Proposal to Designate an Emission
Control Area for Nitrogen Oxides,
Sulfur Oxides and Particulate Matter
Technical Support Document
United States
Environmental Protection
Agency
-------
Proposal to Designate an Emission
Control Area for Nitrogen Oxides,
Sulfur Oxides and Particulate Matter
Technical Support Document
Assessment and Standards Division
Office of Transportation and Air Quality
U.S. Environmental Protection Agency
v>EPA
United States EPA-420-R-09-007
Environmental Protection . ., „„_
Agency April 2009
-------
Table of Contents
1 Executive Summary 1-3
2 Emission Inventory 2-1
2.1 Introduction 2-1
2.2 Modeling Domain and Geographic Regions 2-2
2.3 Development of 2002 Inventories 2-3
2.4 Development of 2020 Inventories 2-22
2.5 Projected Emission Reductions 2-54
2.6 Conclusion 2-54
Appendices 2-56
3 Impacts of Shipping Emissions on Air Quality, Health and the
Environment 3-3
3.1 Pollutants Reduced by the EGA and their Associated Health Impacts 3-3
3.2 Current and Proj ected Air Quality 3-18
3.3 Impacts on Ecosystems 3-39
Appendices 3-93
4 Quantified Health Impacts Analysis 4-1
4.1 Health Impacts Analysis Results for the Proposed EGA 4-4
4.2 Methodology 4-6
4.3 Methods for Describing Uncertainty 4-27
5 Costs 5-1
5.1 Fuel Production Costs 5-2
5.2 Engine and Vessel Costs 5-14
5.3 Total Estimated ECA Costs in 2020 5-33
5.4 Cost Effectiveness 5-34
-------
Economic Impacts 6-1
6.1 The Purpose of an Economic Impact Analysis 6-2
6.2 Economic Impact Analysis Methodology 6-2
6.3 Expected Economic Impacts of the Proposed EGA 6-4
Appendices 6-10
-------
1 Executive Summary
Introduction
On March 27, 2009, the United States and Canada submitted a joint proposal (MEPC
59/6/5) to the International Maritime Organization to designate an Emission Control Area (EGA)
for specific portions of U.S. and Canadian coastal waters. This action would control emissions
of nitrogen oxides (NOx), sulfur oxides (SOx), and particulate matter (PM) from ships.
Designation of the proposed EGA is necessary to protect public health and the environment in
the United States and Canada by reducing exposure to harmful levels of air pollution resulting
from these emissions. The burden on international shipping is small compared to the
improvements in air quality, reductions in premature mortality and other benefits resulting from
designation of the proposed EGA.
This Technical Support Document provides a comprehensive presentation of the many
in-depth technical analyses performed by the U.S. Government, in developing the EGA proposal.
Emission Inventory
Chapter 2 describes how U.S. emission inventories were developed to describe air
emissions from ships operating in waters within the proposed EGA. These inventories provide
the foundation upon which all the subsequent analyses were built, and address Criterion 6 of
Section 3, Appendix III to MARPOL Annex VI. Beyond the level of detail provided in MEPC
59/6/5, Chapter 2 explains how the inputs were developed and what assumptions were made in
assessing what the emissions are from ships currently (2002 base year), what the emissions
would look like in 2020 without the proposed EGA, and what reductions can be expected from
the proposed EGA.
Chapter 2 describes the "bottom-up" methodology that was used, based on the latest state
of the art models and inputs. This chapter describes which port-related emissions were included
and why, and how emissions were obtained for ships while underway in U.S. waters. This
chapter explains in great detail each parameter that went into the modeling and analyses,
including which ships are included, which fuels are used by those ships, which other (non-ECA)
emission controls are in place for each scenario, and what growth rates are expected,
incorporating forecasts of the demand for marine transportation services in 2020.
Impacts of Emissions on Air Quality, Human Health and the Environment
Chapter 3 describes in great detail most of the analyses conducted in support of Criteria
2, 3, 4 and 5 of Section 3, Appendix III to MARPOL Annex VI. For organizational reasons, the
analyses conducted to assess the impacts of ships' emissions on human health are presented in
Chapter 4, summarized below. Chapter 3 contains several sub-sections, outlined here for ease of
reference.
1-1
-------
Impacts of Pollutants on Human Health
Section 3.1 describes the human health impacts of the pollutants proposed for control in
the U.S./Canada EGA. The proposed EGA would not only reduce direct emissions of NOx, SOx
and PM, but also secondarily formed ambient PM and ground-level ozone. Section 3.1.1
describes the nature of these pollutants, formation processes, and relationship to ship emissions.
Section 3.1.2 presents the health effects associated with exposure to NOX, SOX, PM and ground-
level ozone, summarizing the key scientific literature.
Impacts of Ships' Emissions on Air Quality and Benefits of EGA to Air Quality
Section 3.2 describes the effects of NOx, SOx and PM emissions on ambient air quality
under the same scenarios for which emission inventories were developed, presented in terms of
ground-level ozone and PM. This section also describes the multi-pollutant modeling platform
that was used to assess the impacts of reduced marine emissions from the application of the
proposed EGA. Appendix A to Chapter 3 describes the relevant meteorological conditions that
contribute to at-sea emissions being transported to populated areas and contributing to harmful
human health and ecological impacts, and which formed inputs to the modeling platform.
