exact match for the bin at the
modeled port. While the same methodology was used, it is possible that the bin matched for the C3 RIA
might be different to the bin matched in this analysis.
Table A-28. Typical Port Bin Mismatches
Modeled Port Bin
Typical Port Bin
Port
Ship Type
Engine
Type
DWT Range
Ship Type
Engine
Type
DWT Range
CONTAINER SHIP
SSD
> 90,000
CONTAINER SHIP
SSD
45,000 - 90,000
Port of New
PASSENGER
MSD-ED
10,000 - 20,000
PASSENGER
MSD-ED
< 10,000
Orleans
TANKER
MSD-ED
< 30,000
TANKER
MSD
< 30,000
30,000 - 60,000
30,000 - 60,000
CONTAINER SHIP
SSD
35,000 - 45,000
CONTAINER SHIP
SSD
25,000 - 35,000
45,000 - 90,000
Port of Miami
MSD-ED
< 10,000
PASSENGER
10,000 - 20,000
PASSENGER
MSD
< 10,000
GT-ED
10,000 - 20,000
BULK CARRIER
MSD
45,000 - 90,000
BULK CARRIER
SSD
45,000 - 90,000
> 90,000
> 90,000
Port of
CONTAINER SHIP
SSD
35,000 - 45,000
CONTAINER SHIP
SSD
25,000 - 35,000
Mobile
45,000 - 90,000
MSD
45,000 - 90,000
MISCELLANEOUS
MSD
All
MISCELLANEOUS
SSD
All
MSD-ED
All
AUTO CARRIER
SSD
> 30,000
AUTO CARRIER
SSD
20,000 - 30,000
Port of
CONTAINER SHIP
SSD
> 90,000
CONTAINER SHIP
SSD
45,000 - 90,000
Baltimore
PASSENGER
SSD
< 10,000
PASSENGER
ST
< 10,000
RORO
MSD
10,000 - 20,000
RORO
MSD
20,000 - 30,000
Port of
CONTAINER SHIP
SSD
> 90,000
CONTAINER SHIP
SSD
45,000 - 90,000
Norfolk
PASSENGER
MSD-ED
10,000 - 20,000
PASSENGER
MSD-ED
< 10,000
(Hampton
SSD
< 10,000
ST
< 10,000
Roads)
TANKER
SSD
120,000 -150,000
TANKER
SSD
> 150,000
AUTO CARRIER
SSD
> 30,000
AUTO CARRIER
SSD
20,000 - 30,000
Port of
Philadelphia
CONTAINER SHIP
SSD
35,000 - 45,000
CONTAINER SHIP
SSD
25,000 - 35,000
45,000 - 90,000
RORO
MSD
10,000 - 20,000
RORO
MSD
< 10,000
SSD
20,000 - 30,000
SSD
10,000 - 20,000
AUTO CARRIER
SSD
20,000 - 30,000
AUTO CARRIER
SSD
10,000 - 20,000
> 30,000
Port of
BULK CARRIER
MSD-ED
45,000 - 90,000
BULK CARRIER
SSD
45,000 - 90,000
Corpus Christi
MISCELLANEOUS
MSD-ED
All
MISCELLANOUS
MSD
All
TANKER
ST
60,000 - 90,000
TANKER
ST
30,000 - 60,000
TUG
MSD
All
MISCELLANEOUS
MSD
All
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
36
-------
Appendix A: Baseline Emission Inventory Methodology
Port
Modeled Port Bin
Typical Port Bin
Ship Type
Engine
Type
DWT Range
Ship Type
Engine
Type
DWT Range
Port Tampa
Bay
CONTAINER SHIP
SSD
45,000 - 90,000
CONTAINER SHIP
MSD
45,000 - 90,000
MISCELLANEOUS
MSD-ED
All
MISCELLANEOUS
SSD
All
PASSENGER
MSD-ED
< 10,000
PASSENGER
MSD
< 10,000
GT-ED
10,000 - 20,000
REEFER
SSD
20,000 - 30,000
REEFER
SSD
10,000 - 20,000
RORO
MSD
10,000 - 20,000
RORO
SSD
10,000 - 20,000
Port of
Savannah
AUTO CARRIER
SSD
> 30,000
AUTO CARRIER
SSD
20,000 - 30,000
CONTAINER SHIP
SSD
> 90,000
CONTAINER SHIP
SSD
45,000 - 90,000
TANKER
MSD-ED
60,000- 90,000
TANKER
MSD
30,000 - 60,000
SSD
120,000- 150,000
SSD
> 150,000
ST
60,000- 90,000
ST
30,000 - 60,000
90,000- 120,000
Port of San
Juan
CONTAINER SHIP
SSD
35,000 - 45,000
CONTAINER SHIP
SSD
25,000 - 35,000
SSD
45,000 - 90,000
MSD
45,000 - 90,000
ST
25,000 - 35,000
SSD
25,000 - 35,000
MISCELLANEOUS
ST
All
MISCELLANEOUS
SSD
All
PASSENGER
MSD-ED
10,000- 20,000
PASSENGER
MSD
< 10,000
SSD
< 10,000
ST
< 10,000
GT-ED
10,000- 20,000
ST
10,000 - 20,000
A.6.8. Matching Lloyd's Data to USACE Entrances and Clearance Data
To match activity data with the correct emission factors, the Entrances and Clearances data were
matched to Lloyd's Register of Ships. The Entrances and Clearances data65 contain the following
information for each major port or waterway:
¦ date a vessel made entry into (entrance record) or cleared (clearance record) the U.S. Customs port;
¦ vessel's full name;
¦ type of vessel by one digit rig type or International Classification of Ships by Type (ICST) code;
¦ vessel's flag of registry;
¦ vessel's previous (entrance record) or next (clearance record) port of call, whether the port was
foreign or domestic;
¦ vessel's Net and Gross Registered Tonnage;
¦ vessel's draft (feet); and
¦ vessel's International Maritime Organization number.
65 Available at http://www.navieationdatacenter.us/data/dataclen.htm.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
37
-------
Appendix A: Baseline Emission Inventory Methodology
Since this does not contain any call time-in-mode information, average time in mode and speeds need to
be used with the USACE data to estimate emissions at ports (see section A.6.4).
It is important to note that these data only represent foreign cargo movements, and does not account
for U.S. ships delivering domestic cargo from one U.S. port to another U.S. port (covered under the
Jones Act66). However, U.S. flagged ships carrying foreign cargo from a foreign port to a U.S. port or from
a U.S. port to a foreign port are accounted for in the data as these are considered foreign cargo
movements. At most ports, domestic commerce is carried out by Category 2 ships, although there are a
few exceptions, particularly on the West Coast. A study by ICF found that the USACE Entrances and
Clearances data accounted for more than 90% of the emissions from Category 3 ships calling on U.S.
ports, so neglecting Jones Act ships is assumed to be small.67
Additionally, the Entrances and Clearances data do not always match MEPA data because Entrances and
Clearances include cargo movements from both public and private terminals at a port while the MEPA
data usually only cover calls at public terminals, as Port Authorities generally do not have jurisdiction
over private terminals.
Entrances and Clearances data for 2011 contained over 100,000 individual entrances or clearances by
ships, tugs, and barges for U.S. ports or waterways. The ports of interest were matched to Entrances and
Clearances data as shown in Table A-29.
Table A-29. Corresponding USACE Port Names
Port Name
USACE Port #
USACE Name
Port of New York and New Jersey
398
Consolidated Port of New York
Port of New Orleans
2251
Port of New Orleans, LA
Port of Miami
2164
Miami Harbor, FL
Port of South Louisiana
2253
Port of South Louisiana (LA)
Port of Seattle
4722
Seattle Harbor, WA
Port of Baton Rouge
2252
Port of Baton Rouge, LA
Port Arthur
2416
Port Arthur, TX
Port of Portland
4644
Port of Portland, OR
Port of Mobile
2005
Mobile Harbor, AL
Port of Houston
2012
Houston Ship Channel, TX (Houston, TX)
Port of Baltimore
700
Baltimore Harbor and Channels, MD
Port of Hampton Roads (Norfolk)
744
Elizabeth River, VA
Port of Philadelphia
552
Philadelphia Harbor, PA
Port of Charleston
773
Charleston Harbor, SC
Port of Corpus Christi
2414
Corpus Christi, TX
Port Tampa Bay
2021
Tampa Harbor, FL
Port of Savannah
776
Savannah Harbor, GA
Port of Coos Bay, OR
4660
Coos Bay, OR
Port of San Juan, PR
2136
San Juan Harbor, PR
66 Merchant Seaman Protection and Relief 46 USCS Appx § 688 (2002) Title 46. Appendix. Shipping Chapter 18.-
67 ICF International, Inventory Contribution of U.S. Flagged Vessels, June 2008.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
38
-------
Appendix A: Baseline Emission Inventory Methodology
As done in the C3 RIA, barges that are not self-propelled were removed from the data. These are indicated by
the Rig field, where Rig = 4 indicates a dry barge while Rig = 5 represents a liquid barge. All barges that were
not part of an integrated tug-barge (ITB) were eliminated.
After eliminating barges and calls at ports outside the scope of this assessment, there were over 77,000
records for 7,696 vessels. Of these vessels, almost 99% could be matched to Lloyd's data using IMO number
and ship name. 18 vessels had incorrect IMO numbers, 15 of which were matched by IMO number and gross
tons or by vessel name and gross tons. 86 vessels did not have IMO numbers; 21 of these were matched by
vessel name and gross tons or by vessel name and ship type. This left 68 unmatched vessels, of which 60
were known to be less than 100 gross tons.
Of the 7,696 unique vessels, 93% were determined to be Category 3 by calculating cylinder displacement
from bore size and stroke length. 0.2% were steam turbine driven and 0.1% were gas turbine driven.68 Of the
Category 3 engines, gas turbines, and steam turbines, 6,459 were slow speed diesels (SSD) 2-stroke engines,
607 were medium speed diesels (MSD) 4-stroke engines, 97 were medium speed diesels with electric drive
(MSD-ED), nine were gas turbine electric drive (GT-ED), and 16 were steam turbine driven (ST).
After determining the engine type for each vessel in the Entrances and Clearances data, the vessels were
assigned a ship type based upon Lloyd's categorization as shown in Table A-30.
Table A-30. Corresponding Ship Types
Ship Type
Lloyd's Ship Type
Auto Carrier
Vehicles Carrier
Barge Carrier69
Bulk Carrier
Bulk Carrier (with Vehicle Decks)
Bulk Carrier, Laker Only
Bulk Carrier, Self-discharging
Cement Carrier
Bulk Carrier
Fish Carrier
Fish Factory Ship
Heavy Load Carrier
Heavy Load Carrier, semi-submersible
Livestock Carrier
Ore Carrier
Rail Vehicles Carrier
Wood Chips Carrier
Container Ship (Fully Cellular with RORO Facility)
Container Ship
Container Ship (Fully Cellular)
Container/RORO Cargo Ship
68 While steam and gas turbines are not diesel engines and thus do not fall into normal Categories 1, 2, or 3, they were included
in the analysis as OGV propulsion engines.
69 Barge carriers were originally separated in the C3 RIA but are such a small category that they are combined here with bulk
carriers.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
39
-------
Appendix A: Baseline Emission Inventory Methodology
Ship Type
Lloyd's Ship Type
General Cargo Ship
General Cargo
General Cargo Ship (with RORO facility)
Open Hatch Cargo Ship
Palletized Cargo Ship
Anchor Handling Tug Supply
Offshore Support Vessel
Pipe Layer
Pipe Layer Crane Vessel
Miscellaneous
Research Survey Vessel
Trailing Suction Hopper Dredger
Training Ship
Trawler
Yacht Carrier, semi-submersible
Passenger
Passenger/Cruise
Passenger/RORO Ship (Vehicles)
Reefer
Fruit Juice Carrier, Refrigerated
Refrigerated Cargo Ship
RORO
Logistics Vessel (Naval RORO Cargo)
RORO Cargo Ship
Asphalt/Bitumen Tanker
Bulk/Oil Carrier (OBO)
Chemical Tanker
Chemical/Products Tanker
Combination Gas Tanker (LNG/LPG)
Crude Oil Tanker
Tanker
Crude/Oil Products Tanker
FPSO, Oil
LNG Tanker
LPG Tanker
LPG/Chemical Tanker
Molten SulfurTanker
Products Tanker
Shuttle Tanker
Tug
Articulated Pusher Tug
Tug
The final step in the matching process was to fill data gaps. For example, three ships did not have service
speeds. The average value for the Lloyd's ship type was used for those three vessels as was done in the
C3 RIA. Finally, DWT ranges were assigned to the various vessels as was done in 2002 in the C3 RIA.
A.6.9. Additional Details for RSZ, Maneuvering, and Hoteling Calculations
The Port of Houston has two reduced speed zones to accommodate different speeds within the Houston
Ship Channel. These are shown in Figure A-3. The first is in Galveston Bay starting at Barbours Cut (see
red line). The longer distance is for those ships coming from the ship channel past Barbours Cut while
the shorter distance is for those ships leaving Bayport. The second is from where the Houston Ship
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
40
-------
Appendix A: Baseline Emission Inventory Methodology
Channel widens around Channelview to Barbours' Cut (see green line). These two RSZs were combined
in the C3 RIA but were separated out in this analysis to provide more detail.
Figure A-3. Houston Reduced Speed Zones70
gpTte
Atkin, on Island;
Bayport
It is assumed that all ships except for container ships and passenger ships enter the Houston ship
channel and head toward the Turning Basin. They travel over both RSZ 1 and RSZ 2 as well as maneuver
to and from the Turning Basin. Passenger ships are assumed to all stop at Bayport, so they only travel
down a portion of RSZ 1 and then maneuver into Bayport. Most container ships stop at either Bayport or
Barbours Cut but some smaller container ships head toward the turning basin. Those that stop at
Barbours Cut travel all the way down RSZ 1 but only 0.7 km down RSZ 2 and then maneuver into
Barbours Cut. The distribution of container ships that stop at Bayport, Barbours Cut, and the Turning
Basin are shown in Table A-31. These were determined from the Entrances and Clearances data and
Lloyd's data by examining the operating company for each ship and assigning it to one of the stops
based upon the Port of Houston website.71
Table A-31. Distribution of Container Ship Stops at Port of Houston
DWT
Barbours Cut
Bayport
Turning Basin
< 25,000
4.3%
65.7%
30.1%
25,000 - 35,000
55.6%
44.4%
0.0%
35,000 - 45,000
60.6%
39.4%
0.0%
45,000 - 90,000
24.9%
75.1%
0.0%
> 90,000
0.0%
100.0%
0.0%
70 Map data: Google.
71 Available at: http://www.portofhoustori.com/container-terminals.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
41
-------
Appendix A: Baseline Emission Inventory Methodology
Maneuvering and hoteling times for the Ports of Seattle and New York/New Jersey came from updated
port inventories that did not provide these data for the same bins used for the other ports.72 This
information was calculated for the Ports of Seattle and New York/New Jersey from movement data.
The movement data taken from the two inventories are shown in Table A-32.
Table A-32. Arrivals, Departures, and Shifts for Ports of Seattle and New York/New Jersey from Published Inventories
Port
Ship Type
Bin
Arrivals
Departures
Shifts
Bulk
All
2
76
105
1000 TEU
108
116
15
2000 TEU
73
89
24
3000 TEU
29
30
1
4000 TEU
111
115
6
Container
5000 TEU
124
126
5
6000 TEU
17
68
51
Seattle
7000 TEU
83
81
1
8000 TEU
130
135
8
9000 TEU
2
2
0
10000 TEU
1
1
0
General Cargo
All
14
39
26
Passenger
All
196
196
0
Tanker
All
0
5
0
Tug
All
0
1
4
Auto Carrier
All
266
265
93
Bulk Carrier
All
59
58
75
Container
All
2,033
2,032
71
New York/New
General Cargo
All
30
29
8
Jersey
Passenger
All
97
97
0
Reefer
All
46
46
2
RORO
All
90
91
55
Tanker
All
76
76
126
To calculate maneuvering time at the two ports, calls were defined as the maximum of published
arrivals and departures as used in this analysis. Estimated port entrance to berth distances (one-way
maneuvering distance) and shift distances were estimated using Google Earth. Berthing times were
estimated at 0.5 hours. These assumptions are shown in Table A-33 for the two ports. An average
maneuvering speed of 4 knots was used in the calculations to simulate the stop/start nature of
72 Maneuvering times for Port of Charleston also came from an updated inventory; however, these were listed in the published
inventory documentso no calculations for maneuvering time needed to be done.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
42
-------
Appendix A: Baseline Emission Inventory Methodology
maneuvering in a port area. The maneuvering time calculated just includes maneuvering movements.
No anchorage time is included because it is not provided in the two inventories.
Table A-33. Estimated Distances and Times for Maneuvering Calculations
Port
Item
Value
Port of Seattle
One way maneuvering distance
1.8 nm
Berthing time
0.5 hrs per call
Shifting Distance
1.75 nm
Port of New York and New
Jersey
One way maneuvering distance
4.0 nm
Berthing time
0.5 hr per call
Shifting Distance
7.25 nm
The information in Table A-32 is used together with the published inventory arrival, departure, and shift
information in Table A-33 to calculate maneuvering times using the equation below:
Maneuvering Time (hr/call) = ((Arr+Dep)*OWMD/4 + Calls*BT+S*SD/4)/Calls Eq. A-8
Where
Arr = number of arrivals,
Dep = number of departures,
OWMD = one-way maneuvering distance in nm,
BT = berthing time in hours,
S = number of shifts,
SD = average shift distance in nm,
Calls = maximum of arrivals and departures, and
4 = average maneuvering speed in knots.
Both the one-way maneuvering distance multiplied by the sum of arrivals and departures and the
number of shifts times the average shift distance are divided by the average maneuvering speed of 4
knots to obtain hours of maneuvering. Berthing time in hours per call is then multiplied by calls and
added to the maneuvering total. The total is then divided by the number of calls to determine
maneuvering time per call in hours. Calculated maneuvering times are shown in Table A-34 for the two
ports.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
43
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-34. Calculated Maneuvering Times for Port of Seattle
Port
Ship Type
Bin
Maneuvering Time (hrs)
Bulk
All
1.57
1000 TEU
1.43
2000 TEU
1.44
3000 TEU
1.40
4000 TEU
1.41
Container
5000 TEU
1.41
6000 TEU
1.39
Port of Seattle
7000 TEU
1.39
8000 TEU
1.41
9000 TEU
1.40
10000 TEU
1.40
General Cargo
All
1.40
Passenger
All
1.40
Tanker
All
0.95
Tug
All
2.70
Auto Carrier
All
1.55
Bulk Carrier
All
1.95
Container
All
1.42
Port of New York/New
General Cargo
All
1.50
Jersey
Passenger
All
1.40
Reefer
All
1.42
RORO
All
1.66
Tanker
All
2.13
Maneuvering times for container ships for the Port of Seattle had to be translated from the TEU bins listed in
the published inventory to the DWT bins used in this analysis. To accomplish this, container ships at Port of
Seattle in the Entrances and Clearances data were defined both ways and weighting factors determined for
each category. These weighting factors are shown in Table A-35. For example, to calculate maneuvering and
hoteling time for the 25,000-35,000 DWT bin, maneuvering and hoteling times for 1000 TEU and 2000 TEU
container ships were weighted by 35.8% and 64.2%, respectively.