Impacts of Ships' Emissions on Ecosystems and Benefits of EGA to Ecosystems
Section 3.3 describes the impacts of emissions from ships on terrestrial and aquatic
ecosystems such as visibility, ozone uptake, eutrophication, acidification, loss of forest biomass,
and overall forest health. Using the same scenarios as for the other analyses, improvements in
environmental conditions for many types of ecosystems were evaluated. Unlike the analyses for
human health, there are a larger number of pollutants of concern to ecosystems. Thus, deposition
of many chemical forms of NOx, SOx and PM are discussed in this section, as well as the
biogeochemical cycles of interrelated pollutants such as mercury.
Impacts of Ships' Emissions on Human Health and Benefits of ECA to Human Health
Chapter 4 presents quantified U.S.-related health impacts for PM and ozone associated
with emissions from ships, both in terms of the expected contribution of overall ship emissions to
adverse health impacts on land and the reductions in adverse health impacts that can be expected
to occur from the adoption of the proposed ECA.
The health impacts modeling presented in Chapter 4 is based on peer-reviewed studies of
air quality and health and welfare effects associated with improvements in air quality. This
chapter also describes the computer program used to estimate health benefits by integrating a
number of modeling elements (e.g., interpolation functions, population projections, health impact
functions, valuation functions, analysis and pooling methods) to translate modeled air
concentration estimates into health effect incidence estimates.
Cost Analyses
Chapter 5 describes our estimates of the costs associated with the reduction of SOx, NOx,
and PM emissions from ships, not only to the shipping industry but also to marine fuel suppliers
and companies who rely on the shipping industry. This chapter provides additional detail
1-2
-------
regarding the analyses conducted in support of Criteria 7 and 8 of Section 3, Appendix III to
MARPOL Annex VI. This chapter describes the analyses used to evaluate the cost impact of
Tier III NOx requirements combined with low sulfur fuel use on vessels operating within the
proposed EGA, including estimates of low sulfur fuel production costs, vessel hardware costs,
and operating costs. This chapter also presents cost per ton estimates for EGA-based NOx and
fuel sulfur standards and compares these with the costs of established land-based control
programs.
Economic Impact Analysis
Chapter 6 examines the economic impacts of the projected EGA costs on shipping
engaged in international trade. This chapter provides additional detail in support of Criterion 8
of Section 3, Appendix III to MARPOL Annex VI. This chapter describes the econometric
methodology that was used in estimating two aspects of the economic impacts: social costs and
how they are shared across stakeholders, and market impacts for the new engine and new vessel
markets.
1-3
-------
2 Emission Inventory
2.1 Introduction
Ships (i.e., ocean-going vessels) are significant contributors to the total United States (U.S.)
mobile source emission inventory. The U.S. ship inventory reported here focuses on Category 3
(C3) vessels, which use C3 engines for propulsion. C3 engines are defined as having displacement
above 30 liters per cylinder (L/cyl). The resulting inventory includes emissions from both
propulsion and auxiliary engines used on these vessels, as well as those on gas and steam turbine
vessels.
Most of the vessels operating in U.S. ports that have propulsion engines less than 30 liters
per cylinder are domestic and are already subject to strict national standards affecting NOx, PM,
and fuel sulfur content. As such, the inventory does not include any ships, foreign or domestic,
powered by Category 1 or Category 2 (i.e., <30 L/cyl) engines. In addition, as discussed in Sections
2.3.2.3.9 and 2.3.3.3, this inventory is primarily based on activity data for ships that carry foreign
cargo. Category 3 vessels carrying domestic cargo that operate only between U.S. ports are only
partially accounted for in this inventory.1 Emissions due to military vessels are also excluded.
The regional and national inventories for C3 vessels presented in this chapter are sums of
independently constructed port and interport emissions inventories. Port inventories were
developed for 89 deep water and 28 Great Lake ports in the U.S.2 While there are more than 117
ports in the U.S., these are the top U.S. ports in terms of cargo tonnage. Port-specific emissions
were calculated with a "bottom-up" approach, using data for vessel calls, emission factors, and
activity for each port. Interport emissions and emissions for the remaining ports were obtained
using the Waterway Network Ship Traffic, Energy and Environment Model (STEEM).3'4 STEEM
also uses a "bottom-up" approach, estimating emissions from C3 vessels using historical North
American shipping activity, ship characteristics, and activity-based emission factors. STEEM was
used to quantify and geographically (i.e., spatially) represent interport vessel traffic and emissions
for vessels traveling generally within 200 nautical miles (nm) of the U.S.
The detailed port inventories were spatially merged into the STEEM gridded inventory to
create a comprehensive inventory for Category 3 vessels. For the 117 ports, this involved removing
the near-port portion of the STEEM inventory and replacing it with the detailed port inventories.
For the remaining U.S. ports for which detailed port inventories are not available, the near-port
portion of the STEEM inventory was simply retained. This was done for a base year of 2002.
Inventories for 2020 were then projected using regional growth rates5'6 and adjustment factors to
account for the International Maritime Organization (IMO) Tier I and Tier II NOx standards and
NOx retrofit program.2 Inventories incorporating additional Tier III NOx and fuel sulfur controls
within the proposed Emission Control Area (EGA) were also developed for 2020.