Table A-35. Container Ship Size Translations for Port of Seattle
DWT Range
TEUs
Percent
< 25,000
1000
100.0%
25,000 - 35,000
1000
35.8%
2000
64.2%
35,000 - 45,000
2000
91.6%
3000
8.4%
45,000 - 90,000
2000
0.0%
3000
0.9%
4000
33.9%
5000
47.8%
6000
17.3%
7000
0.0%
> 90,000
6000
1.1%
7000
37.1%
8000
59.6%
9000
1.8%
10000
0.4%
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
44
-------
Appendix A: Baseline Emission Inventory Methodology
The result of weighting the maneuvering times shown in Table A-34 by the weights in Table A-35 is
shown in Table A-36. There was no container ship detail on maneuvering times (by TEU) in the New
York/New Jersey inventory.
Table A-36. Translated Container Ship Maneuvering Times for Port of Seattle
DWT Bin
Maneuver Time (hr)
< 25,000
1.43
25,000 - 35,000
1.43
35,000 - 45,000
1.43
45,000 - 90,000
1.40
> 90,000
1.40
Hoteling time was also taken from the Ports of Seattle, New York/New Jersey, and Charleston published
inventories and translated into the DWT bins used in this analysis. Hoteling times for Port of Seattle and
Port of New York/New Jersey are provided in Table A-37.
Table A-37. Published Hoteling Times for the Ports of Seattle and New York/New Jersey
Ship Type
Bin
Average Hoteling Time (hrs)
Seattle
New York
Auto Carrier
All
nia
15
Bulk
All
88.0
35
1000 TEU
24.2
18
2000 TEU
30.3
16
3000 TEU
31.8
22
4000 TEU
31.6
20
Container
5000 TEU
30.0
24
6000 TEU
28.7
30
7000 TEU
27.8
36b
8000 TEU
38.3
41
9000 TEU
33.3
40
10000 TEU
32.1
ni
General Cargo
All
30.1
14
Passenger
All
10.1
10
Chemical
N/A
29
Tanker
HandySize
ni
2
PanaMax
ni
5
Tug
All
110.2
ni
a ni = not included; N/A = not available
b Value not listed for 7000 TEU. Taken as the average of 6000 TEU and 8000 TEU.
The Seattle inventory shows a value of "N/A" for tanker ship hoteling time. Since tankers represent a
small portion of the Seattle inventory, no tanker hoteling emissions were calculated for Seattle in this
analysis.
Hoteling times for container ships for both ports and tanker ships for New York/New Jersey were
translated to the DWT bins used in this analysis. To accomplish this for the Port of New York/New Jersey,
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
45
-------
Appendix A: Baseline Emission Inventory Methodology
container and tanker ships in the Entrances and Clearances data for that port were processed as
described above for Seattle. The results are given below in Table A-38.
Table A-38. Container and Tanker Ship Size Translations for Port of New York/New Jersey
Ship Type
DWT Range
TEUs
Percent
< 25,000
1000
100.0%
25,000 - 35,000
1000
35.8%
2000
64.2%
35,000 - 45,000
2000
91.6%
3000
8.4%
3000
0.9%
Container Ship
45,000 - 90,000
4000
33.9%
5000
47.8%
6000
17.3%
6000
1.1%
7000
37.1%
> 90,000
8000
59.6%
9000
1.8%
10000
0.4%
< 30,000
Chemical
8.3%
HandySize
91.7%
Tankers
30,000 - 60,000
Chemical
0.2%
HandySize
99.8%
> 60,000
PanaMax
100.0%
The container hoteling times for Port of Seattle translated into the DWT range bins used in this analysis
are shown in Table A-39. The translated container and tanker ship hoteling times for Port of New
York/New Jersey are shown in Table A-40.
Table A-39. Translated Container Ship Hoteling Times for Port of Seattle
DWT Bin
Hoteling Time (hr)
< 25,000
24.2
25,000 - 35,000
28.1
35,000 - 45,000
30.4
45,000 - 90,000
30.3
> 90,000
34.2
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
46
-------
Appendix A: Baseline Emission Inventory Methodology
Table A-40. Translated Container and Tanker Ship Hoteling Times for Port of New York/New Jersey
Ship Type
DWT Bin
Hoteling Time (hr)
Container Ship
< 25,000
18.0
25,000 - 35,000
16.5
35,000 - 45,000
17.8
45,000 - 90,000
21.5
> 90,000
40.5
Tankers
< 30,000
4.2
30,000 - 60,000
2.1
> 60,000
5.0
The Charleston Inventory included hoteling time by ships and calls. This was mapped to the ships and
calls found in the Entrances and Clearances data for the Port of Charleston and averaged by ship
type/engine type/DWT bin. The results from this analysis are shown in Table A-41.
Table A-41. Calculated Hoteling Times for Port of Charleston
Ship Type
Main Engine
DWT Range
Hoteling Time (hr)
Auto Carrier
SSD
10,000 - 20,000
38.0
20,000 - 30,000
38.0
> 30,000
38.0
Bulk Carrier
MSD
< 25,000
20.3
SSD
< 25,000
20.3
25,000 - 35,000
20.3
35,000 - 45,000
13.0
45,000 - 90,000
28.2
Container Ship
MSD
< 25,000
11.1
SSD
< 25,000
13.8
25,000-35,000
14.1
35,000 - 45,000
14.9
45,000 - 90,000
14.7
> 90,000
29.9
General Cargo
MSD
< 25,000
26.8
SSD
< 25,000
36.3
25,000 - 35,000
30.3
35,000 - 45,000
22.8
45,000 - 90,000
21.5
Passenger
MSD
< 10,000
46.2
MSD-ED
< 10,000
46.2
10,000 - 20,000
46.2
SSD
< 10,000
46.2
Reefer
SSD
< 10,000
62.5
RORO
MSD
< 10,000
26.8
SSD
10,000 - 20,000
13.3
> 30,000
13.3
GT-ED
> 30,000
17.5
Tanker
MSD
< 30,000
63.7
SSD
< 30,000
54.4
30,000 - 60,000
66.2
60,000 - 90,000
73.5
90,000 -120,000
79.6
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
47
-------
Appendix B: Business as Usual Emission Inventory Methodology
Appendix B. Business as Usual Emission Inventory
Methodology
B.l. Introduction
This assessment included the development of representative, national scale inventories for the baseline
and Business as Usual (BAU) cases for different pollutants and years, followed by the analysis of various
strategies to reduce port-related mobile source emissions. This appendix details the methodology used
to develop the BAU emission inventories for the calendar years 2020, 2030, and 2050.
Separate inventories for various pollutants were developed for the drayage trucks, rail, cargo handling
equipment (CHE), harbor craft, and ocean going vessels (OGV) sectors. The following pollutants were
included in these inventories for 2020 and 2030: nitrogen oxides (NOx), fine particulate matter (PM25),
volatile organic compounds (VOCs), sulfur dioxide (S02), carbon dioxide (C02), black carbon (BC),
acetaldehyde, benzene, and formaldehyde. Note that the selected air toxics (acetaldehyde, benzene,
and formaldehyde) were only analyzed for the non-OGV sectors and S02 was only analyzed for the OGV
sector. Additionally, inventories were developed for 2050 for C02 only. In general, inventories were
developed for each port analyzed in this assessment using national scale methodology and data,
although port-specific data and adjustments were included where available and are noted where
appropriate in this appendix. This assessment does not provide specific projections for local decision-
making at individual ports or specific neighborhoods.
B.2. Projecting the Baseline Inventory
The baseline inventory was projected using port and sector specific growth factors. Then, adjustments
were made at some ports due to recent or planned changes that are expected to change future
emissions.
B.2.1. Growth Factors
For the OGV sector, growth was based on regional annual bunker fuel growth rates from 2002 to 2020 in
a 2008 study by Research Triangle Institute.73 Average annual growth factors by region that were
derived from that study are listed in the C3 RIA Table 3-69. A subset of those, listed in Table B-l, were
used for OGV sector growth for both 2020 and 2030.
Growth for other sectors was based on international trade growth factors from the same study.
Compound annual growth rates (CAGR) for years 2020 and 2030 were calculated from commodity
movements, which are imports plus exports. CAGR were determined relative to the 2011 baseline year
73 Research Triangle I nstitute, Global Trade and Fuel Assessment - Future Trends and Effects of Requiring Cleaner Fuels in the
Marine Sector, EPA Report EPA420-R-08-021, November 2008.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
48
-------
Appendix B: Business as Usual Emission Inventory Methodology
and then aggregated into the four conveyance methods shown in Table B-2. The 2011 US Army Corps of
Engineers (USACE) commodity throughput74 at each port of interest was used to weight the various
categories in Table B-2 to determine the 2020 and 2030 CAGR for each port. For example, if the 2011
commodity throughput at Port X on the Atlantic Coast was 50% containers, 35% bulk, and 15% liquid,
the 2020 CAGR for Port X would be calculated as:
^50% x 4% + 35% x 3.2% + 15% x 0.5%^
V 50% + 35% + 15% ) ~ 3'2%
Table B-l. Regional Bunker Fuel Use Growth Factors for 2020 and 2030
Region
Average Annual Growth Rate
East Coast
4.5%
Gulf Coast
2.9%
South Pacific
5.0%
North Pacific
3.3%
Table B-2. Compound Annual Growth Rates for 2020 and 2030 by Region and Commodity
Conveyance
Category
U.S. ATLANTIC-
Imports + Exports
U.S. PACIFIC NORTH-
Imports + Exports
U.S. PACIFIC SOUTH -
Imports + Exports
U.S.GULF COAST -
Imports + Exports
2020
2030
2020
2030
2020
2030
2020
2030
Bulk
3.2%
2.7%
4.0%
4.0%
3.9%
3.8%
3.3%
3.2%
Container
4.0%
4.4%
4.0%
4.5%
4.3%
4.9%
3.8%
4.1%
Liquid
0.5%
1.1%
1.5%
1.6%
1.1%
1.1%
1.4%
1.6%
Other
5.0%
4.9%
5.0%
4.8%
7.4%
7.2%
3.9%
4.2%
Total
2.7%
2.9%
3.8%
4.0%
3.5%
4.0%
2.2%
2.3%
B.2.2. Adjustments to the Projected BAU
In addition to baseline growth, the projected BAU inventory also considered ongoing or planned
changes in port operations that could substantially change emissions in future years. An example of such
a change could be plans for construction of on-dock rail that would change the mode split and shift
cargo from truck to rail. Where identified, options were assessed for quantifying these changes in the
BAU emission projections.
B.3. Drayage Trucks
This section describes the methodology used for projecting BAU emissions from drayage activity.
74 Available at: http://www.navieationdatacenter.us/db/wcsc/archive/xls/manll/.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
49
-------
Appendix B: Business as Usual Emission Inventory Methodology
B.3.1. Modeling Approach
Truck activity in 2020 and 2030 was estimated by applying the growth factors described in Table B-2 to
the 2011 baseline cargo volumes. The share of cargo throughput moved by truck came from version 3
of the Freight Analysis Framework (FAF).75 The FAF modal splits by commodity class were aggregated to
total tonnage share moved by truck. The same modal splits were used for 2011, 2020, and 2030.
DrayFLEET76 was run with the baseline 2011 cargo volumes and the national 2020 age distribution from
MOVES2010b.77 These intermediate emission inventories were then scaled by the ratio of projected
truck tonnage in 2020 to the baseline 2011 truck tonnage to calculate the 2020 drayage BAU inventory.
This was then repeated with the national 2030 age distribution from MOVES2010b and projected 2030
tonnage. However, since the version of DrayFLEET used in this analysis was not capable of running
calendar year 2030, the model was run as it was for 2020, but with a modified 2030 age distribution. To
get the model to accept the 2030 age distribution, all 2021-2030 model year vehicles were labeled as
2020. This was not expected to impact the results, as the heavy duty vehicle standards modeled here
are identical for model years 2020-2030.78 To calculate the 2050 C02 inventories, the 2030 inventories
were scaled using the anticipated growth in tonnage.
B.3.2. Additional Pollutants
The DrayFLEET Model estimates emissions for PM2.5, NOx, HC, CO, and C02. VOC emissions were
estimated as equal to the HC emissions. Emissions for the air toxics formaldehyde, benzene, and
acetaldehyde were estimated as a fraction of VOC emissions based on diesel speciation profiles
calculated from running MOVES2010b79, as shown in Table B-3. These fractions vary depending on
whether or not VOC is controlled (model year 2007 and later). Weighted speciation factors were
calculated for 2020 and 2030 based on the percent of 2007 model year and greater trucks in the fleet.
Approximately 50% of drayage trucks in 2020 and 11% in 2030 were model year 2006 or earlier.
Black carbon (BC) emissions were estimated as a 77% of PM2.5 emissions, consistent with EPA's Report
to Congress on Black Carbon.80
75 Available at: http://www.ops.fhwa.dot.gov/freight/freight analvsis/faf/.
76 U.S. Environmental Protection Agency, SmartWay DrayFLEET: Truck Drayage Environment and Energy Model, Version 2.0
User's Guide, EPA-420-B-12-065, June 2012.
77 U.S. Environmental Protection Agency, Motor Vehicle Emission Simulator (MOVES): User Guide for MOVES2010b, EPA-420-B-
12-001b, June 2012.
78 This modeling relies on a previous version of MOVES that does not include EPA's heavy-duty engine and vehicle GHG
regulations.
79 U.S. Environmental Protection Agency, MOVES2010b: Additional Toxics Added to MOVES. EPA-420-B-12-029a, May 2012, Sec
3.1.1. Available at: http://www.epa.gov/otaa/models/moves/documents/420bl2029a.pdf.
80 U.S. Environmental Protection Agency, Report to Congress on Black Carbon, EPA-450/R-12-001, March 2012, p. 87.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
50
-------
Appendix B: Business as Usual Emission Inventory Methodology
Table B-3. Diesel Truck Air Toxic Speciation Profiles Based on MOVES2010b
Pollutant
Toxic/VOC without Control
Toxic/VOC with Control
Acetaldehyde
0.035559
0.06934
Benzene
0.007835
0.01291
Formaldehyde
0.078225
0.21744
B.4. Rail
This section describes the methodology used for projecting BAU emissions from port-related rail activity.
B.4.1. Modeling Approach
The baseline 2011 rail sector activity was grown to 2020 and 2030 using the commodity growth rates
shown in Table B-2. Emission factors were calculated from the baseline inventories, adjusted for
expected changes in the future fleet, and then applied to the projected activity to determine the BAU
inventories.
Gross emission factors81 were calculated from the baseline 2011 rail inventory using the following
equation:
EF2011 = E2oii/(C2oh x S) Eq. B-l
Where:
EF = Emission factor for a specific pollutant, port, and locomotive type (g/ton),
E = Total annual emissions for a specific pollutant, port, and locomotive type (g),
C = Total cargo throughput for a specific port (tons), and
S = Share of cargo throughput moved by rail for a specific port (percent of total cargo tonnage).
To calculate the gross emission factors, the total annual emissions came from the baseline rail
inventories (see Appendix A). The total cargo throughput came from USACE's Waterborne Commerce
Statistics.82 The share of cargo throughput moved by rail came from version 3 of the Freight Analysis
Framework (FAF).83 The FAF modal splits by commodity class were aggregated to total tonnage share
moved by rail. The same modal splits were used for 2011, 2020, and 2030. Combining all of these results
in gross emission factors that are valid for the 2011 locomotive fleet at each port.
However, since fleets turn over to newer models in future years that meet stricter emission standards,
the gross emission factors were scaled for use in 2020 and 2030 based on projected future fleet
81 Used here, a "gross emission factor" is an estimate of emissions per unit goods moved. It is described as "gross" to
distinguish from more refined factors, such as engine-, equipment-, operations-, or technology-specific emission factors
determined from sources such as engine certification or emissions models.
82 Available at: http://www.navigationdatacenter.us/db/wcsc/archive/xls/manll/.
83 Available at: http://www.ops.fhwa.dot.eov/freieht/freieht analvsis/faf/.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
51
-------
Appendix B: Business as Usual Emission Inventory Methodology
emission factors listed in EPA's 2008 Locomotive and Marine Emission Standards Rulemaking.84 The
emission factors given in the rulemaking were not used directly because they are in terms of grams per
gallon, and the units required here were grams per ton of cargo moved. However, their values as used in
this scaling analysis are presented in Table B-4. Please note that only NOx, PMio, and HC were available
for this technique; the other pollutants were calculated separately. This scaling approach is consistent
with EPA's 2011 Air Quality Modeling Platform's Technical Support Document.85 Therefore, future year
gross emission factors were calculated using the following equations:
EFzozo = EF2M1 X Eq. b.2
EF2030 = EF2M1 X [IH§!=] Eq. B-3
Where:
EF = Emission factor for a specific pollutant, port, and locomotive type (g/ton) and
FuelEF = Fuel based emission factor by pollutant and locomotive type (Table B-4).
Table B-4. Projected Emission Factors (g/gal) from 2008 EPA Locomotive and
Marine Emission Standards Rulemaking
Calendar Year
Pollutant
Large Line-haul
Large Switch
2011
NOx
149
235
PMio
4.4
5.3
HC
7.7
14.0
2020
NOx
99
187
PMio
2.3
4.1
HC
3.6
10.5
2030
NOx
53
119
PMio
1.0
2.5
HC
1.9
6.2
Since the BAU inventories are grouped by rail lines and rail yards, it was assumed that baseline
emissions associated with rail yards were from switcher locomotives, while all emissions associated with
the rail line segments were from line-haul locomotives.
To calculate the BAU inventories, the future year emission factors were combined with grown rail
activity as shown in the following equations:
84 U.S. Environmental Protection Agency, Emission Factors for Locomotives, EPA-420-F-09-025, April 2009, Tables 5-7.
85 U.S. Environmental Protection Agency, Technical Support Document (TSD): Preparation of Emissions Inventories for the
Version 6.0, 2011 Emissions Modeling Platform, Section 4.4.1, February 26, 2014. Available at:
http://www.epa.eov/ttn/chief/emch/2011v6/outreach/2011v6 2018base EmisMod TSD 26feb2014.pdf.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
52
-------
Appendix B: Business as Usual Emission Inventory Methodology
E2020 —
EF2020 X C2011 X (1 + G)(2020-2011) X S
Eq. B-4
^2030 — EF2030 x ^2011 X (1 + G)^2030 2011) x S
Eq. B-5
Where:
E = Total annual emissions for a specific pollutant, port, and locomotive type (g),
EF = Emission factor for a specific pollutant, port, and locomotive type (g/ton),
C = Total cargo throughput for a specific port (tons),
G = Commodity-based CAGR for the region in which the port is located (see Table B-2), and
S = Share of cargo throughput moved by rail for a specific port (percent of total cargo tonnage).
Separate emissions were calculated using the specific emission factors for line-hauls and switchers, both
of which were associated with total cargo throughput handled by rail. Total rail emissions were then the
sum of projected switcher plus line-haul emissions. Note that emission factors and activity data were
used separately, rather than projection factors as used in the Mobile Source Air Toxics (MSAT) Rule.86
National projection factors (including both growth and control) were not appropriate at the port level
and would not accommodate the regional growth factors used here.
B.4.2. Additional Pollutants
The methodology described above provides inventories for NOx, PM10, and HC only; therefore, VOC,
PM2.5, C02, BC, benzene, acetaldehyde, and formaldehyde needed be calculated separately.
B.4.2.1. VOC
VOCs were estimated to be 1.053 times HC emissions, which is consistent with EPA's 2008 Locomotive
and Marine Emission Standards Rulemaking.
B.4.2.2. PM2.5
PM2.5 emissions were estimated to be 0.97 times the PM10 emissions, which is consistent with EPA's
2008 Locomotive and Marine Emission Standards Rulemaking.
B.4.2.3. BC
BC emissions were estimated to be 0.77 times the PM2.5 emissions, which is consistent with EPA's Report
to Congress on Black Carbon.87
86 U.S. Environmental Protection Agency, National Scale Modeling of Air Toxics for the Mobile Source Air Toxics Rule: Technical
Support Document, EPA 454/R-06-002, January 2006.
87 U.S. Environmental Protection Agency, Report to Congress on Black Carbon, EPA-450/R-12-001, March 2012, p. 87.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
53
-------
Appendix B: Business as Usual Emission Inventory Methodology
B.4.2.4. C02
Future year C02 emission factors were assumed to remain constant and equal to the baseline year. For
the 2050 C02 inventory, the 2030 CAGR were applied to grow the cargo throughput values to 2050
levels.
B.4.2.5. Air Toxics
Emissions for the air toxics formaldehyde, benzene, and acetaldehyde were estimated as a fraction of
VOC emissions based on diesel speciation profiles calculated from running MOVES2010b. MOVES was
used for the rail sector due to a lack of other sources more directly applicable to rail: NONROAD does
not include locomotives and SPECIATE and the NEI do not include projections. The MOVES speciation
profiles are shown in Table B-5. These fractions vary depending on whether or not VOC is controlled
(model year 2007 and later). For nonroad engines, this relates primarily to engines mandated to use Tier
4 emission controls. Weighted speciation factors were calculated for 2020 and 2030 based on Tier 4
versus pre-Tier 4 engine distributions from the 2008 Locomotive and Marine RIA.88 This is not consistent
with the MSAT Rule or the Emissions Modeling Platform, which use constant speciation factors for
future years.
Table B-5. Diesel Truck Air Toxic Speciation Profiles Based on MOVES2010b Applied to Rail89
Pollutant
Toxic/VOC without control
Toxic/VOC with Control
Acetaldehyde
0.035559
0.06934
Benzene
0.007835
0.01291
Formaldehyde
0.078225
0.21744
88 U.S. Environmental Protection Agency, Control of Emissions of Air Pollution from Locomotive Engines and Marine
Compression Ignition Engines Less than 30 Liters Per Cylinder, EPA420-R-08-001, March 2008.
89 U.S. Environmental Protection Agency, MOVES2010b: Additional Toxics Added to MOVES. EPA-420-B-12-029a, May 2012, Sec
3.1.1. Available at: http://www.epa.eov/otaq/models/moves/documents/420bl2029a.pdf.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
54
-------
Appendix B: Business as Usual Emission Inventory Methodology
B.5. Cargo Handling Equipment
This section describes the methodology used for determining BAU CHE emissions from port-related
activity.
B.5.1. Modeling Approach
The baseline 2011 CHE sector activity was grown to 2020 and 2030 using the commodity growth rates
shown in Table B-2. Emission factors were calculated from the baseline inventories, adjusted for
expected changes in the future fleet, and then applied to the projected activity to determine the BAU
inventories.
Gross emission factors90 were calculated from the baseline 2011 CHE inventory using the following
equation:
EF2011 = E2oii/(C2oh x S) Eq. B-6
Where:
EF = Emission factor for a specific pollutant and port (g/ton),
E = Total annual emissions for a specific pollutant and port (g), and
C = Total cargo throughput for a specific port (tons).
To calculate the gross emission factors, the total annual emissions came from the baseline CHE
inventories (see Appendix A). The total cargo throughput came from USACE's Waterborne Commerce
Statistics.91 However, since fleets turn over to newer models in future years that meet stricter emission
standards, the gross emission factors were scaled for use in 2020 and 2030 based on projected future
fleet emission factors obtained from national scale runs of EPA's NONROAD model.92 Emission factors
for all CHE were determined by dividing the sum of national emissions for port-related cargo handling
equipment (a list of applicable SCCs is given in Table B-6) by the sum of the national populations of the
same equipment. These emission factors calculated from the model runs were not used directly because
they are in terms of grams per vehicle per year, and the units required here were grams per ton of cargo
moved. However, their values as used in this scaling analysis are presented in Table B-7. Please note that
this technique was only used for NOx, PM2.5, and HC inventories; the other pollutants were calculated
separately.
90 Used here, a "gross emission factor" is an estimate of emissions per unit goods moved. It is described as "gross" to
distinguish from more refined factors, such as engine-, equipment-, operations-, or technology-specific emission factors
determined from sources such as engine certification or emissions models.
91 Available at: http://www.navigationdatacenter.us/db/wcsc/archive/xls/manll/.
92 Available at: https://www.epa.eov/otaq/nonrdmdl.htm.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
55
-------
Appendix B: Business as Usual Emission Inventory Methodology
Table B-6. CHE Types by SCC from NONROAD
SCC
Equipment Type
Equipment Category
2270002015
Rollers
Construction and Mining Equipment
2270002027
Signal Boards/Light Plants
Construction and Mining Equipment
2270002036
Excavators
Construction and Mining Equipment
2270002045
Cranes
Construction and Mining Equipment
2270002051
Off-highway Trucks
Construction and Mining Equipment
2270002060
Rubber Tire Loaders
Construction and Mining Equipment
2270002066
Tractors/Loaders/Backhoes
Construction and Mining Equipment
2270002069
Crawler Tractor/Dozers
Construction and Mining Equipment
2270002072
Skid Steer Loaders
Construction and Mining Equipment
2270002075
Off-highway Tractors
Construction and Mining Equipment
2270003010
Aerial Lifts
Industrial Equipment
2270003020
Forklifts
Industrial Equipment
2270003030
Sweepers/Scrubbers
Industrial Equipment
2270003050
Other Material Handling Equipment
Industrial Equipment
2270003070
Terminal Tractors
Industrial Equipment
2270006005
Generator Sets
Commercial Equipment
2270006010
Pumps
Commercial Equipment
2270006015
Air Compressors
Commercial Equipment
2270006025
Welders
Commercial Equipment
Table B-7. Projected Emission Factors (Annual Grams per Equipment) from NONROAD
Calendar Year
NOx
HC
PM
2011
57,909
14,505
4,660
2020
23,268
6,816
1,817
2030
14,667
5,692
830
The future year gross emission factors were calculated using the following equations:
EF202o = EF2011 X Eq. B-7
EF203o = EF2011 X Eq. B-8
Where:
EF = Emission factor for a specific pollutant and port (g/ton) and
PopEF = Population based emission factor for a specific pollutant (Table B-7).
To calculate the BAU inventories, the future year emission factors were combined with grown CHE
activity as shown in the following equations:
E2020 = EF2020 X C2011 X (1 + G)(2020-2011) Eq. B-9
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
56
-------
Appendix B: Business as Usual Emission Inventory Methodology
E2030 = EF203o X C2011 X (1 + G)(2030-2011) Eq. B-10
Where:
E = Total annual emissions for a specific pollutant and port (g),
EF = Emission factor for a specific pollutant and port (g/ton),
C = Total cargo throughput for a specific port (tons), and
G = Commodity-based CAGR for the region in which the port is located (see Table B-2).
B.5.2. Additional Pollutants
The methodology described above provides inventories for NOx, PM2.5, and HC only; therefore, VOC, BC, C02,
benzene, acetaldehyde, and formaldehyde needed be calculated separately.
B.5.2.1. VOC
The future year proportional changes in VOC emissions were assumed to be equal to the proportional changes in
HC emissions.
B.5.2.2. BC
BC emissions were estimated to be 0.77 times the PM2.5 emissions, which is consistent with EPA's Report to
Congress on Black Carbon.93
B.5.2.3. C02
Future year C02 emission factors were assumed to remain constant and equal to the baseline year. For the 2050
C02 inventory, the 2030 CAGR were applied to grow the cargo throughput values to 2050 levels.
B.5.2.4. Air Toxics
Emissions for the air toxics formaldehyde, benzene, and acetaldehyde were estimated as a fraction of VOC
emissions based on speciation profiles calculated from running EPA's NMIM.94 These speciation profiles are shown
in Table B-8.
Table B-8. CHE Air Toxic Speciation Profiles from VOC Based on NMIM
Pollutant
2020
2030
Acetaldehyde
0.0155
0.0126
Benzene
0.0277
0.0286
Formaldehyde
0.0333
0.0268
93 U.S. Environmental Protection Agency, Report to Congress on Black Carbon, EPA-450/R-12-001, March 2012, p. 87.
94 Available at: https://www.epa.eov/otaq/nmim.htm.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
57
-------
Appendix B: Business as Usual Emission Inventory Methodology
B.6. Harbor Craft
This section describes the methodology used for determining BAU harbor craft emissions.
B.6.1. Modeling Approach
The projected 2020 and 2030 BAU emission inventories were developed as the product of emission
factors and activity data. To facilitate this, the sector was split into two categories: goods-moving and
non-goods moving. For vessels directly tied to goods movement, such as the various categories of tug,
tow, and push, the activity growth was grown to 2020 and 2030 using the commodity growth rates
shown in Table B-2.
For goods-moving harbor craft, gross emission factors95 were calculated from the baseline 2011 harbor
craft inventory using the following equation:
Where:
EF = Goods-moving emission factor for a specific pollutant and port (g/ton),
E = Goods-moving annual emissions for a specific pollutant and port (g), and
C = Total cargo throughput for a specific port (tons).
To calculate the gross emission factors for goods-moving harbor craft, the goods-moving annual
emissions came from the baseline harbor craft inventories (see Appendix A). The total cargo throughput
came from USACE's Waterborne Commerce Statistics.96 Combining these results in gross emission
factors that are valid for the 2011 goods-moving harbor craft fleet at each port.
For non-goods moving harbor craft, gross fuel-based emission factors were calculated from the baseline
inventory using the following equation:
Where:
EF = Non-goods moving emission factor for a specific pollutant and port (g/gallon),
E = Non-goods moving annual emissions for a specific pollutant and port (g), and
FC = Non-goods moving annual fuel consumption (gallons).
95 Used here, a "gross emission factor" is an estimate of emissions per unit goods moved. It is described as "gross" to
distinguish from more refined factors, such as engine-, equipment-, operations-, or technology-specific emission factors
determined from sources such as engine certification or emissions models.
96 Available at: http://www.navieationdatacenter.us/db/wcsc/archive/xls/manll/.
EF2011 — E2011/C2011
Eq. B-ll
EF2011 — E2011/FC2011
Eq. B-12
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
58
-------
Appendix B: Business as Usual Emission Inventory Methodology
The non-goods moving portion of the total annual emissions came from the baseline inventory for vessel
types such as ferries, support, fishing, and government. The fuel consumption was estimated from the
non-goods moving baseline C02 inventories: ECo2 [g] / (26.34% [fuel carbon content] * 3207 [g/gal] *
3.664 [C02 to C ratio]). Combining these results in gross emission factors that are valid for the 2011 non-
goods moving harbor craft fleet at each port.
However, since fleets turn over to newer models in future years that meet stricter emission standards,
both sets of emission factors needed to be adjusted. Therefore, the gross emission factors for both
goods-moving and non-goods moving were scaled for use in 2020 and 2030 based on projected
emissions per vessel as calculated from EPA's 2008 Locomotive and Marine Emission Standards
Rulemaking.97 Total emissions for the control case for 2020 and 2030 were divided by the total number
of CI and C2 engines in those years to determine average emissions of each pollutant per marine
engine.
The emission factors calculated from the rulemaking were not used directly because they varied by age,
engine type (main or auxiliary), engine power, and engine displacement; this level of disaggregation was
impractical to repeat for this activity. As such, the calculated values as used in this scaling analysis are
presented in Table B-9. Please note that only NOx, PM2.5, and VOC were available for this technique; the
other pollutants were calculated separately.
Table B-9. Projected Emission Factors (Annual Grams per Marine Engine) from 2008 Locomotive and Marine RIA
Calendar Year
NOx
VOC
PM2.5
2011
5,654,548
123,910
198,748
2020
3,592,204
79,765
111,930
2030
2,022,078
43,223
64,705
Future year gross emission factors were calculated using the following equations:
EF2„2„ = EF2011 X [£2E|MS] Eq. B-13
EF2030 = EF2'I I I >< GIIeFj™] ^ B"14
Where:
EF = Emission factor for a specific pollutant and port (g/ton) and
PopEF = Population based emission factor for a specific pollutant (Table B-9).
97 U.S. Environmental Protection Agency, Control of Emissions of Air Pollution from Locomotive Engines and Marine
Compression Ignition Engines Less than 30 Liters per Cylinder, EPA420-R-08-001, March 2008.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
59
-------
Appendix B: Business as Usual Emission Inventory Methodology
To calculate the goods-moving BAU inventories, the future year emission factors were combined with
grown cargo tonnage moved at each port as shown in the following equations:
E2020 = EF2020 X C2011 X (1 + G)(2020-2011) Eq. B-15
E2030 = EF203o X C2011 X (1 + G)(2030-2011) Eq. B-16
Where:
E = Goods-moving emissions for a specific pollutant and port (g),
EF = Goods-moving emission factor for a specific pollutant and port (g/ton),
C = Total cargo throughput for a specific port (tons), and
G = Commodity-based CAGR for the region in which the port is located (see Table B-2).
For the non-goods moving BAU inventories, the activity was assumed to be inelastic to changes in cargo
movement and therefore assumed to have no growth. In other words, vessels such as ferries, support
(offshore & research), and fishing were assumed to operate at 2011 activity levels for all future years.
This assumption was used in lieu of better data as no nationwide projected values for harbor craft are
readily available. The 2008 study by the Research Triangle Institute, the basis of the growth rates shown
in Table B-2, discusses global historical and projected fuel consumption for non-goods moving vessels
but does not provide domestic values. The California Air Resources Board (ARB) OFFROAD model
predicts no change in the statewide fleet and fuel consumption for crew and supply vessels for all years
through 2025 and no change in the number of commercial fishing vessels from 2009 through 2030,
while showing a substantial increase in the number of "commercial" boats. However, it also predicts the
number of tugs to remain constant over this period. This approach is also similar to that used in the
Energy Information Administration's (ElA's) National Energy Modeling System, which determines fuel
demand for goods movement activities in the Freight Transport Module in two main categories: freight
and recreational. However, those categories are not good matches to those used here and NEI
projections are not likely to be consistent with the factors used here for vessels involved in goods
movements.
Therefore, to calculate the non-goods moving BAU inventories, the future year emission factors were
combined with the 2011 non-goods moving fuel consumption at each port as shown in the following
equations:
E2020 = EF202o x FC20ii Eq. B-17
E2030 = EF203o x FC20ii Eq. B-18
Where:
E = Goods-moving emissions for a specific pollutant and port (g),
EF = Goods-moving emission factor for a specific pollutant and port (g/gallon), and
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
60
-------
Appendix B: Business as Usual Emission Inventory Methodology
FC = Non-goods moving 2011 fuel consumption (gallons).
B.6.2. Additional Pollutants
The methodology described above provides inventories for NOx, PM2.5, and VOC only; therefore, BC,
C02, benzene, acetaldehyde, and formaldehyde needed be calculated separately.
B.6.2.1. BC
BC emissions were estimated to be 0.77 times the PM2.5 emissions, which is consistent with EPA's Report
to Congress on Black Carbon.98
B.6.2.2. C02
Future year C02 emission factors were assumed to remain constant and equal to the baseline year. For
the goods-moving 2050 C02 inventory, the 2030 CAGR were applied to grow the cargo throughput
values to 2050 levels. For the non-goods moving 2050 inventory, the 2011 non-goods moving activity
values were used.
B.6.2.3. Air Toxics
Emissions for the air toxics formaldehyde, benzene, and acetaldehyde were estimated as a fraction of
VOC emissions based on diesel speciation profiles calculated from running MOVES2010b. MOVES was
used for the harbor craft sector due to a lack of other sources more directly applicable to harbor craft:
NONROAD does not include commercial marine and SPECIATE and the NEI do not include projections.
The MOVES speciation profiles are shown in Table B-10. These fractions vary depending on whether or
not VOC is controlled (model year 2007 and later). For marine engines, this relates primarily to engines
mandated to use Tier 4 emission controls. Weighted speciation factors were calculated for 2020 and
2030 based on Tier 4 versus pre-Tier 4 engine distributions from the 2008 Locomotive and Marine RIA."
This is not consistent with the MSAT Rule or the Emissions Modeling Platform, which use constant
speciation factors for future years.
Table B-10. Diesel Truck Air Toxic Speciation Profiles Based on MOVES2010b Applied to Harbor Craft
Pollutant
Toxic/VOC without control
Toxic/VOC with Control
Acetaldehyde
0.035559
0.06934
Benzene
0.007835
0.01291
Formaldehyde
0.078225
0.21744
98 U.S. Environmental Protection Agency, Report to Congress on Black Carbon, EPA-450/R-12-001, March 2012, p. 87.
99 U.S. Environmental Protection Agency, Control of Emissions of Air Pollution from Locomotive Engines and Marine
Compression Ignition Engines Less than 30 Liters Per Cylinder, EPA420-R-08-001, March 2008.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
61
-------
Appendix B: Business as Usual Emission Inventory Methodology
B.7. Ocean Going Vessels
This section describes the methodology used for determining BAU OGV emissions.
B.7.1. Modeling Approach
The business as usual inventories for OGV are based primarily upon the methodology used for the
Category 3 Marine Engine Rulemaking100 (C3 RIA). Using the C3 RIA modeling approach, the OGV
emission inventories were calculated using energy-based emission factors combined with activity
profiles for vessels calling at each port.
Essentially, the baseline 2011 number of calls at each port included in this assessment was grown to
2020 and 2030 using the annual average growth rate as listed in Table B-l for the appropriate region.
This action alone would proportionally increase each of the inventories. However, adjustment factors
were also applied to the calculations for NOx, PM, BC, S02 and C02 to account for fleet turnover and
lower sulfur fuels.
The IMO adopted NOx limits in Annex VI to the International Convention for Prevention of Pollution
from Ships in 1997. These NOx limits apply for all marine engines over 130 kW for engines built on or
after January 1, 2000, including those that underwent a major rebuild after January 1, 2000. For the C3
RIA, EPA determined the effect of the IMO standard to be a reduction in NOx emission rate of 11%
below that the standard for engines built before 2000. Therefore, for engines built between 2000 and
2010 (Tier I), a NOx emission adjustment of 0.89 was applied to the calculation of NOx emissions for
both propulsion and auxiliary engines. IMO Tier II NOx emission standards came into effect in 2011 and
represent a 2.5 g/kWh reduction over Tier I engines. Tier III came into effect for engines built in 2016,
which represents an 80% reduction from Tier I. Thus Tier III emission factors are 20% of Tier I emission
factors. All emission factors are consistent with the C3 RIA.
In addition to the MARPOL Annex VI emission limits that apply to all ships engaged in international
transportation, US vessels must also comply with EPA's Clean Air Act requirements for engines and fuels.
The NOx emission limits for Category 3 engines are equivalent to the MARPOL Annex VI NOx limits.
EPA's sulfur limit for distillate locomotive or marine (LM) diesel fuel sold in the United States is more
stringent (15 ppm sulfur) than the ECA fuel sulfur limit (1000 ppm sulfur starting 2015); the sulfur limit
for ECA fuel for use on Category 3 marine vessels is equivalent to the MARPOL Annex VI SOx limits. EPA
also has emission standards for C3 engines,101 which are generally the same or more stringent but
almost all C3 engines used in international shipping fall under IMO regulations.
100 U.S. Environmental Protection Agency, Regulatory Impact Analysis: Control of Emissions of Air Pollution from Category 3
Marine Diesel Engines, EPA Report EPA-420-R-09-019, December 2009. Available at:
http://www.epa.gov/otaq/regs/nonroad/marine/ci/420r09019.pdf.
101 U.S. Environmental Protection Agency, Control of Emissions from New Marine Compression-Ignition Engines at or Above 30
Liters per Cylinder, Federal Register, Vol 75, No 83, April 30, 2010.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
62
-------
Appendix B: Business as Usual Emission Inventory Methodology
In addition, as part of the new IMO standards, marine diesel engines built between 1990 and 1999 that
are 90 liters per cylinder or more need to be retrofitted by 2020 to meet Tier 1 emission standards upon
engine rebuild if a retrofit kit is available to the ships. Consistent with the C3 RIA, it was assumed that
80% of all ships > 90L / cylinder will have retrofit kits available.