The inventory estimates reported in this chapter include emissions out to 200 nm from the
U.S. coastline, including Alaska and Hawaii, but not extending into the Exclusive Economic Zone
(EEZ) of neighboring countries. Inventories are presented for the following pollutants: oxides of
nitrogen (NOx), particulate matter (PM2.s and PMio), sulfur dioxide (SO2), hydrocarbons (HC),
carbon monoxide (CO), and carbon dioxide (COz). The PM inventories include directly emitted
PM only, although secondary sulfates and nitrates are taken into account in the air quality modeling.
2-1
-------
2.2 Modeling Domain and Geographic Regions
The inventories described in this chapter reflect ship operations that occur within the area that
extends 200 nautical miles (nm) from the official U.S. baseline but exclude operations in Exclusive
Economic Zones of other countries. The official U.S. baseline is recognized as the low-water line
along the coast as marked on the official U.S. nautical charts in accordance with the articles of the
Law of the Sea. The boundary was mapped using geographic information system (CIS) shapefiles
obtained from the National Oceanic and Atmospheric Administration, Office of Coast Survey.7
The accuracy of the NOAA shapefiles was verified with images obtained from the U.S. Geological
Survey. The confirmed NOAA shapefiles were then combined with a shapefile of the U.S.
international border from the National Atlas.8
The resulting region was further subdivided for this analysis to create regions that were
compatible with the geographic scope of the regional growth rates, which are used to project
emission inventories for the years 2020, as described later in this document.
• The Pacific Coast region was split into separate North Pacific and South Pacific regions
along a horizontal line originating from the Washington/Oregon border (Latitude 46° 15'
North).
• The East Coast and Gulf of Mexico regions were divided along a vertical line roughly drawn
through Key Largo (Longitude 80° 26' West).
• The Alaska region was divided into separate Alaska Southeast and Alaska West regions
along a straight line intersecting the cities of Naknek and Kodiak. The Alaska Southeast
region includes most of the State's population, and the Alaska West region includes the
emissions from ships on a great circle route along the Aleutian Islands between Asia and the
U.S. West Coast.
• For the Great Lakes domain, shapefiles were created containing all the ports and inland
waterways in the near port inventory and extending out into the lakes to the international
border with Canada. The modeling domain spanned from Lake Superior on the west to the
point eastward in the State of New York where the St. Lawrence River parts from U.S. soil.
• The Hawaiian domain was subdivided so that a distance of 200 nm beyond the southeastern
islands of Hawai'i, Maui, O'ahu, Moloka'i, Ni'ihau, Kaua'i, Lanai, and Kahoolawe was
contained in Hawaii East. The remainder of the Hawaiian Region was then designated
Hawaii West.
This methodology resulted in nine separate regional modeling domains that are identified
below and shown in Figure 2-1. U.S. territories are not included in this analysis.
• South Pacific (SP)
• North Pacific (NP)
• East Coast (EC)
• Gulf Coast (GC)
• Alaska Southeast (AE)
• Alaska West (AW)
• Hawaii East (HE)
2-2
-------
• Hawaii West (HW)
• Great Lakes (GL)
Figure 2-1 Regional Modeling Domains
2.3 Development of 2002 Inventories
This section describes the methodology and inputs, and presents the resulting inventories for
the 2002 baseline calendar year. The first section describes the general methodology. The second
section describes the methodology, inputs, and results for near port emissions. The third section
describes the methodology and inputs for emissions when operating away from port (also referred
to as "interport" emissions). The fourth section describes the method for merging the interport and
near port portions of the inventory. Resulting total emissions for the U.S., as well as for nine
geographic regions within the U.S., are then presented.
2.3.1 Outline of Methodology
The total inventory was created by summing emissions estimates for ships while at port
(near port inventories) and while underway (interport inventories). Near port inventories for
calendar year 2002 were developed for 117 U.S. commercial ports that engage in foreign trade.
Based on an analysis of U.S. Government data, these 117 commercial ports encompass nearly all
U.S. C3 vessel calls.9
The outer boundaries of the ports are defined as 25 nm from the terminus of the reduced
speed zone for deep water ports and 7 nm from the terminus of the reduced speed zone for Great
Lake ports. Port emissions are calculated for different modes of operation and then summed.
Emissions for each mode are calculated using port-specific information for vessel calls, vessel
characteristics, and activity, as well as other inputs that vary instead by vessel or engine type (e.g.,
emission factors).
2-3
-------
The interport inventory is estimated using the Waterway Network Ship Traffic, Energy, and
Environmental Model (STEEM).3'4 The model geographically characterizes emissions from ships
traveling along shipping lanes to and from individual ports, in addition to the emissions from
vessels transiting near the ports. The shipping lanes were identified from actual ship positioning
reports. The model then uses detailed information about ship destinations, ship attributes (e.g.,
vessel speed and engine horsepower), and emission factors to produce spatially allocated (i.e.,
gridded) emission estimates for ships engaged in foreign commerce.
The 117 near port inventories are an improvement upon STEEM's near port results in
several ways. First, the precision associated with STEEM's use of ship positioning data may be less
accurate in some locations, especially as the lanes approach shorelines where ships would need to
follow more prescribed paths. Second, the STEEM model includes a maneuvering operational
mode (i.e., reduced speed) that is generally assumed to occur for the first and last 20 kilometers of
each trip when a ship is leaving or entering a port. In reality, the distance when a ship is traveling at
reduced speeds varies by port. Also, the distance a ship traverses at reduced speeds often consists
of two operational modes: a reduced speed zone (RSZ) as a ship enters or leaves the port area and
actual maneuvering at a very low speed near the dock. Third, the STEEM model assumes that the
maneuvering distance occurs at an engine load of 20 percent, which represents a vessel speed of
approximately 60 percent of cruise speed. This is considerably faster than ships would maneuver
near the docks. The single maneuvering speed assumed by STEEM also does not reflect the fact
that the reduced speed zone, and therefore emissions, may vary by port. Fourth, and finally, the
STEEM model does not include the emissions from auxiliary engines during hotelling operations at
the port. The near-port inventories correct these issues.
The regional emission inventories produced by the current STEEM interport model are most
accurate for vessels while cruising in ocean or Great Lakes shipping lanes; the near port inventories
use more detailed local port information and are significantly more accurate near the ports.
Therefore, the inventories in this analysis are derived by merging together: 1) the near port
inventories, which extend 25 nautical miles and 7 nautical miles from the terminus of the RSZ for
deep water ports and Great Lake ports, respectively, and 2) the remaining interport portion of the
STEEM inventory, which extends from the endpoint of the near port inventories to the 200 nautical
mile boundary or international border with Canada, as appropriate. Near some ports, a portion of
the underlying STEEM emissions were retained if it was determined that the STEEM emissions
included ships traversing the area near a port, but not actually entering or exiting the port.
2.3.2 Near Port Emissions
Near port inventories for calendar year 2002 were developed for ocean-going vessels at 89
deep water and 28 Great Lake ports in the U.S. The inventories include emissions from both
propulsion and auxiliary engines on C3 vessels.
This section first describes the selection of the ports for analysis and then provides the
methodology used to develop the near port inventories. This is followed by a description of the key
inputs. Total emissions by port and pollutant for 2002 are then presented.
2-4
-------
2.3.2.1 Selection of Individual Ports to be Analyzed
All 150 deep water and Great Lake ports in the Principal Ports of the United States dataset10
were used as a starting point. Thirty ports which had no foreign traffic were eliminated because the
dataset used to obtain port calls and other ship characteristics has no information about domestic
traffic. Several California ports were also used because the California Air Resources Board (ARE)
provided the necessary data and estimates for those ports. The final list of 117 deep water and
Great Lake ports, along with their coordinates, is given in Appendix 2A.
2.3.2.2 Port Methodology
Near port emissions for each port are calculated for four modes of operation: 1) hotelling, 2)
maneuvering, 3) reduced speed zone (RSZ), and 4) cruise. Hotelling, or dwelling, occurs while the
vessel is docked or anchored near a dock, and only the auxiliary engine (s) are being used to provide
power to meet the ship's energy needs. Maneuvering occurs within a very short distance of the
docks. The RSZ varies from port to port, though generally the RSZ would begin and end when the
pilots board or disembark, and typically occurs when the near port shipping lanes reach
unconstrained ocean shipping lanes. The cruise mode emissions in the near ports analysis extend 25
nautical miles beyond the end of the RSZ lanes for deep water ports and 7 nautical miles for Great
Lake ports.
Emissions are calculated separately for propulsion and auxiliary engines. The basic
equation used is as follows:
Equation 2-1
Emission&oAe[eng] = (call§ x (P[eng]) x (hrs/callmoj x (LFmode[eng]) x (EF[eng]) x (Adj) x (1(T6 tannedg)
Where:
- Emissionsmode [eng] = Metric tonnes emitted by mode and engine type
Calls = Round-trip visits (i.e., one entrance and one clearance is considered a call)
P[eng] = Total engine power by engine type, in kilowatts
- hrs/callmode = Hours per call by mode
- LFmode [eng] = Load factor by mode and engine type (unitless)
EF[eng] = Emission factor by engine type for the pollutant of interest, in g/kW-hr
(these vary as a function of engine type and fuel used, rather than activity mode)
- Adj = Low load adjustment factor, unitless (used when the load factor is below 0.20)
106 = Conversion factor from grams to metric tonnes
Main engine load factors are calculated directly from the propeller curve based upon the
cube of actual speed divided by maximum speed (at 100% maximum continuous rating [MCR]). In
addition, cruise mode activity is based on cruise distance and speed inputs. Appendix 2B provides
the specific equations used to calculate propulsion and auxiliary emissions for each activity mode.
2.3.2.3 Inputs for Port Emission Calculations
The following inputs are required to calculate emissions for the four modes of operation
(cruise, RSZ, maneuvering, and hotelling):
2-5
-------
• Number of calls
• Main engine power
• Cruise (vessel service) speed
• Cruise distance
• RSZ distance for each port
• RSZ speed for each port
• Auxiliary engine power
• Auxiliary load factors
• Main and auxiliary emission factors
• Low load adjustment factors for main engines
• Maneuvering time-in-mode (hours/call)
• Hotelling time-in-mode (hours/call)
Note that load factors for main engines are not listed explicitly, since they are calculated as a
function of mode and/or cruise speed. This section describes the inputs in more detail, as well as
the sources for each input.