To calculate NOx reductions due to fleet turnover, NOx adjustment factors were calculated for 2020 and
2030 based upon all ports examined in this analysis combined. This simplification allowed the same
factors to be applied to all 19 ports. To accomplish this, installed power age profiles by engine type for
propulsion engines and by vessel type for auxiliary engines were developed using the 2011 Entrances
and Clearances data and Lloyd's vessel characterization data. For propulsion engines, installed power by
engine type was calculated for each model year based upon the sum of the total propulsion power of
Category 3 vessels over the Entrances and Clearances data. In addition, to calculate the effect of
retrofitting Tier 0 engines of more than 90 liters per cylinder, installed power was also calculated for
slow speed diesel (SSD) and medium speed diesel (MSD) engines that were over 90 liters per cylinder.
Ages were determined by subtracting the build year from 2011. This 2011 age profile was then used in
both 2020 and 2030 adjusting model years to fit the age profile. For example, a five-year-old engine in
2011 is a 2006 model year, but in 2020 is a 2015 model year and in 2030 a 2025 model year. This same
methodology was used in the C3 RIA.
For auxiliary engines, auxiliary power was calculated from the propulsion power using the auxiliary
power to propulsion power ratios by ship type found in Table A-20. This is a slight variation from the C3
RIA, which instead used the propulsion installed power to calculate auxiliary engine NOx adjustment
factors. Auxiliary engines were only segregated into passenger ships and other ships because in 2011
different residual oil (RO) to marine gas oil (MGO) ratios were used. The installed power by age and
engine type (or vessel type for auxiliary engines) is shown in Table B-ll.
Table B-ll. 2011 Installed Power by Engine Type (kW)
Age (yrs)
Propulsion Engines
Auxiliary Engines
MSD
SSD
GT
ST
Passenger
Other
58
-
883
-
-
-
238
56
4,415
155,328
-
-
43,181
1,188
54
60,044
-
-
-
-
16,152
51
5,880
-
-
-
-
1,123
47
-
-
-
28,318
-
6,287
46
1,104
-
-
-
-
211
45
5,663
-
-
-
-
1,082
44
23,055
-
-
11,400
-
7,562
43
23,536
-
-
-
6,543
-
41
3,884
-
-
-
-
742
40
142,385
156,512
-
235,360
-
118,761
39
683,299
-
-
-
-
171,013
38
40,578
52,950
-
94,144
9,812
33,484
37
13,245
96,350
-
23,536
-
26,111
36
113,264
237,351
-
-
-
69,933
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
63
-------
Appendix B: Business as Usual Emission Inventory Methodology
Age (yrs)
Propulsion Engines
Auxiliary Engines
MSD
SSD
GT
ST
Passenger
Other
35
61,451
41,192
-
-
-
25,096
34
683,005
818,491
-
-
-
330,714
33
419,242
962,313
-
63,254
-
326,991
32
1,416,412
1,307,594
-
-
-
691,957
31
1,289,949
1,589,319
-
-
-
674,436
30
168,980
3,951,662
-
-
9,681
895,328
29
1,583,342
2,503,409
-
-
-
918,034
28
102,456
6,291,171
-
-
-
1,571,537
27
36,852
17,197,530
-
-
-
3,927,271
26
235,329
7,271,460
-
-
-
1,694,493
25
53,936
8,010,446
-
-
-
1,806,526
24
134,791
6,790,457
-
-
-
1,590,847
23
170,487
11,922,756
-
-
11,743
2,699,496
22
2,316,184
14,011,939
-
-
636,028
3,150,791
21
6,409,732
4,072,808
-
-
1,616,225
1,081,023
20
1,341,768
14,379,033
-
-
299,533
3,252,828
19
4,850,116
12,301,414
-
-
1,262,876
2,858,540
18
537,682
22,408,866
-
-
76,861
5,494,131
17
1,824,020
28,760,465
-
-
413,430
6,973,053
16
12,471,612
44,043,180
-
-
3,402,311
9,750,108
15
18,714,120
50,532,802
-
-
5,067,754
11,270,238
14
10,139,100
55,342,547
-
-
2,411,583
12,629,186
13
25,079,400
49,681,652
-
-
6,361,730
11,478,444
12
23,817,894
32,483,814
-
-
5,965,327
7,798,999
11
23,603,136
60,132,802
4,701,972
-
6,975,766
13,978,348
10
27,380,891
65,091,176
13,058,032
-
10,594,636
14,892,250
9
13,195,720
87,355,544
-
105,920
3,083,740
19,632,342
8
16,670,650
75,944,446
10,315,048
211,824
6,921,840
17,047,496
7
31,990,234
85,901,288
1,265,000
116,208
8,168,129
19,686,618
6
22,460,220
97,966,162
-
-
5,752,821
21,910,074
5
14,017,483
111,010,740
-
116,208
2,196,800
25,779,985
4
18,001,132
129,071,986
-
107,104
3,482,793
29,685,530
3
18,759,733
103,070,751
-
57,330
3,682,955
24,017,466
2
3,278,815
86,978,426
-
55,200
349,279
19,553,144
1
13,620,786
93,117,757
-
52,992
3,199,891
21,069,329
0
2,370,454
24,647,140
-
-
557,446
5,617,455
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports 64
-------
Appendix B: Business as Usual Emission Inventory Methodology
Emission factors for propulsion engines by engine type, fuel type, and emission tier are shown in Table
B-12. This table combines the base emission factors as discussed in section A.6.6 and these adjustments
discussed above.
Table B-12. NOx Emission Factors by Engine Type
Engine Type
Tier
Engine Model Years
Emission Factor (g/kWh)
RO
MDO/MGO
0
Pre-2000
14.0
13.2
MSD
1
2000 -2010
12.5
11.7
2
2011-2015
10.0
9.2
3
2016+
-
2.3
0
Pre-2000
18.1
17.0
SSD
1
2000 -2010
16.1
15.1
2
2011-2015
13.6
12.6
3
2016+
-
3.0
GT
0
All
6.1
5.7
ST
0
All
2.1
2.0
0
Pre-2000
14.7
13.9
Auxiliary
1
2000 -2010
13.1
12.4
2
2011-2015
10.6
9.9
3
2016+
-
2.5
An ARB survey published in 2005102 found that almost all ships used RO in their main propulsion engines, and
that only 29% of all ships (except passenger ships) used distillate (MGO/MDO) in their auxiliary engines, with
the remaining 71% using RO. Only 8% of passenger ships used distillate in their auxiliary engines, while the
other 92% used RO. Even though these two fuels are not blended in any given vessel, the emission factor
used in the analysis represents an average of the two fuels, weighted by the relative market share of each.
For all other ships, 29% used distillate and 71% used RO. Table B-13 shows the NOx emission factors by Tier
for the two ship types.
Table B-13. Auxiliary Engine NOx Emission Factors by Ship Type
Calendar Year
Tier
Engine MY
Emission Factor (g/kWh)
Passenger
Other
0
Pre-2000
14.6
14.5
2011
1
2000 -2010
13.0
12.9
2
2011
10.5
10.4
0
Pre-2000
13.9
13.9
2020 and later
1
2000 -2010
12.4
12.4
2
2011-2015
9.9
9.9
3
2016+
2.5
2.5
102 California Air Resources Board, 2005 Oceangoing Ship Survey, Summary of Results, September 2005
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
65
-------
Appendix B: Business as Usual Emission Inventory Methodology
Using the above information, average NOx emission factors were calculated for the years 2020 and 2030.
Engine model years were assigned to the age profile based upon calendar year. For 2020, age 0 was model
year 2020 and for 2030, age 0 was model year 2030. Next, emission tiers and NOx emission factors from
Table B-12 were assigned to the various model years. Then, the effect of the retrofit requirement for engines
built between 1990 and 1999 were taken into account. Average NOx emission factors were determined for
2011, 2020, and 2030 by taking the sum of the installed power times the emission factor for each model year
and dividing the sum-product by the sum of the installed power and the results are shown in Table B-14. A
similar process was used for auxiliary engines, and the resulting average NOx emission factors are given in
Table B-15. This is different from the C3 RIA, where age distributions were broken out by Great Lakes/Deep
Sea Ports instead of ship type. NOx adjustment factors were then calculated by dividing the 2020 or 2030
average NOx emission factor by the 2011 average NOx emission factor for a given engine or vessel type.
These are given in Table B-16. These adjustment factors were then applied to the calculation of NOx
emissions by engine type and port.
Table B-14. Average Propulsion Engine NOx Emission Factor (g/kWh) by Engine Type
Year
MSD
SSD
GT
ST
2011
13.0
16.6
6.1
2.1
2020
9.4
10.6
5.7
2.0
2030
3.7
5.0
5.7
2.0
Table B-15. Average Auxilliary Engine NOx Emission Factor (g/kWh) by Ship Type
Year
Passenger
Other
2011
13.6
13.3
2020
10.3
8.6
2030
3.7
4.1
Table B-16. NOx Adjustment Factors
Year
Propulsion Engines
Auxiliary Engines
MSD
SSD
GT
ST
Passenger
Other
2020
0.7233
0.6389
0.9344
0.9524
0.7582
0.6505
2030
0.2856
0.2988
0.9344
0.9524
0.2736
0.3064
B.7.2. Additional Pollutants
Fuel changes from 2.7% sulfur RO to 0.1% sulfur MDO starting in 2015 affects PMio, PM2.5, BC, S02 and
C02 emissions. Emission factors for the two fuels used in propulsion engines are shown in Table B-17
and come from Entec.103 Please note that in the C3 RIA, different RO sulfur levels were used for West
103 Entec UK Limited, Quantification of Emissions from Ships Associated with Ship Movements between Ports in the European
Community, prepared for the European Commission, July 2002.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
66
-------
Appendix B: Business as Usual Emission Inventory Methodology
Coast ports. Since California ports were not included in this analysis and the 2011 Starcrest Inventory for
Port of Seattle used 2.7% sulfur for RO, we have used 2.7% sulfur RO for all ports.
Table B-17. Average Propulsion Engine Emission Factors (g/kWh) by Engine Type
Engine Type
Fuel Type
Sulfur
Emission Factor (g/kWh)
PMio
PM2.5
BC
SO2
CO2
BSFC
SSD
RO
2.70%
1.42
1.31
0.039
10.29
621
195
MDO
0.10%
0.19
0.17
0.010
0.36
589
185
MSD
RO
2.70%
1.43
1.32
0.040
11.24
678
213
MDO
0.10%
0.19
0.17
0.010
0.40
646
203
GT
RO
2.70%
1.47
1.35
0.040
16.10
971
305
MDO
0.10%
0.17
0.15
0.009
0.57
923
290
ST
RO
2.70%
1.47
1.35
0.040
16.10
971
305
MDO
0.10%
0.17
0.15
0.009
0.57
923
290
The fuel changes also affect the same emissions for auxiliary engines. Emission factors for the two fuels
used in auxiliary engines are shown in Table B-18 (also from Entec). BSFC from Entec was used based
upon the ratio of RO versus MDO listed in the table.
Table B-18. Average Auxilliary Engine Emission Factors (g/kWh) by Ship Type
Ship Type
Fuel Type
Sulfur
Emission Factor (g/kWh)
PM10
PM2.5
BC
SO2
CO2
BSFC
Passenger
92% RO/8% Distillate
2.56%
1.36
1.25
0.038
11.36
718
226
100% MGO
0.10%
0.18
0.17
0.010
0.42
691
217
Other
71% RO/29% Distillate
2.21%
1.16
1.07
0.032
9.74
707
224
100% MGO
0.10%
0.18
0.17
0.010
0.42
691
217
Adjustment factors taking into account the change in emission factors for both propulsion and auxiliary
engines due to lower sulfur fuel are shown in Table B-19 and were applied to emissions calculations for both
2020 and 2030. It was assumed that by 2020 and 2030, all vessels will be using distillate fuel at these ports.
Table B-19. Fuel Adjustment Factors for 2020 and 2030
Engine
Type
PM10
PM2.5
BC
SO2
CO2
Propulsion
MSD
0.1329
0.1329
0.2659
0.0351
0.9487
SSD
0.1295
0.1295
0.2591
0.0353
0.9531
GT
0.1134
0.1134
0.2268
0.0352
0.9508
ST
0.1134
0.1134
0.2268
0.0352
0.9508
Auxiliary
Passeneer
0.1340
0.1340
0.2680
0.0373
0.9617
Other
0.1569
0.1569
0.3138
0.0436
0.9772
Finally, it was assumed that there would be an increase in the use of shore power at a limited
number of ports in this assessment. It was assumed that those ships that use shore power only emit
auxiliary engine emissions during the time the shore power cables are being connected to and
disconnected from the ship. This was estimated to take two hours per call. Assuming an average of
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
67
-------
Appendix B: Business as Usual Emission Inventory Methodology
10 hours hoteling per call for passenger ships results in an 80% reduction in all emissions except C02
that use shore power. C02 emissions from the power plant that generates electricity must also be
considered. Based upon current projections of U.S. average generation mix104 using the Argonne
National Laboratories GREET2014 model,105 electricity generation produces 517 g/kWh of C02 at the
plug compared for 2020, 477 g/kWh in 2030 and 461 g/kWh in 2050106 to the 691 g/kWh generated
by the auxiliary engines. Taking into account the time the cables are connected, this results in a 20,
25 and 27% reduction in C02 emissions for 2020, 2030 and 2050, respectively, for ships that use
shore power.107
B.8. BAU Summary Results
The following figures show the 2011 baseline inventories combined with the 2020 and 2030
business as usual projections for each pollutant, aggregated by sector. As noted earlier, S02 was
only calculated for OGV and acetaldehyde, benzene, and formaldehyde were only calculated for the
non-OGV sectors.
Figure B-l. Total NOx Emissions Aggregated by Sector, Tons/Year
50,000
40,000
-
= 20,000
I-
10,000
0
12011
12020
12030
OGV
Harbor Craft
Rail
Mode
CHE
Drayage
104 The C02 emission rates calculated using GREET2014 assume an average U.S. generation mix. At some ports, the generation
mix is significantly different and would thus have a different emission factor. For example, in the Northwest, much of the
electricity comes from hydropower therefore utilities emit less C02 overall.
105 Argonne National Laboratories, GREET Model 2014. Available at: https://ereet.es.anl.eov/.
106 GREET2014 only extrapolates to 2040, so the 2040 C02 emission rate for power plants was also used for 2050.
107 Note that cargo loading and unloading occur while the connection is being made and removed, so the total hoteling time
estimate is expected to be unchanged by shore power, although on a first call at a new terminal commissioning is required,
which takes much longer. This calculation assumes 2 hours per call for connection and disconnection, but does not include
any commissioning time.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
68
-------
Appendix B: Business as Usual Emission Inventory Methodology
Figure B-2. Total PM2.5 Emissions Aggregated by Sector, Tons/Year
3,000
2,500
£ 2,000
a>
> 1,500
C
° 1,000
500
0
OGV Harbor Craft Rail CHE Drayage
Mode
12011
12020
12030
3,000
2,500
£ 2,000
a>
> 1,500
C
° 1,000
500
0
Figure B-3. Total VOC Emissions Aggregated by Sector, Tons/Year
OGV Harbor Craft
Rail
Mode
CHE
Drayage
12011
12020
12030
Figure B-4. Total BC Emissions Aggregated by Sector, Tons/Year
1200
1000
800
600
400
200
0
OGV Harbor Craft
Rail
Mode
CHE
L
Drayage
12011
12020
12030
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
69
-------
Appendix B: Business as Usual Emission Inventory Methodology
Figure B-5. Total C02 Emissions Aggregated by Sector, Tons/Year
8,000,000
7,000,000
6,000,000
> 15,000
C
° 10,000
5,000
0
OGV
Mode
12011
12020
12030
60
50
40
30
20
10
0
Figure B-7. Total Acetaldehyde Emissions Aggregated by Sector, Tons/Year
12011
12020
12030
Harbor Craft
Rail
CHE
Drayage
Mode
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
70
-------
Appendix B: Business as Usual Emission Inventory Methodology
16
14
12
S 10
£ 8
O 6
I-
4
2
0
Figure B-8. Total Benzene Emissions Aggregated by Sector, Tons/Year
12011
12020
12030
Harbor Craft
Rail
CHE
Drayage
Mode
Figure B-9. Total Formaldehyde Emissions Aggregated by Sector, Tons/Year
120
100
80
60
40
20
0
12011
12020
12030
Harbor Craft Rail CHE Drayage
Mode
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
71
-------
Appendix C: Analysis of Emission Reduction Scenarios
Appendix C. Analysis of Emission Reduction Scenarios
C.l. Overview
This appendix describes the methodology and other assumptions that support the scenario analysis
results presented in Section 6 of the final report.
C.l.l. Intent of Scenarios
The strategy scenarios were developed based on the screening-level assessment described in Section 5
where a range of potential technological and operational strategies were evaluated, in addition to EPA's
existing expertise on port-related strategies and consultation with port stakeholder experts. Strategy
scenarios were developed for each mobile source sector for the years 2020 and 2030 for all pollutants
and for only C02 in 2050. Although the specific strategies differ between sectors, the purpose of all
strategy scenarios are as follows:
¦ Scenario A was intended to reflect an increase in the introduction of newer technologies in port
vehicles and equipment beyond what would occur through normal fleet turnover. Operational
strategies in Scenario A reflect a reasonable increase in expected efficiency improvements for
drayage truck, rail, and OGV sectors. For the OGV sector, moderate levels of fuel switching and
other emission control strategies are also analyzed. All of the strategies included in Scenario A may
be supported by a moderate increase in public and private funding.
¦ Scenario B reflected a more aggressive suite of strategies as compared to Scenario A. Scenario B
would necessitate a major public and private investment to accelerate introduction of zero
emissions vehicles, equipment, and vessels, in addition to different fuels and other technologies.
Operational strategies in Scenario B assume further operational efficiency improvements beyond
Scenario A.
In selecting strategies, EPA qualitatively considered several factors, such as strategy costs,
implementation barriers, feasibility, and potential for market penetration by analysis year. However, an
in-depth cost-benefit analysis was not conducted.
The remainder of this appendix provides more details regarding the methodology and assumptions used
to estimate the strategy scenario reductions in Section 6.
C.2. Drayage Trucks
C.2.1. Scenarios
As discussed in Section 5, the scenarios were based on the future year distributions of drayage trucks
consistent with the national default fleet turnover rates in EPA's MOVES2010b model, which is
presented in Table C-l.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
72
-------
Appendix C: Analysis of Emission Reduction Scenarios
Table C-l. BAU Distribution of Trucks by Model Year
Model Year
2011
2020
2030
2050
pre-1991
20%
5%
0%
0%
1991-93
9%
6%
0%
0%
1994-97
21%
13%
0%
0%
1998-2003
24%
16%
7%
0%
2004-06
12%
9%
5%
0%
2007-09
10%
8%
5%
0%
2010+
4%
44%
84%
100%
Total
100%
100%
100%
100%
Table C-2 shows the strategy scenarios that were analyzed in this assessment. Note that model year
references in this table indicate the emission standards that began with that model year (e.g., a 2007
truck means a truck that meets the EPA standards effective for model years 2007-2009).
Table C-2. Drayage Truck Strategy Scenarios
Strategy
2020/A
2020/B
2030/A
2030/B
2050/A
2050/B
Replace all
Replace all
Replace
Replace 100%
Replace 25%
Replace 50%
pre-1994
pre-1998
100% of pre-
of pre-2007
of post-
of post-2010
trucks with
trucks with
2004 trucks
trucks with
2010 trucks
trucks with
50% post-
50% 2007,
with 2010
50% 2010 and
with PHEV
PHEV
Technological
1998, 30%
40% 2010,
trucks.