2.3.2.3.1 Calls and Ship Characteristics (Propulsion Engine Power and Cruise Speed)
For this analysis, U.S. Army Corps of Engineers (USAGE) entrance and clearance data for
2002,u together with Lloyd's data for ship characteristics,12 were used to calculate average ship
characteristics and calls by ship type for each port. Information for number of calls, propulsion
engine power, and cruise speed were obtained from these data.
2.3.2.3.1.1 Bins by Ship Type, Engine Type, and DWT Range
The records from the USAGE entrances and clearances data base were matched with
Lloyd's data on ship characteristics for each port. Calls by vessels that have either Category 1 or 2
propulsion engines were eliminated from the data set. The data was then binned by ship type,
engine type and dead weight tonnage (DWT) range. The number of entrances and clearances in
each bin are counted, summed together and divided by two to determine the number of calls (i.e.,
one entrance and one clearance was considered a call). For Great Lake ports, there is a larger
frequency of ships either entering the port loaded and leaving unloaded (light) or entering the port
light and leaving loaded. In these cases, there would only be one record (the loaded trip into or out
of the port) that would be present in the data. For Great Lake ports, clearances were matched with
entrances by ship name. If there was not a reasonable match, the orphan entrance or clearance was
treated as a call.
Propulsion power and vessel cruise speed are also averaged for each bin. Auxiliary engine
power was computed from the average propulsion power using the auxiliary power to propulsion
power ratios discussed in section 2.3.2.3.4.
2.3.2.3.1.2 Removal of Category 1 and 2 Ships
Since these inventories were intended to cover Category 3 propulsion engine ships only, the
ships with Category 1 and 2 propulsion engines were eliminated. This was accomplished by
matching all ship calls with information from Lloyd's Data, which is produced by Lloyd's Register-
Fairplay Ltd.12 Over 99.9 percent of the calls in the entrances and clearances data were directly
2-6
-------
matched with Lloyd's data. The remaining 0.1 percent was estimated based upon ships of similar
type and size.
Engine category was determined from engine make and model. Engine bore and stroke
were found in the Marine Engine 2005 Guide13 and displacement per cylinder was calculated.
Ships with Category 1 or 2 propulsion engines were eliminated from the data.
Many passenger ships and tankers have either diesel-electric or gas turbine-electric engines
that are used for both propulsion and auxiliary purposes. Both were included in the current
inventory.
2.3.2.3.2 Cruise Distance
Cruise mode emissions are calculated assuming a 25 nautical mile distance into and out of
the port for deep water ports and 7 nautical miles into and out of the port for Great Lake ports
outside of the reduced speed and maneuvering zones.
2.3.2.3.3 RSZ Distances and Speeds by Port
Reduced speed zone (RSZ) distance and speed were individually determined for each port.
For deep water ports, the RSZ distances were developed from shipping lane information contained
in the USAGE National Waterway Network.14 The database defines waterways as links or line
segments that, for the purposes of this study, represent actual shipping lanes (i.e., channels,
intracoastal waterways, sea lanes, and rivers). The sea-side endpoint for the RSZ was selected as
the point along the line segment that was judged to be far enough into the ocean where ship
movements were unconstrained by the coastline or other vessel traffic. The resulting RSZ distance
was then measured for each deep water port. The final RSZ distances and endpoints for each port
are listed in Appendix 2C. The RSZ for each Great Lake port was fixed at three nautical miles.
2.3.2.3.4 Auxiliary Engine Power and Load Factors
Since hotelling emissions are a large part of port inventories, it is important to distinguish
propulsion engine emissions from auxiliary engine emissions. In the methodology used in this
analysis, auxiliary engine maximum continuous rating power and load factors were calculated
separately from propulsion engines and different emission factors (EFs) applied. All auxiliary
engines were treated as Category 2 medium-speed diesel (MSB) engines for purposes of this
analysis.
Auxiliary engine power is not contained in the USAGE database and is only sparsely
populated in the Lloyd's database; as a result, it must be estimated. The approach taken was to
derive ratios of average auxiliary engine power to propulsion power based on survey data. The
California Air Resources Board (ARE) conducted an Oceangoing Ship Survey of 327 ships in
January 2005 that was principally used for this analysis.15 Average auxiliary engine power to
propulsion power ratios were estimated by ship type and are presented in Table 2-1. These ratios by
ship type were applied to the propulsion power data to derive auxiliary power for the ship types at
each port.
2-7
-------
Table 2-1 Auxiliary Engine Power Ratios (ARE Survey, except as noted)
Ship Type
Auto Carrier
Bulk Carrier
Container Ship
Passenger Shipa
General Cargo
Miscellaneous15
RORO
Reefer
Tanker
Average
Propulsion
Engine (kW)
10,700
8,000
30,900
39,600
9,300
6,250
11,000
9,600
9,400
Average Auxiliary Engines
Number
2.9
2.9
3.6
4.7
2.9
2.9
2.9
4.0
2.7
Power
Each
(kW)
983
612
1,889
2,340
612
580
983
975
735
Total
Power
(kW)
2,850
1,776
6,800
11,000
1,776
1,680
2,850
3,900
1,985
Engine Speed
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Auxiliary to
Propulsion
Ratio
0.266
0.222
0.220
0.278
0.191
0.269
0.259
0.406
0.211
a Many 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.
Miscellaneous ship types were not provided in the ARE methodology, so values from the Starcrest Vessel
Boarding Program were used.
Auxiliary engine to propulsion engine power ratios vary by ship type and operating mode
roughly from 0.19 to 0.40. Auxiliary load, shown in Table 2-2, is used together with the total
auxiliary engine power to calculate auxiliary engine emissions. Starcrest's Vessel Boarding
Program16 showed that auxiliary engines are on all of the time, except when using shoreside power
during hotelling.
Table 2-2 Auxiliary Engine Load Factor Assumptions
Ship-Type
Auto Carrier
Bulk Carrier
Container Ship
Passenger Ship
General Cargo
Miscellaneous
RORO
Reefer
Tanker
Cruise
0.13
0.17
0.13
0.80
0.17
0.17
0.15
0.20
0.13
RSZ
0.30
0.27
0.25
0.80
0.27
0.27
0.30
0.34
0.27
Maneuver
0.67
0.45
0.50
0.80
0.45
0.45
0.45
0.67
0.45
Hotel
0.24
0.22
0.17
0.64
0.22
0.22
0.30
0.34
0.67
2.3.2.3.5 Fuel Types and Fuel Sulfur Levels
There are primarily three types of fuel used by marine engines: residual marine (RM),
marine diesel oil (MDO), and marine gas oil (MGO), with varying levels of fuel sulfur.5 MDO and
MGO are generally described as distillate fuels. For this analysis, RM and MDO fuels are assumed
2-8
-------
to be used. Since PM and S02 emission factors are dependent on the fuel sulfur level, calculation of
port inventories requires information about the fuel sulfur levels associated with each fuel type, as
well as which fuel types are used by propulsion and auxiliary engines.
Based on an ARE survey,15 average fuel sulfur level for residual marine was set to 2.5
percent for the west coast and 2.7 percent for the rest of the country. A sulfur content of 1.5 percent
was used for MDO.17 While a more realistic value for MDO used in the U.S. appears to be 0.4
percent, given the small proportion of distillate fuel used by ships relative to RM, the difference
should not be significant. Sulfur levels in other areas of the world can be significantly higher for
RM. Table 2-3, based on the ARE survey, provides the assumed mix of fuel types used for
propulsion and auxiliary engines by ship type.
Table 2-3 Estimated Mix of Fuel Types Used by Ships
Ship Type
Passenger
Other
Fuel Used
Propulsion
100% RM
100% RM
Auxiliary
92% RM/8% MDO
71%RM/29%MDO
2.3.2.3.6 Propulsion and A uxiliary Engine Emission Factors
An analysis of emission data was prepared and published in 2002 by Entec.17 The resulting
Entec emission factors include individual factors for three speeds of diesel engines (slow-speed
diesel (SSD), medium-speed diesel (MSB), and high-speed diesel (HSD)), steam turbines (ST), gas
turbines (GT), and two types of fuel used here (RM and MDO). Table 2-4 lists the propulsion
engine emission factors for NOx and HC that were used for the 2002 port inventory development.
The CO, PM, SOz and COz emission factors shown in the table come from other data sources as
explained below.
Table 2-4 Emission Factors for OGV Main Engines using RM, g/kWh
Engine
SSD
MSD
ST
GT
All Ports
NOX
18.1
14.0
2.1
6.1
CO
1.40
1.10
0.20
0.20
HC
0.60
0.50
0.10
0.10
CO2
620.62
668.36
970.71
970.71
West Coast Ports
PM10
1.4
1.4
1.4
1.4
PM25
1.3
1.3
1.3
1.3
SO2
9.53
10.26
14.91
14.91
Other Ports
PM10
1.4
1.4
1.5
1.5
PM25
1.3
1.3
1.4
1.4
SO2
10.29
11.09
16.10
16.10
CO emission factors were developed from information provided in the Entec appendices
because they are not explicitly stated in the text. HC and CO emission factors were confirmed with
a recent U.S. Government review.18
2-9
-------
PMioA values were determined based on existing engine test data in consultation with
ARE. 19 GT PMio emission factors were not part of the U.S. Government analysis but assumed here
to be equivalent to ST PMio emission factors. Test data shows PMio emission rates as dependent
upon fuel sulfur levels, with base PMio emission rates of 0.23 g/kw-hr with distillate fuel (0.24%
sulfur) and 1.35 g/kw-hr with residual fuel (2.46% sulfur).20 The equation used to generate
emission factors based on sulfur content is shown below. PM2.s is assumed to be 92 percent of
PMio. While the US Government NONROAD model uses 0.97 for such conversion based upon low
sulfur fuels, a reasonable value seems to be closer to 0.92 because higher sulfur fuels in medium
and slow speed engines would tend to produce larger particulates than high speed engines on low
sulfur fuels.