50% PHEV.
2007, 20%
and 10%
Replace 20%
Replace 10%
2010 or
PHEV
of 2004-09
of post-2010
newer trucks
trucks with
with PHEV
PHEV
Reduce gate
Reduce gate
Reduce gate
Reduce gate
Reduce gate
Reduce gate
Operational
queues by
queues by
queues by
queues by
queues by
queues by
25%
50%
25%
50%
25%
50%
Note that for the Operational Strategies, the percent of truck operating time spent in gate queues did
not vary for the BAU inventories in 2020, 2030, and 2050. However, the number of drayage trucks
increased with each future year, so the BAU emissions associated with gate queues also increased with
each future year.
C.2.2. Technological Strategy Scenarios: Methodology and Assumptions
C.2.2.1. Relative Reduction Factors
The DrayFLEET model was used to determine a fleet-average relative reduction factor (RRF) for each
scenario strategy. This is different from how the BAU inventories for drayage trucks were developed,
where a default truck fleet age distribution from MOVES2010b was used in the DrayFLEET model. For
the strategy scenario analysis, the BAU age distribution was replaced with alternative distributions
consistent with each scenario. Since plug-in hybrid electric vehicles are not included in MOVES, these
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
73
-------
Appendix C: Analysis of Emission Reduction Scenarios
trucks were accounted for outside the model. For each scenario, a fleet-average RRF was calculated as
the scenario emissions divided by the BAU emissions. To simplify calculations, a generic, national
average fleet emissions RRF representing an average port (a "typical port") was used, rather than
generating a separate model for each port.
Emission factors for conventional diesel drayage trucks for most pollutants were drawn from the
emission standards reported in EPA's Heavy-Duty Highway Compression-Ignition Engines and Urban
Buses Exhaust Emission Standards.108 VOC emission rates were determined from the hydrocarbon (THC)
standards, using EPA THC to VOC conversion factors.109 Emission rates for the select air toxics
acetaldehyde, benzene, and formaldehyde were calculated by applying speciation factors for diesel
engines to VOC emissions. The analysis relied on EPA speciation profiles for on-road engines from the
MOVES model.110 Well-to-wheel C02 emission factors came from GREET 2015111 using the 2015 HDV
emission factor for CIDI Combination Short-Haul Conventional Diesel. For these calculations, no change
in fuel economy or C02 emission rates were assumed between model year standards.112 For PHEVs,
emission factors were developed for NOx, PM2.5, and C02 based on percent reduction values from
SCAG's Comprehensive Regional Goods Movement Plan.113 RRFs for VOCs were based on emission
benefit percent reductions from the Gateway Cities Air Quality Action Plan.114 RRFs for black carbon (BC)
are assumed equal to that for PM2.5 consistent with the BAU emission inventory methodology.
Because the DrayFLEET model cannot readily model each of these scenarios, the fleet-average emission
factor was calculated from the emission standards described above weighted by truck population
distributions specific to each scenario. To develop the truck population distributions for each scenario,
the default truck fleet age distribution from the BAU methodology was used (originally taken from
MOVES2010b). No PHEV trucks were assumed in the BAU fleet mixes. The population distributions were
adjusted based on the scenarios. Table C-3 shows the model distributions that resulted from applying
these scenarios to a hypothetical population of 1,000 trucks.
108 U.S. Environmental Protection Agency, Emission Standards Reference Guide. Available at:
http://www.epa.gov/otaq/standards/heavv-dutv/hdci-exhaust.htm.
109 U.S. Environmental Protection Agency, Conversion Factors for Hydrocarbon Emission Components (July 2010), Report EPA-
420-R-10-015 NR-002d.
110 U.S. Environmental Protection Agency, MOVES2010b: Additional Toxics Added to MOVES. EPA-420-B-12-029a, May 2012,
Sec 3.1.1. Available at: http://www.epa.gov/otaq/models/moves/documents/420bl2029a.pdf.
111 GREET2015 was released October 2, 2015. All calculations were updated to GREET2015 results. For information on the
model, see https://greet.es.anl.gov/.
112 Note that BAU inventories are based on DrayFLEET results. That model relies on a previous version of MOVES, which does
not include EPA's heavy-duty engine and vehicle GHG regulations.
113 ICF International, Comprehensive Regional Goods Movement Plan, Task 10.2: Evaluation of Environmental Mitigation
Strategies, prepared for the Southern California Association of Governments, 2012.
114 ICF International, Gateway Cities Air Quality Action Plan, Task 7: New Measures Analysis, prepare for the Los Angeles County
Metropolitan Transportation Authority and the Gateway Cities Council of Governments, 2013.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
74
-------
Appendix C: Analysis of Emission Reduction Scenarios
Table C-3. Example of Drayage Truck Model Year Distribution by Scenario
Model Year
2020 BAU
2020/A
2020/B
2030 BAU
2030/A
2030/B
2050 BAU
2050/A
2050/B
pre-1991
50
0
0
0
0
0
0
0
0
1991-93
50
0
0
0
0
0
0
0
0
1994-97
130
130
0
0
0
0
0
0
0
1998-2003
160
185
160
60
0
0
0
0
0
2004-06
90
115
90
50
40
0
0
0
0
2007-09
80
110
195
50
40
50
0
0
0
2010+
440
460
532
840
900
811
1,000
750
500
PHEV
0
0
23
0
20
139
0
250
500
Total
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
An average emission factor for the fleet described by each scenario and for the BAU vehicle fleet was
determined for each pollutant with the emissions factors (EFs) described above. The scenario RRF was
then calculated as:
RRF = 1-Scenario EF/BAUEF Eq. C-l
Note that use of emission factors to determine RRFs in this manner implies that other technical and
operational parameters, such as engine load and power, are unchanged between the BAU and scenario
analysis. Table C-4 shows the resulting Technological RRFs, applicable to a typical port.115
Table C-4. Emission Relative Reduction Factors for Drayage Technological Scenarios
Scenario
Overall Emission Reductions (%)
NOx
PM2.5
voc
BC
CO2
Acetaldehyde
Benzene
Formaldehyde
2020/A
19%
43%
14%
43%
0%
10%
11%
7%
2020/B
48%
62%
35%
62%
1%
21%
25%
12%
2030/A
48%
34%
33%
34%
0%
20%
23%
14%
2030/B
60%
52%
39%
52%
4%
22%
26%
15%
2050/A
-
-
-
-
6%
-
-
-
2050/B
-
-
-
-
13%
-
-
-
C.2.2.2. Application of Emission Relative Reduction Factors
The typical port RRFs were applied to each port's BAU drayage truck emission inventory, determined
under the BAU methodology. To approximate changes at the few ports where adjustments were made
to the BAU drayage age distribution to account for local programs, the revised age distribution
described by the scenario was compared to the local distribution. These ports have local programs in
effect that would exceed the scenarios considered here. For those cases, no additional emission
reductions from those included in the BAU case were realized. At all other ports, the RRFs described
115 A "typical port" in this assessment is intended to establish a hypothetical port that that allows EPA to illustrate the relative
impacts of a particular strategy and/or scenario.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
75
-------
Appendix C: Analysis of Emission Reduction Scenarios
above (Equation C-l) by pollutant and scenario were multiplied by each port's BAU emission inventory
to estimate the emission reductions associated with each scenario. There was no change in age
distribution in either 2030 or 2050 at all ports, so the full RRFs were applied.
C.3. Rail
C.3.1. Scenarios
As discussed in Section 5, the rail strategy scenarios were based on the future year BAU distributions of
locomotives, which are shown in Tables C-5 and C-6.
Table C-5. BAU Distribution of Line-Haul Locomotives by Emissions Tier
Tier
2011
2020
2030
2050
Pre-Tier 0
10%
0%
0%
0%
TierO
37%
3%
0%
0%
Tier 0+
19%
33%
10%
0%
Tier 1
4%
0%
0%
0%
Tier 1+
6%
9%
5%
0%
Tier 2
24%
0%
0%
0%
Tier 2+
0%
22%
17%
0%
Tier 3
0%
10%
9%
0%
Tier 4
0%
23%
59%
100%
Total
100%
100%
100%
100%
Table C-6. BAU Distribution of Switcher Locomotives by Tier
Tier
2011
2020
2030
2050
Pre-Tier 0
74%
38%
8%
0%
TierO
7%
1%
0%
0%
Tier 0+
10%
45%
52%
0%
Tier 1
1%
0%
0%
0%
Tier 1+
0%
1%
1%
0%
Tier 2
7%
7%
0%
0%
Tier 2+
0%
0%
6%
0%
Tier 3
1%
3%
3%
0%
Tier 4
0%
5%
29%
100%
Total
100%
100%
100%
100%
Table C-7 shows the strategy scenarios that were analyzed as described in Section 6 and this appendix.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
76
-------
Appendix C: Analysis of Emission Reduction Scenarios
Table C-7. Rail Strategy Scenarios
Strategy
2020/A
2020/B
2030/A
2030/B
2050/A
2050/B
Line-haul—
Technology
Strategies
Replace 50%
of Tier 0+
engines with
Tier 2+
engines
Replace
100% of Tier
0+ engines
with 50% 2+
engines and
50% Tier 4
engines
Replace 100%
of Tier 1+ and
earlier engines
with 50% 2+
engines and
50% Tier 4
engines
Replace all
pre-Tier
engines with
Tier 4 engines.
Replace 10%
of Tier 4 with
zero
emission
locomotive
Replace 25%
of Tier 4 with
zero emission
locomotive
Line-haul—
Operationa
1 Strategies
1%
improvement
in fuel
efficiency
5%
improvemen
t in fuel
efficiency
5%
improvement
in fuel
efficiency
10%
improvement
in fuel
efficiency
10%
improvemen
t in fuel
efficiency
20%
improvement
in fuel
efficiency
Switchers
Replace 50%
of Pre-TierO
engines with
95% Tier 2+
engines and
5% Tier 4
Genset
Replace all
Pre-Tier 0
engines with
90% Tier 2+
and 10% Tier
4 Genset
Replace all
Pre-Tier 0
engines and
20% of Tier 0+
with 90% Tier
2+ engines and
10% Tier 4
Genset
Replace all
Pre-Tier 0
engines and
40% of Tier 0+
with 70% Tier
4 engines and
30% Tier 4
Genset
Assume 30%
Tier 4 Genset
Assume 50%
Tier 4 Genset
C.3.2. Methodology and Assumptions
As described in Section 6, rail emission reductions were calculated by developing an RRF for each
scenario strategy and pollutant, and these RRFs were calculated using average emission rates,
determined as the emission rate for each engine tier weighted by the locomotive engine population
distribution in that tier. An RRF was calculated from the following equation:
RRF = 1 - Scenario EF/BAU EF Eq. C-2
Table C-8 shows an example of how a line-haul locomotive RRF was estimated.
Table C-8. Example of Scenario 2020/A NOx Emission Reduction Factor for Line-Haul Locomotives
Tier
NOx Emission
Pop
Weighted
Pop
Weighted
Relative
Pre-Tier 0
13.0
0%
0.00
0%
0
TierO
8.6
3%
0.27
3%
0.27
Tier 0+
7.2
33%
2.38
17%
1.19
Tier 1
6.7
0%
0.00
0%
0.00
Tier 1+
6.7
9%
0.62
9%
0.62
Tier 2
4.95
0%
0.00
0%
0.00
Tier 2+
4.95
22%
1.08
38%
1.90
Tier 3
4.95
10%
0.50
10%
0.50
Tier 4
1.0
23%
0.23
23%
0.23
100%
5.08
100%
4.71
93%
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
77
-------
Appendix C: Analysis of Emission Reduction Scenarios
In developing these strategies, it was assumed that any new engines would have similar duty cycles,
rated power, and annual usage as the engines they replace, such that emission changes were due solely
to changes in the engine emission rates (e.g., in g/kWh).
For line-haul locomotives, emission rates for NOx, PM2.5, and HC for each line-haul tier were taken from
EPA guidance.116 VOC emission factors were calculated from hydrocarbon (THC) emission rates, using
EPA conversion factors.117 Select air toxics emission factors were computed from the VOC emissions
factors using EPA speciation profiles for on-road engines from the MOVES model.118 Well-to-wheels C02
emission factors (in g/ton-mi) were derived from GREET2015; the Rail emission factors for Freight Rails
were used.119 No change in fuel economy/C02 emission rates between locomotive engine tiers was
assumed. Zero emissions locomotives were assumed to be electric locomotives with only well-to-plug
C02 emissions. RRFs for BC were assumed to be equal to PM2.5 RRFs.
For switcher locomotives, the RRFs were calculated using the emission rate for each tier weighted by the
locomotive engine population distribution in that tier for the BAU emission inventory and strategy
scenarios. While GenSet locomotives can be built to Tier 3 or Tier 4 standards, it was assumed that the
Tier 4 standards were more appropriate for all GenSet locomotives in these scenarios. The RRFs were
calculated as described by Equation C-2 above and the underlying methodology and assumptions were
generally similar to the Line-haul Technology scenarios. For scenarios involving replacements with Tier 4
GenSets, it was assumed the GenSet engines would meet the EPA Nonroad Tier 4 emission standards.
These were assumed to fall under the nonroad emission standards for engine power between 175 and
750 hp.
C.3.3. Application of Relative Reduction Factors
The RRF for each scenario and pollutant was multiplied by the relevant portion of the BAU emissions
inventory for the appropriate analysis year, with the resulting line-haul and switcher locomotive
emissions reductions for each scenario. This method was applied uniformly to all ports within this
national scale analysis, depending upon the level of rail activity.
C.4. Cargo Handling Equipment
The analysis of emission reduction strategies for CHE focused on those equipment types that contribute
the bulk of CHE emissions at most ports: yard tractors, rubber tire gantry (RTG) cranes, and container
116 U.S. Environmental Protection Agency, Technical Highlights: Emission Factors for Locomotives, Report EPA-420-F-09-025,
April 2009.
117 U.S. Environmental Protection Agency, Conversion Factors for Hydrocarbon Emission Components (July 2010), Report EPA-
420-R-10-015 NR-002d.
118 U.S. Environmental Protection Agency, MOVES2010b: Additional Toxics Added to MOVES. EPA-420-B-12-029a, May 2012,
Sec 3.1.1. Available at: http://www.epa.gov/otaq/models/moves/documents/420bl2029a.pdf.
119 GREET2015 was released October 2, 2015. For information on the model, see https://ereet.es.anl.eov/.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
78
-------
Appendix C: Analysis of Emission Reduction Scenarios
handlers (side picks and top picks, which are under the category of "rubber tire loaders" in the NEI). See
Section 5 for further background on CHE strategies.
For each scenario, an RRF was calculated using the emission rate for each engine tier weighted by the
population distribution in that tier. This approach was similar to the drayage and locomotive
replacement strategy scenarios discussed above, such as in Table C-8 for Line-haul Locomotive
Technology scenarios.
C.4.1. Yard Truck Strategy Scenarios
As discussed in Section 5, the yard truck scenarios were based on the NONROAD model's future year
distributions of yard trucks, as shown in Table C-9.
Table C-9. Distribution of Yard Trucks by Tier
Tier
2011
2020
2030
2050
Tier 1
9%
0%
0%
0%
Tier 2
17%
0%
0%
0%
Tier 3
64%
3%
0%
0%
Tier 4
10%
97%
100%
100%
Total
100%
100%
100%
100%
Table C-10 shows the BAU assumptions that are relevant for all yard truck strategy scenarios, since
most, if not all, yard trucks are already assumed to be meeting Tier 4 emission standards in all analysis
years, based on the assumptions for this assessment. Thus, all strategy scenarios focus on the
introduction of battery electric yard truck technologies.
Table C-10. Yard Truck Strategy Scenarios
2020/A
2020/B
2030/A
2030/B
2050/A
2050/B
Replace all Tier
3 with Tier 4
Replace all Tier
3 with Tier 4,
and replace 5%
of Tier 4 with
battery electric
Replace 10%
Tier 4 diesel
with battery
electric
Replace 25%
Tier 4 diesel
with battery
electric
Replace 25% of
Tier 4 diesel
engines with
battery electric
Replace 50% of
Tier 4 diesel
engines with
battery electric
C.4.2. Cranes Strategy Scenarios
As discussed in Section 5, the crane strategy scenarios were also based on the future year distributions
of cranes, as assumed in the NONROAD model and shown in Table C-ll.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
79
-------
Appendix C: Analysis of Emission Reduction Scenarios
Table C-ll. Distribution of RTG Cranes by Tier
Tier
2011
2020
2030
2050
Uncontrolled
6%
1%
0%
0%
Tier 1
27%
3%
0%
0%
Tier 2
20%
5%
1%
0%
Tier 3
38%
17%
2%
0%
Tier 4
9%
74%
98%
100%
Total
100%
100%
100%
100%
Table C-12 shows the RTG crane strategy scenarios that were analyzed, including the significant increase
of electric crane technologies for most analysis years. Since this technology is available today and is
being installed at a number of leading ports, this assessment assumed that electric cranes would be
widely deployed by 2050.
Table C-12. RTG Crane Strategy Scenarios
2020/A
2020/B
2030/A
2030/B
2050/A
2050/B
Replace all
Replace all
Replace all Tier
Replace all Tier
Replace 50%
Replace 75%
Uncontrolled
Uncontrolled,
2 and 3 with
2 and 3 with
Tier 4 with
Tier 4 with
and 50% of Tier
Tier 1 and 2
50% Tier 4 and
50% Tier 4 and
electric
electric
1 and 2 with
with 75% Tier 4
50% electric.
50% electric.
50% Tier 3 and
and 25% electric
Replace 10%
Replace 25%
50% Tier 4
Tier 4 with
Tier 4 with
electric
electric
C.4.3. Container Handler Strategy Scenarios
The container handler strategy scenarios were based on the future year distributions shown in Table
C-13.
Table C-13. Distribution of Container Handlers by Tier
Tier
2011
2020
2030
2050
Uncontrolled
2%
0%
0%
0%
Tier 1
26%
1%
0%
0%
Tier 2
23%
2%
0%
0%
Tier 3
44%
15%
0%
0%
Tier 4
5%
81%
100%
100%
Total
100%
100%
100%
100%
Table C-14 shows the container handler strategy scenarios used in this assessment, as described further
in Section 6.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
80
-------
Appendix C: Analysis of Emission Reduction Scenarios
Table C-14. Container Handler Strategy Scenarios
2020/A
2020/B
2030/A
2030/B
2050/A
2050/B
Replace all Tier
Replace Tier 1
Replace 10% of
Replace 25% of
Replace 50% of
Replace 75% of
1 and 2 engines
and 2 engines
Tier 4 diesel
Tier 4 diesel
Tier 4 diesel
Tier 4 diesel
with 50% Tier 3
with Tier 4
engines with
engines with
engines with
engines with
and 50% Tier 4.
engines.
electric engines
electric
electric
electric
Replace Tier 3
with 50% Tier 4
and 50% elec.
engines
C.4.4. Relative Reduction Factors: Methodology and Assumptions for all CHE
For each scenario, an RRF was calculated using the emission rate for each engine tier weighted by the
population distribution in that tier.120 EPA's emission standards were used as criteria pollutant emission
factors for conventional diesel equipment.121 Nonroad emission standards vary based on the rated
power of the engine:122
¦ Yard trucks were assumed to fall in the rated power category of 130 to 225 kW (175 to 300 hp)
based upon an average engine size of 206 hp.
¦ RTG cranes were assumed to fall in the rated power category of 225 to 450 kW (300-600 hp) based
on an average engine size of 453 hp.
¦ Container handlers were assumed within the rated power category of 130 to 225 kW (175 to 300
hp) based on an average engine size of 184 hp for side handlers and 282 hp for top handlers.