Equation 2-2 Calculation of PM10 Emission Factors Based on Fuel Sulfur Levels
PMEF = PMNom + [(SAct - SNom) x BSFC x FSC x MWR x 0.0001]
where:
PMEF = PM emission factor adjusted for fuel sulfur
PMNom = PM emission rate at nominal fuel sulfur level
= 0.23 g/kW-hr for distillate fuel, 1.35 g/kW-hr for residual fuel
SAct = Actual fuel sulfur level (weight percent)
SNom = nominal fuel sulfur level (weight percent)
= 0.24 for distillate fuel, 2.46 for residual fuel
BSFC = fuel consumption in g/kW-hr
= 200 g/kW-hr used for this analysis
FSC = percentage of sulfur in fuel that is converted to direct sulfate PM
= 2.247% used for this analysis
MWR = molecular weight ratio of sulfate PM to sulfur
= 224/32 = 7 used for this analysis
emission factors were based upon a fuel sulfur to SOz conversion formula which was
supplied by ENVIRON.21 Emission factors for SOz emissions were calculated using the formula
assuming that 97.753 percent of the fuel sulfur was converted to S02.22 The brake specific fuel
consumption (BSFC) that was used for SSDs was 195 g/kWh, while the BSFC that was used for
MSDs was 210 g/kWh based upon Lloyds 1995. The BSFC that was used for STs and GTs was
305 g/kWh based upon Entec.1
Equation 2-3 Calculation of SO2 Emission Factors, g/kWh
S02 EF = BSFC x 64/32 x 0.97753 x Fuel Sulfur Fraction
C02 emission factors were calculated from the BSFC assuming a fuel carbon content of 86.7
percent by weight17 and a ratio of molecular weights of COz and C at 3.667.
Equation 2-4 Calculation of CO2 Emission Factors, g/kWh
C02 EF = BSFC x 3.667x0.867
A PM10 is particulate matter of 10 micrometers or less.
B Brake specific fuel consumption is sometimes called specific fuel oil consumption (SFOC).
2-10
-------
The most current set of auxiliary engine emission factors also comes from Entec except as
noted below for PM and S02. Table 2-5 provides these auxiliary engine emission factors.
Table 2-5 Auxiliary Engine Emission Factors by Fuel Type, g/kWh
Engine
MSD
Fuel
RM
MDO
All Ports
NOX
14.70
13.90
CO
1.10
1.10
HC
0.40
0.40
CO2
668.36
668.36
West Coast Ports
PM10
1.4
0.6
PM25
1.3
0.55
SO2
10.26
6.16
Other Ports
PM10
1.4
0.6
PM25
1.3
0.55
SO2
11.09
6.16
emission factors were calculated using Equation 2-3 while PM emissions were
determined using Equation 2-2.
Using the ratios of RM versus MDO use15 as given in Table 2-3 together with the emission
factors shown in Table 2-5, the auxiliary engine emission factor averages by ship type are listed in
Table 2-6. As discussed above, this fuel sulfur level may be too high for the U.S. However, we do
not believe this emission factor has a significant effect on the total emission inventory estimates.
Table 2-6 Auxiliary Engine Emission Factors by Ship Type, g/kWh
Ship Type
Passenger
Others
All Ports
NOX
14.64
14.47
CO
1.10
1.10
HC
0.40
0.40
C02
668.36
668.36
West Coast Ports
PM10
1.3
1.1
PM25
1.2
1.0
S02
9.93
9.07
Other Ports
PM10
1.4
1.2
PM25
1.3
1.1
S02
10.70
9.66
2.3.2.3.7 Low Load Adjustment Factors for Propulsion Engines
Emission factors are considered to be constant down to about 20 percent load. Below that
threshold, emission factors tend to increase as the load decreases. This trend results because diesel
engines are less efficient at low loads and the brake specific fuel consumption (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. (EEA) demonstrated this effect in a study
prepared for the U.S. Government in 2000.23 In the EEA report, equations have been developed for
the various emissions. The low-load emission factor adjustment factors were developed based upon
the concept that the BSFC increases as load decreases below about 20 percent load.
Using these algorithms, fuel consumption and emission factors versus load were calculated.
By normalizing emission factors to 20% load, low-load multiplicative adjustment factors were
calculated for propulsion engines and presented in Table 2-7. Due to how they are operated, there is
no need for a low load adjustment factor for auxiliary engines.