Emission factors for pre-Tier engines were taken from the baseline CHE emission factors used in this
assessment.
VOC emission rates were based off of the hydrocarbon (THC) standards, using EPA conversion factors.123
Select air toxic emission factors were computed from the VOC emission factors using EPA speciation
profiles for on-road engines from the MOVES model.124 Diesel upstream wheel-to-pump C02 emission
factors were calculated using GREET 2015. CHE tailpipe C02 emission factors were obtained from EPA's
120 Note that the baseline and BAU emission inventories for CHE include a mix of fuels. However, the strategy scenarios that
were analyzed imply replacement of diesel engines only. Thus, RRF calculations were made in terms of emission factors for
diesel engines only. However, according to the NONROAD model used, the three types of equipment considered here are
predominately diesel fueled for all years included here, so any bias in the results from this assumption was minimal.
121 U.S. Environmental Protection Agency, Nonroad Compression-Ignition Emission Standards. Available at:
http://www.epa.gov/otaq/standards/nonroad/nonroadci.htm.
122 U.S. Environmental Protection Agency, Current Methodologies in Preparing Mobile Source Port-Related Emission Inventories,
April 2009.
123 U.S. Environmental Protection Agency, Conversion Factors for Hydrocarbon Emission Components (July 2010), Report EPA-
420-R-10-015 NR-002d.
124 U.S. Environmental Protection Agency, MOVES2010b: Additional Toxics Added to MOVES. EPA-420-B-12-029a, May 2012,
Sec 3.1.1. Available at: http://www.epa.eov/otaq/models/moves/documents/420bl2029a.pdf.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
81
-------
Appendix C: Analysis of Emission Reduction Scenarios
Current Methodologies document. For electric equipment, no tailpipe pollutant emissions were
assumed. Upstream electricity emission factors for C02 were calculated using GREET2015 assuming an
average U.S. grid mix for the appropriate scenario years. Well-to-wheels emission factors assumed the
sum of upstream and downstream emissions. RRFs for BC were taken as equal to those for PM2.5.
This method was applied uniformly to all ports in this national scale analysis. However, the calculated
reductions were based on the resolution of the BAU inventories. As with other CHE types, the
calculated reductions are likely to overestimate the potential percent reduction in emissions at ports
that have CHE that is newer than average, while it will underestimate the reductions at ports with older
equipment.
C.4.5. Application of Relative Reduction Factors
Generally, the RRF for each strategy scenario was multiplied by the CHE portion of the BAU emissions
inventory for the appropriate analysis year to determine the emission reductions for each scenario.
In the case of CHE, as described above, the BAU inventory did not include resolution by equipment type
or operating mode with only total CHE emissions presented for each pollutant for each port. This
method was applied uniformly.
C.5. Harbor Craft
The analysis of emission reduction strategies for harbor craft focused on the two types of vessels that
contribute the bulk of harbor craft emissions at most ports: tugs and ferries. See Section 5 for further
background on harbor craft strategies.
C.5.1. Scenarios: Tugs
As described in Section 5, the future year distribution of tugs was estimated using a methodology based
on the growth and scrappage assumptions in EPA's NONROAD model, as shown in Table C-15.
Table C-15. Distribution of Tugs by Tier
Tier
2011
2020
2030
2050
Tier 0/0+
61%
10%
0%
0%
Tier 1/1+
35%
24%
3%
0%
Tier 2/2+
4%
33%
7%
0%
Tier 3/3+
0%
30%
80%
61%
Tier 4
0%
3%
10%
39%
Total
100%
100%
100%
100%
Table C-16 shows the tug strategy scenarios that were analyzed in this assessment. Due to the slower
national distribution of fleet turnover for tugs in the NONROAD model, the strategies analyzed included
more repowers and replacements to cleaner diesel engines for all analysis years, with some opportunity
for hybrid electric technology in 2030 and 2050.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
82
-------
Appendix C: Analysis of Emission Reduction Scenarios
Table C-16. Tug Strategy Scenarios
2020/A
2020/B
2030/A
2030/B
2050/A
2050/B
Repower/
Repower/
Repower/
Repower/
Repower/
Repower/
Replace all Pre-
Replace all Pre-
Replace all Tier
Replace all Tier
Replace 50% of
Replace all Tier
Control engines
Control and Tier
1 and 2 with
1 and 2 with
Tier 3 engines
3 engines with
with Tier 3
1 with Tier 3
Tier 4
Tier 4
with Tier 4
Tier 4 engines
engines
Repower 10% of
Repower/
Repower/
engines
Repower/
Tier 2 with Tier
Replace 25% of
Replace 50% of
Repower/
Replace 25% of
3 hybrid electric
Tier 3 engines
Tier 3 engines
Replace 10% of
Tier 4 with
with Tier 4
with Tier 4
Tier 4 with
hybrid electric
engines
engines
Repower/
Replace 25% of
Tier 4 with
hybrid electric
hybrid electric
C.5.2. Scenarios: Ferries
As with tugs, the future year distributions of ferries were estimated using a methodology based on the
growth and scrappage assumptions in EPA's NONROAD model. Table C-17 shows that distribution.
Table C-17. Distribution of Ferries by Tier
Tier
2011
2020
2030
2050
Tier 0/0+
75%
39%
10%
0%
Tier 1/1+
21%
18%
12%
0%
Tier 2/2+
4%
10%
8%
1%
Tier 3/3+
0%
28%
59%
60%
Tier 4
0%
5%
11%
39%
Total
100%
100%
100%
100%
Table C-18 shows the ferry strategy scenarios. As with tugs, the slower fleet turnover assumed for
ferries in the BAU inventory allowed for cleaner diesel strategies to be analyzed in all analysis years, in
addition to hybrid electric technology in 2030 and 2050.
Table C-18. Ferry Strategy Scenarios
2020/A
2020/B
2030/A
2030/B
2050/A
2050/B
Repower/
Repower/
Repower/
Repower/ Replace
Repower/
Repower/
Replace all
Replace all Pre-
Replace all Tier
all Tier 0,1 and 2
Replace all Tier
Replace all Tier
Pre-Control
Control and Tier
0,1 and 2 with
with Tier 4
2 and 50% of
2 and 3 engines
engines
1 with Tier 3
Tier 4
Repower/ Replace
Tier 3 engines
with Tier 4
with Tier 3
Repower 10% of
Repower/
50% of Tier 3
with Tier 4
engines
engines
Tier 2 with Tier 3
Replace 25% of
engines with Tier 4
engines
Repower/
hybrid electric
Tier 3 engines
engines. Repower/
Repower/
Replace 25% of
with Tier 4
Replace 25% of Tier
Replace 10% of
Tier 4 with
engines
4 with hybrid electric
Tier 4 with
hybrid electric
hybrid electric
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
83
-------
Appendix C: Analysis of Emission Reduction Scenarios
C.5.3. Development and Application of Relative Reduction Factors
For each scenario, an RRF was developed using the emission rate for each engine tier weighted by the
population distribution in that tier. This process is similar to the example for line-haul locomotives
discussed earlier in this appendix. The RRF for each scenario was then multiplied by the BAU emissions
for the given vessel type appropriate analysis year to determine the scenario's emission reductions.
The following paragraphs further describe how this was done in conjunction with more complex
methodology for the harbor craft BAU inventory development.
To determine the BAU emission inventory by vessel type, the relative share of emissions by each vessel
type (tugs or ferries), by pollutant, within the two vessel categories (goods moving or non-goods
moving) was determined for the baseline year (e.g., tugs average 97% of CO emissions from goods
moving vessels nationwide). This share was then assumed to also apply for future years, which was
reasonable since the vessel categories are grown rather than individual vessel types. Next, the share of
emissions by operating mode that were due to vessels in each category was determined. For example,
the product of the tug share of total goods moving emissions by pollutant, along with the share of total
emissions due to goods movements by operating mode, approximated the BAU tug inventory by
pollutant by operating mode in each scenario analysis year. A similar approach was applied for ferries.
Finally, the emission reduction from the application of each strategy was determined as the product of
the vessel and mode-specific BAU inventory and the RRF determined for each strategy scenario. This
method was applied consistently for all strategy scenarios and applied uniformly.
C.6. Ocean Going Vessels
As discussed in Section 6, OGV strategies were grouped in the following scenario categories:
¦ Fuel Changes
¦ Shore Power
¦ Stack Bonnets
¦ Reduced Hoteling Time
Reductions were calculated for all scenarios relative to the BAU inventories, independently. This method
considered reductions separately for each scenario, consistent with other sectors. Note that in practice
results may not be additive as such relative to the BAU case.125 In summary, the reductions presented
here would be reasonable for each individual strategy, but would overestimate the cumulative impact if
multiple strategies were applied simultaneously.
125 For example, if shore power was applied after fuel changes were required, the BAU inventory that would exist in practice for
shore power would be smaller for many pollutants, as would be the reductions.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
84
-------
Appendix C: Analysis of Emission Reduction Scenarios
C.6.1. Fuel Changes Scenarios
The Fuel Change strategy scenarios are found in Table C-19 and Table C-20 for propulsion and auxiliary
engines, respectively.
Table C-19. Fuel Change Strategy Scenarios for OGV Propulsion Engines
Ship Type
2020/A
2020/B
2030/A
2030/B
2050/A
2050/B
Bulk
10% use 500
ppm sulfur
fuel; 2% use
LNG
25% use 500
ppm sulfur
fuel; 10% use
LNG
25% use 200
ppm sulfur
fuel; 4% use
LNG
50% use 200
ppm sulfur
fuel; 15% use
LNG
8% use LNG
25% use LNG
Container
10% use 500
ppm sulfur
fuel; 1% use
LNG
25% use 500
ppm sulfur
fuel; 5% use
LNG
25% use 200
ppm sulfur
fuel; 2% use
LNG
50% use 200
ppm sulfur
fuel; 5% use
LNG
5% use LNG
5% use LNG
Passenger
10% use 500
ppm sulfur
fuel
25% use 500
ppm sulfur
fuel
25% use 200
ppm sulfur
fuel
50% use 200
ppm sulfur
fuel
-
-
Tanker
10% use 500
ppm sulfur
fuel; 2% use
LNG
25% use 500
ppm sulfur
fuel; 10% use
LNG
25% use 200
ppm sulfur
fuel; 4% use
LNG
50% use 200
ppm sulfur
fuel; 15% use
LNG
8% use LNG
25% use LNG
Table C-20. Fuel Change Scenarios for OGV Auxiliary Engines
Ship Type
2020/A
2020/B
2030/A
2030/B
2050/A
2050/B
Bulk
10% use
ULSD; 2% use
LNG
20% use
ULSD; 10%
use LNG
30% use
ULSD; 4% use
LNG
40% use
ULSD; 15%
use LNG
8% use LNG
25% use LNG
Container
10% use
ULSD; 1% use
LNG
20% use
ULSD; 5% use
LNG
30% use
ULSD; 2% use
LNG
40% use
ULSD; 5% use
LNG
5% use LNG
5% use LNG
Passenger
10% use
ULSD
20% use
ULSD
30% use
ULSD
40% use
ULSD
-
-
Tanker
10% use
ULSD; 2% use
LNG
20% use
ULSD; 10%
use LNG
30% use
ULSD; 4% use
LNG
40% use
ULSD; 15%
use LNG
8% use LNG
25% use LNG
LNG is limited in container ships to 5% based upon a study by Lloyds/26 and no LNG use was included for
passenger ships due to passenger safety issues.
126 Lloyds Register Marine, Global Marine Fuel Trends 2030, 2014.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
85
-------
Appendix C: Analysis of Emission Reduction Scenarios
C.6.1.1. Developing and Application of Relative Reduction Factors
RRFs were calculated from the ratio of average emission factors under the BAU scenario and those
under each analysis scenario. RRFs are calculated according to the Equation C-2.
To apply these scenarios to the 2020, 2030, and 2050 BAU emissions inventories, emission factors were
developed by engine type and fuel type. Table C-21 and Table C-22 show these emission factors for
propulsion and auxiliary engines.
Table C-21. Average Emission Factors for Propulsion Engines
Engine Type
Sulfur
Emission Factors (g/kWh)
PM2.5
HC
BC
CO2
SO2
BSFC
MSD
0.10%
0.17
0.5
0.0102
646
0.4
203
0.05%
0.16
0.5
0.0093
646
0.2
203
0.02%
0.15
0.5
0.0088
646
0.08
203
SSD
0.10%
0.17
0.6
0.0104
589
0.36
185
0.05%
0.16
0.6
0.0096
589
0.18
185
0.02%
0.15
0.6
0.0092
589
0.07
185
GT
0.10%
0.25
0.1
0.0151
923
0.57
290
0.05%
0.23
0.1
0.0138
923
0.28
290
0.02%
0.22
0.1
0.0131
923
0.11
290
Otto
LNG
0.03
0.5
0.0105
457
0.003
166
Table C-22. Average Emission Factors for Auxiliary Engines
Sulfur
Emission Factors (g/kWh)
PM2.5
HC
BC
CO2
SO2
BSFC
0.10%
0.17
0.4
0.0101
691
0.42
217
ULSD
0.14
0.4
0.0082
691
0.01
217
LNG
0.03
0.5
0.0105
457
0.003
166
Here, the brake specific fuel consumption (BSFC) for medium speed diesel (MSD), slow speed diesel
(SSD) and gas turbines (GT) was taken from a European Union study (Entec).127 Emission factors and
BSFC for LNG were taken from an IMO study.128 As discussed in Appendix A, PMio was calculated using
the formula:
PMio EF = 0.23 + BSFC x 7 x 0.02247 x (Fuel Sulfur Fraction - 0.0024) Eq. C-3
127 Entec UK Limited, Quantification of Emissions from Ships Associated with Ship Movements between Ports in the European
Community, prepared for the European Commission, July 2002.
128 International Maritime Organization, Third IMO GHG Study, June 2014. Available at:
http://www.imo.org/en/OurWork/Environment/PollutionPrevention/AirPollution/Pages/Relevant-links-to-Third-IMO-GHG-
Studv-2014.aspx.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
86
-------
Appendix C: Analysis of Emission Reduction Scenarios
PM2.5 was taken as 92% of PM10. S02 was calculated using the formula:
SO2 EF = BSFC x 2 x 0.97753 x Fuel Sulfur Fraction Eq. C-4
C02 emission factors were calculated as follows for diesel:
C02 EF = BSFC x 0.868x3.667 Eq. C-5
C02 emission factors were calculated as follows for LNG:129
C02 EF = BSFC x 2.75 Eq. C-6
BC was calculated as 6% of PM2.5 for diesel fuels and 38% of PM2.5 for LNG based upon the EPA's Black Carbon
Report to Congress.130
Note that, consistent with the baseline and BAU emission inventory development, OGV uses total
hydrocarbons (HC) while all other sectors report volatile organic compounds (VOC). HC emission factors for
SSD, MSD, GT and ST engines came from Entec.
Average NOx emission factors by engine type for 2020 and 2030 were calculated as discussed in Appendix B.
They are listed by calendar year and engine type in Table C-23 for propulsion engines and by calendar year
and ship type in Table C-24 for auxiliary engines. LNG NOx emission factors were taken from the IMO study.
Table C-23. Average Propulsion NOx Emission Factors
Engine Type
NOx (g/kWh)
2020
2030
MSD
9.4
3.7
SSD
10.6
5.0
GT
5.7
5.7
LNG Otto
1.3
1.3
Table C-24. Average Auxiliary NOx Emission Factors
Engine Type
NOx (g/kWh)
2020
2030
Passenger
10.3
3.7
Other
8.6
4.1
LNG Ships
1.3
1.3
129 International Maritime Organization, Third IMO GHG Study, June 2014.
130 U.S. Environmental Protection Agency, Report to Congress on Black Carbon, EPA-450/D-11-001, March 2011.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
87
-------
Appendix C: Analysis of Emission Reduction Scenarios
To calculate aggregate emission factors for the BAU and fuel scenarios, engine-type weighting factors
were needed. Using the combined Entrances and Clearances data combined with Lloyd's data for the
ports, weighting factors were calculated by ship type. Table C-25 shows these factors.131
Table C-25. Engine Type Weighting Factors by Ship Type
Ship Type
MSD
SSD
GT
Bulk Carrier
0%
100%
0%
Container
3%
97%
0%
Passenger
89%
0%
11%
Tanker
3%
97%
0%
To calculate emission reductions for the scenarios listed in Table C-19 and Table C-20, BAU emission
inventories were separated into propulsion and auxiliary engine emissions for the four ship types. Emissions
related to propulsion engines during reduced speed zone (RSZ) and maneuvering modes were combined into
the propulsion engine emissions. Emissions related to auxiliary engines during RSZ, maneuvering and hoteling
modes were combined into the auxiliary engine emissions. In addition, hoteling-only emissions (from
auxiliary engines) were also calculated by ship type to use for strategies that these emissions.
Table C-26 presents average emissions factors by ship type calculated from the BAU emission inventories
(i.e., "BAU EF" in Equation C-2). BAU emissions assume the use of 1,000 ppm S MDO/MGO in both
propulsion and auxiliary engines; BAU C02 emission factors stay constant through 2050. Table C-27 through
Table C-31 show the average emission factors by ship type for the scenarios (i.e., "Scenario EF" in Equation C-
2). See Section 6 for the RRFs that were calculated for each scenario.
Table C-26. BAU Average Emission Factors by Ship Type and Calendar Year
Engine
Ship Type
CY
Combined Emission Factors (g/kWh)
NOx
PM2.5
HC
BC
C02
SO2
Propulsion
Bulk
2020
10.6
0.17
0.6
0.010
589
0.36
2030
5.0
0.17
0.6
0.010
589
0.36
Container
2020
10.6
0.17
0.6
0.010
591
0.36
2030
4.9
0.17
0.6
0.010
591
0.36
Passenger
2020
9.0
0.18
0.5
0.011
677
0.42
2030
3.9
0.18
0.5
0.011
677
0.42
Tanker
2020
10.6
0.17
0.6
0.010
591
0.36
2030
4.9
0.17
0.6
0.010
591
0.36
Auxiliary
Bulk
2020
8.6
0.17
0.4
0.010
691
0.42
2030
4.1
0.17
0.4
0.010
691
0.42
Container
2020
8.6
0.17
0.4
0.010
691
0.42
2030
4.1
0.17
0.4
0.010
691
0.42
Passenger
2020
10.3
0.17
0.4
0.010
691
0.42
2030
3.7
0.17
0.4
0.010
691
0.42
Tanker
2020
8.6
0.17
0.4
0.010
691
0.42
2030
4.1
0.17
0.4
0.010
691
0.42
131 Steam turbines (ST) were not significant and included here.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
88
-------
Appendix C: Analysis of Emission Reduction Scenarios
Table C-27. Average Emission Factors by Ship Type for Fuel Scenario 2020/A
Engine
Vessel
Combined Emission Factors (g/kWh)
NOx
PM2.5
HC
BC
C02
SO2
Propulsion
Bulk
10.4
0.17
0.6
0.010
586
0.34
Container
10.5
0.17
0.6
0.010
589
0.34
Passenger
9.0
0.18
0.5
0.011
677
0.39
Tanker
10.4
0.17
0.6
0.010
588
0.34
Auxiliary
Bulk
8.5
0.16
0.4
0.010
686
0.37
Container
8.6
0.16
0.4
0.010
688
0.38
Passenger
10.3
0.16
0.4
0.010
691
0.38
Tanker
8.5
0.16
0.4
0.010
686
0.37
Table C-28. Average Emission Factors by Ship Type for Fuel Scenario 2020/B
Engine
Vessel
Combined Emission Factors (g/kWh)
NOx
PM2.5
HC
BC
C02
SO2
Propulsion
Bulk
9.7
0.16
0.6
0.010
576
0.28
Container
10.1
0.16
0.6
0.010
584
0.3
Passenger
9.0
0.18
0.5
0.011
677
0.36
Tanker
9.7
0.16
0.6
0.010
577
0.28
Auxiliary
Bulk
7.9
0.15
0.4
0.010
667
0.3
Container
8.3
0.15
0.4
0.010
679
0.32
Passenger
10.3
0.16
0.4
0.010
691
0.32
Tanker
7.9
0.15
0.4
0.010
667
0.3
Table C-29. Average Emission Factors by Ship Type for Fuel Scenario 2030/A
Engine
Vessel
Combined Emission Factors (g/kWh)
NOx
PM2.5
HC
BC
C02
SO2
Propulsion
Bulk
4.8
0.16
0.6
0.010
584
0.28
Container
4.9
0.17
0.6
0.010
588
0.28
Passenger
3.9
0.17
0.5
0.010
677
0.33
Tanker
4.8
0.16
0.6
0.010
585
0.28
Auxiliary
Bulk
4.0
0.15
0.4
0.010
681
0.28
Container
4.0
0.16
0.4
0.010
686
0.29
Passenger
3.7
0.16
0.4
0.010
691
0.3
Tanker
4.0
0.15
0.4
0.010
681
0.28
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports 89
-------
Appendix C: Analysis of Emission Reduction Scenarios
Table C-30. Average Emission Factors by Ship Type for Fuel Scenario 2030/B (g/kWh)
Engine
Vessel
Combined Emission Factors (g/kWh)
NOx
PM2.5
HC
BC
CO2
SO2
Propulsion
Bulk
4.4
0.14
0.6
0.010
569
0.16
Container
4.7
0.16
0.6
0.010
584
0.2
Passenger
3.9
0.17
0.5
0.010
677
0.25
Tanker
4.4
0.14
0.6
0.010
570
0.16
Auxiliary
Bulk
3.7
0.13
0.4
0.009
656
0.19
Container
3.9
0.15
0.4
0.009
679
0.24
Passenger
3.7
0.16
0.4
0.009
691
0.26
Tanker
3.7
0.13
0.4
0.009
656
0.19
Table C-31. Average C02 Emission Factors for Fuel Scenarios 2050/A and 2050/B
Engine
Vessel
Scenario EFs (g/kWh)
2050/A
2050/B
Propulsion
Bulk
578
556
Container
584
584
Passenger
677
677
Tanker
580
557
Auxiliary
Bulk
672
632
Container
679
679
Passenger
691
691
Tanker
672
632
C.6.2. Shore Power Scenarios
Table C-32 presents the Shore Power strategy scenarios for container, passenger and reefer ships that
stop at the ports that were part of this national scale analysis.132
Table C-32. Shore Power Strategy Scenarios
Ship Type
2020/A
2020/B
2030/A
2030/B
2050/A
2050/B
Container
1%
10%
5%
20%
15%
35%
Passenger
10%
20%
20%
40%
30%
60%
Reefer
1%
5%
5%
10%
10%
20%
Table C-32 shows the penetration rate for the three ship types for each scenario, defined in terms of installed
auxiliary power. To correspond with CARB's shore power regulation/33 shore power was applied to
container, passenger and reefer ships in this assessment.
132Shore power was not applied to RoRos as hoteling emissions from RoRos was much smaller than the other three ship types.
133 CARB, Airborne Toxic Control Measure for Auxiliary Diesel Engines Operated on Ocean-Going Vessels At- Berth in a California
Port, Final Regulation Order, 2010. Available at: http://www.arb.ca.gov/ports/shorepower/finalregulation.pdf.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
90
-------
Appendix C: Analysis of Emission Reduction Scenarios
C.6.2.1. Defining Frequent Callers
Percentages in Table C-32 represent the percentage of installed auxiliary power that shore power is applied to
for frequent by ship type at each port from the 2011 Entrances and Clearances data.134 Installed power directly
related to emissions for a given ship type, so by specifying the percent of installed power related to frequent
callers, the amount of eligible frequent caller emissions was estimated.
Table C-33 shows the resulting percentages due to frequent callers by port and ship type. Port-vessel
combinations marked N/A had no ships of that ship type stop at the port.
Table C-33. Frequent Caller135 Percentages by Port and Ship Type
Port
Ship Type
% Frequent
Caller
Port
Ship Type
% Frequent
Caller
New York / New Jersey
Container
62%
Baltimore
Container
31%
Passenger
93%
Passenger
97%
Reefer
87%
Reefer
N/A
New Orleans
Container
57%
Norfolk
Container
48%
Passenger
99%
Passenger
84%
Reefer
0%
Reefer
N/A
Miami
Container
62%
Philadelphia
Container
40%
Passenger
98%
Passenger
N/A
Reefer
83%
Reefer
56%
South Louisiana
Container
0%
Charleston
Container
57%
Passenger
N/A
Passenger
83%
Reefer
N/A
Reefer
0%
Seattle
Container
65%
Corpus Christi
Container
N/A
Passenger
97%
Passenger
N/A
Reefer
N/A
Reefer
N/A
Baton Rouge
Container
0%
Tampa
Container
72%
Passenger
N/A
Passenger
100%
Reefer
N/A
Reefer
0%
Port Arthur
Container
N/A
Savannah
Container
53%
Passenger
N/A
Passenger
0%
Reefer
N/A
Reefer
N/A
Portland
Container
75%
Coos Bay
Container
N/A
Passenger
N/A
Passenger
N/A
Reefer
N/A
Reefer
N/A
Mobile
Container
41%
San Juan
Container
75%
Passenger
97%
Passenger
93%
Reefer
0%
Reefer
95%
Houston
Container
61%
Port Average
Container
56%
Passenger
0%
Passenger
96%
Reefer
0%
Reefer
72%
134 U.S. Army Corps of Engineers, Vessel Entrances and Clearances. Available at:
http://www.navigationdatacenter.us/data/dataclen.htm.
135 Frequent callers were defined for this assessment as individual vessels calling at a port 6 times or more times per year, and
for passenger ships, 5 calls or more per year.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
91
-------
Appendix C: Analysis of Emission Reduction Scenarios
C.6.2.2. Relative Reduction Factor
Emission reductions for each ship type at the port are calculated as the BAU emissions times the RRF
where RRF is defined as:
RRF = FC x PR x Eff Eq. C-7
Where
RRF is the relative reduction factor,
FC is the percent of installed power for frequent callers,
PR is the technology penetration levels, and
Eff is the emission reduction effectiveness (shore power emission reduction per call).
The assessment assumed approximately 2 hours to connect and disconnect cables during a call, and the
strategy's effectiveness was based upon the number of hours connected versus the total hoteling time.
Average hoteling times by vessel type were used to calculate effectiveness by ship type, and then the
resulting valued to all ports with applicable vessel types. The same share of installed power by ship type
by port was also applied for all future years. Shore power effectiveness was based on the number of
hours connected divided by total average hoteling time. The number of hours connected was calculated
as the total average hoteling time minus 2 hours.
Table C-34 shows per call effectiveness for shore power by ship type, considering only emissions from
the vessels themselves.
Table C-34. Shore Power Effectiveness for Vessel Emissions Only, per call
Ship Type
Average Hoteling Time (hrs)
Shore Power Reduction (%)
Container
30.7
93%
Passenger
10.1
80%
Reefer
64.3
97%
In addition to vessel emissions, C02 and criteria pollutant emissions were assumed to be generated by
power plants generating electricity for the shore power technology. Based upon the default U.S.
average generation mix136 using the Argonne National Laboratory's GREET 1 2015 model137, emission
136 The C02 emission rates for this assessment were calculated based on GREET, assuming an average U.S. generation mix. In
practice, the generation mix could be significantly different. For example, in the Northwest, much of the electricity comes
from hydropower, and therefore, those utilities emit less C02 overall.
137 Argonne National Laboratories, GREET Model 2015, https://ereet.es.anl.eov/.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
92
-------
Appendix C: Analysis of Emission Reduction Scenarios
factors for electricity generation138 139 are shown in Table C-35. These should be compared to the
auxiliary engine emission factors to assess the net effectiveness of shore power, considering the time
spent plugged in for a given call.
Table C-35. Power Plant Emission Factors at Plug (g/kWh)
Year
NOx
PMio
PM2.5
HC
BC
co2
SO2
2020
0.119
0.037
0.015
0.004
0.001
489
0.67
2030
0.124
0.040
0.016
0.005
0.001
478
0.633
2050
-
-
-
-
-
460
-
Similar to the special cases defined for drayage vehicles, the BAU emissions accounted for the use of
shore power currently planned, as described in the BAU methodology.140 BAU emissions for the two
ports were modified to separate out power plant emissions from auxiliary engine emissions for those
ships that use shore power and power plant criteria pollutant emissions were added. See Section 6 for
more information on the resulting RRFs and Shore Power strategy scenario results.
C.6.3. Advanced Marine Emission Control System Scenarios
Advanced Marine Emission Control Systems (AMECS, and sometimes referred to as "stack bonnets") can
also provide emission reductions while a ship is at berth. Table C-36 shows the penetration rates for the
qualifying vessels for each AMECS strategy scenario analyzed in this assessment. Again, the percentages
represent installed auxiliary power as a surrogate for hoteling emissions.
Table C-36. AMECS Strategy Scenarios
Ship Type
2020/A
2020/B
2030/A
2030/B
Container
1%
5%
5%
10%
Tanker
1%
5%
5%
10%
As discussed in Section 6, the AMECS strategy scenarios were targeted to only non-frequent callers for
container and tanker ship types. Table C-37 shows the percentages due to non-frequent callers by port
and ship type in this national scale analysis. Port-vessel combinations marked N/A had no ships of that
ship type stop at the port. Overall, non-frequent callers were 47%.
138 GREET2015 only extends to 2040, so the 2040 C02 emission rate for power plants was used for 2050.
139 Note that cargo loading and unloading occur while the connection is being made and removed, so the total hoteling time
estimate is expected to be unchanged by shore power, although on a first call at a new terminal commissioning is required,
which takes much longer. This calculation assumes 2 hours per call for connection and disconnection, but does not include
any commissioning time.
140 Only the Ports of Seattle and New York/New Jersey had installed shore power at the time of this assessment, and only for
passenger ships.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
93
-------
Appendix C: Analysis of Emission Reduction Scenarios
Table C-37. Non-Frequent Caller Percentages by Port and Ship Type
Port
Ship Type
% Non-
Frequent
Caller
Port
Ship Type
% Non-
Frequent
Caller
New York / New Jersey
Container
38%
Baltimore
Container
69%
Tanker
83%
Tanker
100%
New Orleans
Container
43%
Norfolk
Container
52%
Tanker
97%
Tanker
100%
Miami
Container
38%
Philadelphia
Container
60%
Tanker
100%
Tanker
87%
South Louisiana
Container
100%
Charleston
Container
43%
Tanker
94%
Tanker
95%
Seattle
Container
35%
Corpus Christi
Container
N/A
Tanker
100%
Tanker
87%
Baton Rouge
Container
100%
Tampa
Container
28%
Tanker
87%
Tanker
66%
Port Arthur
Container
N/A
Savannah
Container
47%
Tanker
61%
Tanker
100%
Portland
Container
25%
Coos Bay
Container
N/A
Tanker
100%
Tanker
N/A
Mobile
Container
59%
San Juan
Container
25%
Tanker
95%
Tanker
50%
Houston
Container
39%
Port Average
Container
44%
Tanker
70%
Tanker
81%
See Section 6 for further details on the AMECS strategy scenarios and results.
C.7. Sector-by-sector Review of Results
C.7.1. Absolute Emission Reductions
The following figures show the absolute emission reductions obtained from applying each strategy as
described above.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
94
-------
Appendix C: Analysis of Emission Reduction Scenarios
° 200,000
100,000
0
Figure C-l. Absolute Emissions Reductions from the Drayage Sector
NOv PM
2.5
3,000 250
2,500
i- 2,000
a>
> 1,500
£
I- 1,000
500
0
200
<3 150
>
¦
V)
O 100
1—
50
0
I Technological
I Operational
¦ I
2020/A 2020/B 2030/A 2030/B 2020/A 2020/B 2030/A 2030/B
co2 voc
600,000 160
140
500,000 ¦
120
- 400,000 ¦ i-
ro ¦ ro 100
(U ¦ 80
J.L l 1.1
2020/A 2020/B 2030/A 2030/B 2050/A 2050/B 2020/A 2020/B 2030/A 2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
95
-------
Appendix C: Analysis of Emission Reduction Scenarios
0.2
0.0
Figure C-l. Absolute Emissions Reductions from the Drayage Sector (Continued)
¦ Technological
BC Acetaldehyde
¦ Operational
200
180
160
140
jg 120 H H S 3
£ !00 " "
o 80
I I
60
40
20
0
2020/A 2020/B 2030/A 2030/B 2020/A 2020/B 2030/A 2030/B
Benzene Formaldehyde
1.2 8
1.0
i- 0.8 __ _ r
ro (D 5
0-6
c _
o 3
l- 0.4 •-
2020/A 2020/B 2030/A 2030/B 2020/A 2020/B 2030/A 2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
96
-------
Appendix C: Analysis of Emission Reduction Scenarios
25
(0
£ 20
U)
= 15
o
I—
10
5
0
Figure C-2. Absolute Emissions Reductions from the Rail Sector
¦ Line Haul Operational
NOx ^^2 5 ¦ Line Haul Technology
600 16
14
500 ~
12
- 400 __ -
S ¦¦ S 10
> 300
c
o
I- 200
100
I Switcher
.1 II .1 II ll II .1
0 "" ~ ~ ~ ~ ~ ~ ~ 0
2020/A 2020/B 2030/A 2030/B 2020/A 2020/B 2030/A 2030/B
VOC BC
35 12
30
10
.1 I .
2020/A 2020/B 2030/A 2030/B 2020/A 2020/B 2030/A 2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
97
-------
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-2. Absolute Emissions Reductions from the Rail Sector (Continued)
50,000
45,000
40,000
35,000
i- '
$ 30,000
> 25,000
O 20,000
15,000
10,000
5,000
0
0.35
0.3
0.25
(D
£ 0.2
U)
= 0.15
o
I—
0.1
0.05
0
CO,
2020/A
I _
2020/B
I _
2030/A
Benzene
In
.hi
2030/B 2050/A 2050/B
II ll I
1.4
1.2
1
(D
« 0.8
U)
= 0.6
o
I—
0.4
0.2
0
ro
ai
>-
3
2.5
2
1.5
1
0.5
0
Acetaldehyde
I Line Haul Operational
I Line Haul Technology
I Switcher
I
I I J
2020/A
2020/B
2030/A
2030/B
Formaldehyde
2020/A
2020/B
2030/A
2030/B
,1
2020/A
2020/B
.1
2030/A
2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
98
-------
Appendix C: Analysis of Emission Reduction Scenarios
600
o
200
100
0
18
16
14
II
Figure C-3. Absolute Emissions Reductions from the CHE Sector
¦ Container Handler
NOx PM
700 20
18
16
500 ¦ 14
fU OJ n
0) 400 ^ _ « 12
300
2.5 ¦ RTG Crane
¦ Yard Tractor
III ill .1. ill i ill ill all
2020/A 2020/B 2030/A 2030/B 2020/A 2020/B 2030/A 2030/B
VOC BC
20 16
(0 -in (0 10
dj Iz 10
O 8
14
12
III .11 .1.
2020/A 2020/B 2030/A 2030/B 2020/A 2020/B 2030/A 2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
99
-------
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-3. Absolute Emissions Reductions from the CHE Sector (Continued)
450,
400,
350,
300,
250,
200,
150,
100,
50,
-50,
000
000
000
000
000
000
000
000
000
0
000
CO,
2020/A
..I
2020/B 2030/A
0.25
0.2
$ 0.15
>-
l/>
o o.l
I-
0.05
0
Acetaldehyde
I Container Handler
I RTG Crane
I Yard Tractor
2030/B 2050/A 2050/B
2020/A
2020/B
2030/A
2030/B
0.6
0.5
0.4
0.3
0.2
0.1
0
Benzene
II. Ill
ro
a>
>-
0.6
0.5
0.4
0.3
0.2
0.1
0
Formaldehyde
2020/A
2020/B
2030/A
2030/B
2020/A
2020/B
2030/A
2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
100
-------
Appendix C: Analysis of Emission Reduction Scenarios
200
$ 150
>-
(/)
O 100
I-
50
0
Figure C-4. Absolute Emissions Reductions from the Harbor Craft Sector
¦ Ferry
NOx PM2.5
I Tug
9,000 400
8,000 350
7,000 300
>- 6,000 __ __ _
as as 250
a) a)
> 5,000
c 4,000
° 3,000
2,000
¦¦ ¦¦ ¦¦ ¦¦
_
¦
1
1 ¦
1 ¦
2020/A
2020/B 2030/A
voc
2030/B
1
1
1
¦
.1
.1
.1
£ 200
O 150
100
250 300
250
200
150
100
50
.1 .1
2020/A 2020/B 2030/A 2030/B
BC
.1 .1 .1
2020/A 2020/B 2030/A 2030/B 2020/A 2020/B 2030/A 2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
101
-------
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-4. Absolute Emissions Reductions from the Harbor Craft Sector (Continued)
200,
180,
160,
140,
$ 120,
> 100,
O 80,
60,
40,
20,
000
000
000
000
000
000
000
000
000
000
0
CO,
2020/A
¦
2020/B
, J
7
6
ro 5
0)
£ 4
c
,2 3
2
1
0
Acetaldehyde
I Ferry
I Tug
2030/A 2030/B 2050/A 2050/B
.1
2020/A
I .1
2020/B
2030/A
2030/B
1.8
1.6
1.4
i- 1.2
ra
^ 1.0
"i" 0.8
o
I- 0.6
0.4
0.2
0.0
Benzene
i
I
18
16
14
12
10
8
6
4
2
0
Formaldehyde
2020/A
2020/B
2030/A
2030/B
.1
2020/A
2020/B
2030/A
2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
102
-------
1,400
1,200
1,000
800
600
400
200
0
160
140
120
100
80
60
40
20
0
-20
-40
-60
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-5. Absolute Emissions Reductions from the Ocean Going Vessels Sector
.1.
2020/A
2020/A
NO„
¦ i
ll .L. ll.
2020/B
HC
..
!)20/B
2030/A
2030/B
... 1.1
2030/A |)30/B
100
90
80
70
60
50
40
30
20
10
0
4
3
3
1_
(0
(Ll 9
> Z
lAMECS ¦ Fuel Change (Auxiliary) ¦ Fue) Change {Propulsion)
PM
2.5
1
1
=t
... J
Li.l
Li
11
.
2020/A
2020/B
2030/A
BC
2030/B
1
.1 ll ll-ll ll
Ll
ll.
1
2020/A 2020/B 2030/A
I Reduced Hotel in g Time I Shore Power
2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
103
-------
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-5. Absolute Emissions Reductions from the Ocean Going Vessels Sector (Continued)
CO,
so,
300,
250,
200,
ra
£ 150
100
50
-50
000
000
000
000
000
000
0
000
JL
1,1
l,
... i_fa _. n l_ln
LI
.1
2020/A 2020/B 2030/A 2030/B 2050/A 2050/B
¦ AMECS ¦ Fuel Change (Auxiliary) ¦ Fue) Change (Propulsion) BReduced HotelingTime ¦ Shore Power
900
800
700
600
£ 500
^ 400
£ 300
° 200
100
0
-100
-200
2020/A
I
2020/B
2030/A
2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
104
-------
Appendix C: Analysis of Emission Reduction Scenarios
C.7.2. Relative Emission Reductions
The following figures show the percentage reductions obtained from applying each strategy as described above. Note that the reductions are
relative to the applicable and relevant portion of the BAU inventories.
Figure C-6. Relative Emissions Reductions from the Drayage Sector
70%
NO„
70%
PM
2.5
I Technological
I Operational
3 60%
<
m
g 50%
o
£ 40%
C
•£ 30%
u
(U tso/o
cc
s? 10%
o%
I
2020/A
2020/B
2030/A
2030/B
3 60%
<
m
g 50%
o
£ 40%
C
•£ 30%
u
(U tso/o
cc
s? 10%
o%
2020/A
2020/B
2030/A
2030/B
40%
D 35%
m 30%
O 25%
4—
= 20%
« 15%
T3
10%
^ 5%
0%
voc
I
70%
3 60%
<
m
g 50%
o
£ 40%
c
•B 30%
(U tso/o
cc
s? 10%
o%
BC
2020/A
2020/B
2030/A
2030/B
2020/A
2020/B 2030/A 2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
105
-------
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-6. Relative Emissions Reductions from the Drayage Sector (Continued)
14.0%
^ 12.0%
CO
£ 10.0%
O
£ 8.0%
TS
-------
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-7. Relative Emissions Reductions from the Rail Sector
70%
3 60%
<
CO
£ 50%
o
£ 40%
£
•£ 30%
0
1 20%
cc
S? 10%
0%
70%
3 60%
<
CO
£ 50%
g
¦t 40%
£
•£ 30%
u
I 20%
cc
S? 10%
0%
NO„
.1
2020/A
ll .1
2020/B 2030/A 2030/B
voc
ll
2020/A
80%
D 70%
<
00 60%
E
O 50%
M-
= 40%
u 30%
« 20%
S? 10%
0%
80%
D 70%
<
130 60%
E
g 50%
M-
= 40%
u 30%
« 20%
S? 10%
0%
ll
2020/A
2020/B
2030/A
2030/B
ll
2020/A
PM
2.5
2020/B
I Line Haul Operational
I Line Haul Technology
I Switcher
ll
2030/A
2020/B
¦I
2030/A
2030/B
2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
107
-------
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-7. Relative Emissions Reductions from the Rail Sector (Continued)
25%
D
< 20%
CO
E
2 15%
C
o
tJ 10%
3
"O
ai
* 5%
0%
CO,
1.1
100%
90%
80%
70%
60%
| 50%
ts 40%
30%
20%
10%
0%
D
<
CO
E
p
T3
ai
cc
Acetaldehyde
I Line Haul Operational
I Line Haul Technology
I Switcher
¦I ll .1 ll
2020/A 2020/B 2030/A 2030/B 2050/A 2050/B
2020/A
2020/B
2030/A
2030/B
12%
< 10%
CO
I 8%
S 6%
¦O
-------
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-8. Relative Emissions Reductions from the CHE Sector
70%
3 60%
<
CO
£ 50%
o
£ 40%
£
•£ 30%
0
1 20%
cc
S? 10%
0%
NO„
70%
3 60%
<
CO
£ 50%
o
£ 40%
£
•£ 30%
0
1 20%
cc
S? 10%
0%
2020/A
2020/B
2030/A
2030/B
ill
2020/A
PM
2.5
I Container Handler
I RTG Crane
I Yard Tractor
ill
2030/A
2030/B
voc
30%
< 25%
CO
I 20%
= 15%
TS
-------
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-8. Relative Emissions Reductions from the CHE Sector (Continued)
60%
< 50%
CO
I 40%
= 30%
-a
a>
cc
20%
10%
0%
CO,
2020/A 2020/B 2030/A 2030/B 2050/A 2050/B
30%
25%
20%
ca
E
o
£ 15%
£
o
10%
u
3
"S 5%
0%
Acetaldehyde
-5%
2020/A
2020/B
I Container Handler
I RTG Crane
I Yard Tractor
ll. Ill
2030/A
2030/B
30%
< 25%
CO
I 20%
= 15%
TS
-------
Appendix C: Analysis of Emission Reduction Scenarios
= 30%
= 20%
T5
(D
^ 10%
Figure C-9. Relative Emissions Reductions from the Harbor Craft Sector
NOx PM2.5
70% 70%
3 60% H 3 60%
CO CO
£ 50% h H £ 50%
40% ¦ ¦ H 40%
Ilg I. II
22»
_ ¦¦ ¦ ¦ ¦¦
50%
40%
•j5 30%
I. II II II != I. II II
I
I Ferry
I Tug
0% 0%
2020/A 2020/B 2030/A 2030/B 2020/A 2020/B 2030/A 2030/B
VOC BC
60% — 70%
< 50% _ I < 60%
| 40%
0% 0%
2020/A 2020/B 2030/A 2030/B 2020/A 2020/B 2030/A 2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
111
-------
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-9. Relative Emissions Reductions from the Harbor Craft Sector (Continued)
3.5%
3 3.0%
<
CO
£ 2.5%
O
£ 2.0%
£
£ 1.5%
0
1 1-0%
cc
S? 0.5%
0.0%
CO,
I IIII
2020/A 2020/B 2030/A 2030/B 2050/A 2050/B
45%
D 40%
m 35%
| 30%
^ 25%
C
•2 20%
-i 15%
cc 10%
^ 5%
0%
Acetaldehyde
I Ferry
I Tug
I I
2020/A
2020/B
2030/A
2030/B
Benzene
50%
D 45%
< 40%
CO
£ 35%
2 30%
| 25%
13 20%
¦i 15%
ai
oe 10%
^ 5%
0%
I
I
35%
3 30%
<
ca
£ 25%
g
¦t 20%
£
•£ 15%
u
I 10%
cc
S? 5%
0%
Formaldehyde
2020/A
2020/B
2030/A
2030/B
2020/A
2020/B
2030/A
2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
112
-------
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-10. Relative Emissions Reductions from the Ocean Going Vessels Sector
NOx PM2.5
8% 12%
10%
D 7% D
< <
00 s% 00
E E
8%
2 5% m O
M-
4% — ¦ = 6%
i 3%
| 2% ¦ .1 1
S? 1%
4%
¦III II I llll ! la mlml ilii I
0% ™ ™ ~ ~ ~ ~ ™ ~ ~ ~ ~ 0%
2020/A
2020/B
HC
2030/A
2030/B
7% 5%
6% m— < 4%
D ^ 4%
2 5% ^ 3%
E 4% £ 3o/o
3= 3% .2 2%
O io/_ 3 2%
2020/A
2020/B
co2
2030/A
2030/B
. TO/
+3 Z/o ¦¦ ¦ !¦ -o
« ¦ (U 1%
» 1% I I I I I I I *1
^ 0% ¦ ¦ ¦ ¦ ¦ ¦ Q%
3
1
II 1
II II
||
.10/ 2020/A |)20/B 2030/A |)30/B ^ ^
«o\ opv .oV ,o\
-2% # # ^ #
¦ AMECS ¦ F uel Change (AuxiIiary) ¦ FueIChange (Propulsion) ¦ R educed HoielingTime I Shore Power
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
113
-------
Appendix C: Analysis of Emission Reduction Scenarios
Figure C-10. Relative Emissions Reductions from the Ocean Going Vessels Sector (Continued)
BC S02
2020/A
2020/B
2030/A
2030/B
2020/A
2020/B
2030/A
2030/B
¦ AMECS ¦ Fuel Change (Auxiliary) ¦ Fue) Change (Propulsion) ¦ Reduced Hotel in g Time I Shore Power
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
114
-------
Appendix D: Stratified Summary of Results
Appendix D. Stratified Summary of Results
D.l. Additional Details for Stratification Analysis
The results of this assessment were stratified in a number of ways to examine which types of
strategies may have more potential to reduce emissions at different kinds of ports. This analysis was
performed separately for the OGV and non-OGV sectors due to the nature of the various strategies
applied in each sector. This appendix provides more details on some aspects of this analysis as well
as additional charts of stratification results.
The ports were classified as "container" if their cargo throughput was greater than 100,000 twenty-
foot equivalent units (TEUs). Container ports were further classified as "small" if their cargo
throughput was less than 1 million TEUs and "large" if it was more. Additionally, ports were
considered "bulk" if their non-container throughput was greater than 20,000 tons per year (tpy);
the cutoff between large and small bulk ports was 50,000 tpy. Finally, ports were classified as
"passenger" based on engineering judgement. Large passenger ports were ports with more than
750,000 annual passengers. The TEU and tonnage data by port came from U.S. Army Corps of
Engineers' Waterborne Commerce Statistics Center.
Each classification was made independently of the others, so that each port might fall into any
number of categories and may have different size distinctions. For example, a port could be labeled
as both a small passenger port and a large container port. However, it is important to note that
these classifications and distinctions are not official determinations, but are simply used in the
stratification analysis to differentiate generally between the different kinds of ports included in this
assessment. The distinctions "large" and "small" only serve to compare between ports in this
assessment and do not facilitate other comparisons. The cutoff points between the two distinctions
were chosen such that the large and small ports within a classification contained a roughly equal
number of ports.
A listing of ports that fall under each type and size category may be found in Table D-l. This analysis
aggregated the emissions across these 19 port areas to examine the potential impacts of emission
reduction strategy scenarios at the national scale; this assessment (including the stratification
analysis) does not provide specific data for local decision-making at individual ports or specific
neighborhoods.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
115
-------
Appendix D: Stratified Summary of Results
Table D-l. Port Classification by Type and Size for Stratification Analysis Only
Type
Size
Port
Port of New York and New Jersey
Port of South Louisiana
Port of Savannah
Large
Port of Seattle
Port of Hampton Roads (Norfolk)
Port of Houston
Container
Port of Charleston
Port of San Juan, PR
Port of Miami
Port of Baltimore
Small
Port of New Orleans
Port of Philadelphia
Port of Portland, OR
Port of Mobile
Port of South Louisiana
Port of Houston
Port of New York and New Jersey
Large
Port of New Orleans
Port of Corpus Christi
Port of Baton Rouge
Bulk
Port of Mobile
Port of Baltimore
Port of Hampton Roads (Norfolk)
Port of Savannah
Small
Port Tampa Bay
Port of Philadelphia
Port Arthur
Port of Portland, OR
Port of Miami
Large
Port of New Orleans
Port Tampa Bay
Passenger
Port of Seattle
Port of New York and New Jersey
Small
Port of Baltimore
Port of San Juan, PR
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
116
-------
Appendix D: Stratified Summary of Results
D.2. Additional Charts for Stratification Analysis of OGV Sector
D.2.1. Entire OGV Sector
Figure D-l. Comparing NOx Relative Reduction Potential of the OGV Sector
.3
6%
4%
2%
0%
'
w
i
c 6%
OJ
g4%
.2 2%
a 0%
"O
£
6%
496
2%
¦ Uil^
—¦ I—11
1 -U J h*
¦ U-U.U
A
Scenario
¦
2020/A
¦
2020/B
2030/A
2030/B
to
LU
5
<
Figure D-2. Comparing PM2.5 Relative Reduction Potential of the OGV Sector
15%
10%
O%
I 15%
U 1(5%
£
5%
0%
MM
* ±
Scenario
I 2020/A
I 2020/B
| 2030/A
I 2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
117
-------
Appendix D: Stratified Summary of Results
Figure D-3. Comparing HC Relative Reduction Potential of OGV Sector
7.5ft
2.5ft
7.5ft
2.5 ft
£
7.5%
5.0%
2.5%
0.(3%
-2.5%
Scenario
n
I 2020/A
D
OJ
I 2020/B
=J
III
| 2030/A
1 2030/B
ft
LU
<
01
t£l
Figure D-4. Comparing BC Relative Reduction Potential of OGV Sector
Scenario
n
1 202D/A
o
OJ
| 2020/B
d
m
1 2030/A
| 2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
118
-------
Appendix D: Stratified Summary of Results
Figure D-5. Comparing S02 Relative Reduction Potential of OGV Sector
«- -C;:
a 30*
£
AO%
3056
20%
10%
o%
4%
2%
0*
.2
11
£ 256
c
.a 054
t3
3
¦o
St
4%
2%
0%
~
LU
<
Figure D-6. Comparing C02 Relative Reduction Potential of OGV Sector
-
¦- I ¦_ I I
ill ^Jl
Scenario
I 2020/A
I 202D/B
| 2030/A
I 203-3,;B
Scenario
| 2020/A
| 2020/B
| 2030/A
I 2030/B
2050/A
I 2050/B
5
<
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
119
-------
Appendix D: Stratified Summary of Results
D.2.2. Container Ports
Figure D-7. NOx Relative Reduction Potential of the OGV Sector for Container Ports
tj 6%
Scenario
| 2020/A
I 2020/B
H 2030/A
I 2030/B
i D-8. PM2.5 Relative Reduction Potential of the OGV Sector for Container Ports
Figure
Scenario
| 2020/A
| 2020/B
l| 2030/A
H 2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
120
-------
Appendix D: Stratified Summary of Results
Figure D-9. HC Relative Reduction Potential of OGV Sector for Container Ports
7.5%
2.S-/Z
SL, 0.0%
Scenario
2020/A
ti -2.5%
2020/B
2030/A
2030/B
5.C :
J. 5 - :
0.0%
-2.5%
Figure D-10. BC Relative Reduction Potential of OGV Sector for Container Ports
Scenario
I 2020/A
I 2020/B
| 2030/A
¦ 2030/B.
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
121
-------
Appendix D: Stratified Summary of Results
Figure D-ll. S02 Relative Reduction Potential of OGV Sector for Container Ports
AQ%
30%
20%
ig.10%
H- 0%
m
£
40%
£
30%
20%
lost
0%
dJ
0
LU
<
01
4%
2%
:o%
.3
£
c
.S
I
iTuTiil
- 4%
£
2%
0%
ill -ul iil J
Scenario
I 2020/A
I 2020/B
| 2030/A
I 2030/B
Figure D-12. C02 Relative Reduction Potential of OGV Sector for Container Ports
Scenario
I 2020/A
I 2020/B
| 2030/A
J 2030/B
I 2050/A
I 2050/B
a
LU
<
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
122
-------
Appendix D: Stratified Summary of Results
D.2.3. Bulk Ports
Figure D-13. NOx Relative Reduction Potential of OGV Sector for Bulk Ports
Scenario
2020/A
2020/B
2030,'A
2030/B
Figure D-14. PM Relative Reduction Potential of OGV Sector for Bulk Ports
15%
10%
.a
e
o
£
e
.2
z
a
-o
&
Scenario
2020/A
2020: B
1556
10M
056
2030/A
2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
123
-------
Appendix D: Stratified Summary of Results
Figure D-15. HC Relative Reduction Potential of OGV Sector for Bulk Ports
Scenario
I 2020/A
I 2020/B
| 2030/A
I 203-3,;B
Figure D-16. BC Relative Reduction Potential of OGV Sector for Bulk Ports
Scenario
I 2020/A
I 2020/B
| 2030/A
B 2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
124
-------
Appendix D: Stratified Summary of Results
Figure D-17. S02 Relative Reduction Potential of OGV Sector for Bulk Ports
40%
30%
20%
¦J3
5^10% ¦ Scenario
.a
o% ^^^™ H 2020/A
£ —
£ 2020/B
I 2030/A
= 40%
¦s
£
30%
20%
10%
0%
2030/B
Figure D-18. C02 Relative Reduction Potential of OGV Sector for Bulk Ports
Scenario
2020/A
V*
2%
0%
-III Ul ill
2030/A
2030/B
20BO/A
2050/B
0 ?
lu
^ aj
Ul
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
125
-------
Appendix D: Stratified Summary of Results
D.2.4. Passenger Ports
Figure D-19. NOx Relative Reduction Potential of OGV Sector for Passenger Ports
8%
6%
4%
II ..
Scenario
I 2020/A
| 2020/B
| 2030/A
II 2030/B
Figure D-20. PM Relative Reduction Potential of OGV Sector for Passenger Ports
Scenario
H2020/A
| 2020/B
l| 2030/A
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
126
-------
Appendix D: Stratified Summary of Results
Figure D-21. HC Relative Reduction Potential of OGV Sector for Passenger Ports
7.5%
5.0%
,— 2.5%
§
.a o.o%
e
CJ
£
i 7.5%
£
5.0%
2.556
o.o*;
J
Scenario
| 2020/A
| 2020/B
| 2030/A
I 2030/B
Figure D-22. BC Relative Reduction Potential of OGV Sector for Passenger Ports
a^2«
.a
Scenario
I 2020/A
I 2020/B
| 2030/A
¦ 2030/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
127
-------
Appendix D: Stratified Summary of Results
Figure D-23. S02 Relative Reduction Potential of OGV Sector for Passenger Ports
30*
205€
glO*
¦J OS6
J J
Scenario
I 2020/A
£ ^112<32°/B
I 2030/A
C40«
s
| 30%
20S6
10S6
o%
J A
2030/B
a 5
LU 3
>
^ OJ
bfl
4%
3%
2%
1%
sS
3 OS
Figure D-24. C02 Relative Reduction Potential of OGV Sector for Passenger Ports
idl -ul Hkjl
Scenario
I 2020/A
I 2020/B
| 2030/A
| 2030/B
I 2050/A
I 2050/B
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
128
-------
Appendix D: Stratified Summary of Results
D.3. Additional Charts for Stratification Analysis of Non-OGV Sector
70%
60%
re
'¦§ 50%
o
2 40%
¦2 30%
20%
Figure D-25. NOx Relative Reduction Potential of Non-OGV Sector
ttL
10%
0% ¦ 1
ir i
i ii
r i
Year
| 2020/A
12020®
| 2030/A
I 2030/B
C=
CD
X
o
LU
X
o
CD
c
CD
o
O
I—
cr
LU
X
o
>-
X
o
CD
CD
CD
O
c
CD
O
X
o
X
CD
cn
Figure D-26. PM2.5 Relative Reduction Potential of Non-OGV Sector
70%
0% -
¦ III
¦ II
P P"
2020/A
12020/B
| 2030/A
I 2030/B
a
CD
X
o
LU
X
o
CD
a
CD
o
O
I—
cr
LJU
X
o
X
o
CD
CT.
CD
o
c
CD
O
O
X
\—
CD
cr
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
129
-------
Appendix D: Stratified Summary of Results
Figure D-27. Comparing BC Relative Reduction Potential of Non-OGV Sectors
70%
60%
50%
40%
~ 30%
o
"H 20%
10% H
0% -¦
I I
¦ III
¦ II
f r~
Year
I 2020/A
I 2020/B
| 2030/A
I 2030/Q
TD
C
O
X
o
p
o
I—
tr
x
o
>-
LU
X
o
Q
-------
Appendix D: Stratified Summary of Results
Figure D-29. Comparing VOC Relative Reduction Potential of Non-OGV Sectors
60%
50%
£
¦*->
o
CL
C
o
o
3
O
QL
40%
30%
20%
ml
I ¦
__H I Year
Juil I
10% -I
0% -¦
11 ¦ ¦
r r
p in
i r-
r
T=l
C
o
X
o
p
o
I—
tr
i
o
>-
LU
X
o
-
X
o
CD
o
o
c:
CO
:=-
O
o
X
I-
CD
CU
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
131
-------
Appendix D: Stratified Summary of Results
Figure D-31. Comparing Benzene Relative Reduction Potential of Non-OGV Sectors
50%
— 40%
re
30%
c
4-«
o
Q.
C
¦g 20%
¦c
0
01 10%
i. ,l
IJ I
0% -¦
r r
ini
i
r
Year
I 2020/A
12020/B
| 2030/A
I 2030/B
TD
C
O
X
o
p
o
I—
tr
x
o
>-
LU
X
o
Q
-
X
o
CD
o
o
c:
CO
:=-
O
o
X
f-
CD
£3C
National Port Strategy Assessment: Reducing Air Pollution and Greenhouse Gases at U.S. Ports
132
------- |