2-11
-------
Table 2-7 Calculated Low Load Multiplicative Adjustment Factors
Load (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
NOX
11.47
4.63
2.92
2.21
1.83
1.60
1.45
1.35
1.27
1.22
1.17
1.14
1.11
1.08
1.06
1.05
1.03
1.02
1.01
1.00
HC
59.28
21.18
11.68
7.71
5.61
4.35
3.52
2.95
2.52
2.20
1.96
1.76
1.60
1.47
1.36
1.26
1.18
1.11
1.05
1.00
CO
19.32
9.68
6.46
4.86
3.89
3.25
2.79
2.45
2.18
1.96
1.79
1.64
1.52
1.41
1.32
1.24
1.17
1.11
1.05
1.00
PM
19.17
7.29
4.33
3.09
2.44
2.04
1.79
1.61
1.48
1.38
1.30
1.24
1.19
1.15
1.11
1.08
1.06
1.04
1.02
1.00
S02
5.99
3.36
2.49
2.05
1.79
1.61
1.49
1.39
1.32
1.26
1.21
1.18
1.14
1.11
1.09
1.07
1.05
1.03
1.01
1.00
C02
5.82
3.28
2.44
2.01
1.76
1.59
1.47
1.38
1.31
1.25
1.21
1.17
1.14
1.11
1.08
1.06
1.04
1.03
1.01
1.00
2.3.2.3.8 Use of Detailed Typical Port Data for Other Inputs
There is currently not enough information to readily calculate time-in-mode (hours/call) for
all 117 ports during the maneuvering and hotelling modes of operation. As a result, it was
necessary to review and select available detailed emission inventories that have been estimated for
selected ports to date. These ports are referred to as typical ports. The typical port information for
maneuvering and hotelling time-in-mode (as well as maneuvering load factors for the propulsion
engines) was then used for the typical ports and also assigned to the other modeled ports. A
modeled port is the port in which emissions are to be estimated. The methodology that was used to
select the typical ports and match these ports to the other modeled ports is briefly described in
Appendix 2D, and more fully described in an ICF report.2
2.3.2.3.9 Port Domestic Traffic
One of the concerns with using USAGE entrances and clearances data is that it only contains
foreign cargo movements moved by either a foreign flag vessel or a U.S. flag vessel. As a result,
U.S. flag ships carrying domestic cargo (i.e., Jones Act) ships are not included in the USAGE data.
Determining the contribution of Jones Act ships is difficult as most data sources include Category 1
and 2 Jones Act ship movements with Category 3 ships and do not provide either enough data or a
method for separating them.
Under contract to the U.S. Government, ICF conducted an analysis to estimate the amount
of Category 3 Jones Act ships calling at the 117 U.S. ports. This was done by analyzing marine
2-12
-------
exchange data obtained from port authorities for eleven typical ports and using this information to
estimate the Jones Act ship contribution for the remaining ports. Based on this limited analysis,
Jones Act ships are estimated to account for 9.2% of the total installed power calling on U.S. ports.
Approximately 30% of these ships, largely in the Alaska and Pacific regions, have been included in
the 2002 baseline inventory. Based on this analysis, Jones Act ships excluded from this inventory
constitute roughly 6.5% of total installed power.24 This results in an underestimation of the port
ship inventory and therefore the benefits of the EGA program reported in this chapter are also
underestimated.
2.3.2.4 2002 Near Port Inventories
This section provides a summary of the total port emissions for 2002. Table 2-8 provides a
breakout of the total port emissions by auxiliary and propulsion engines. Table 2-9 provides the
breakout by mode of operation, while Table 2-10 provides a summary of port emissions by ship
type.
Table 2-8 2002 Port Emissions Summary by Engine and Port Type (metric tonnes)
Engine Type
Propulsion
Auxiliary
All
Port Type
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Grand Total
Metric Tonnes
NOX
64,288
248
64,536
57,317
302
57,619
121,606
549
122,155
PM10
5,478
25
5,503
5,052
25
5,077
10,530
50
10,580
PM2.5
5,034
23
5,057
4,597
23
4,620
9,631
46
9,677
HC
2,532
11
2,543
1,615
8
1,624
4,148
19
4,167
CO
6,329
22
6,351
4,306
23
4,328
10,635
45
10,680
S02
52,676
187
52,863
41,232
202
41,433
93,908
389
94,297
C02
2,360,435
11,267
2,371,702
2,635,436
13,944
2,649,380
4,995,871
25,210
5,021,082
Auxiliary emissions at ports are responsible for 39-48% of the total port inventory,
depending on the pollutant. Hotelling, cruise, and RSZ modes of operation are all important
contributors to emissions.
2-13
-------
Table 2-9 2002 Port Emissions Summary by Mode and Port Type (metric tonnes)
Mode
Cruise
RSZ
Maneuvering
Hotelling
All
Port Type
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Grand Total
Metric Tonnes
NOx
34,193
183
34,376
34,427
45
34,472
7,383
70
7,452
45,603
252
45,855
121,606
549
122,155
PM10
2,826
17
2,843
2,887
4
2,891
758
7
765
4,060
21
4,081
10,530
50
10,580
PM2.5
2,623
16
2,639
2,657
4
2,661
625
7
632
3,726
19
3,745
9,631
46
9,677
HC
1,141
6
1,148
1,280
2
1,281
440
4
444
1,287
7
1,294
4,148
19
4,167
CO
2,651
14
2,665
3,804
4
3,808
724
8
732
3,456
19
3,475
10,635
45
10,680
SO2
21,186
137
21,323
35,148
33
35,181
4,356
50
4,406
33,218
168
33,386
93,908
389
94,297
C02
1,314,146
8,313
1,322,459
1,318,897
2,052
1,320,950
266,262
3,213
269,476
2,096,566
11,631
2,108,197
4,995,871
25,210
5,021,082
Table 2-10 2002 Port Emissions Summary by Ship Type and Port Type (metric tonnes)
Ship Type
Auto Carrier
Barge Carrier
Self-Unloading
Bulk Carrier
Other Bulk
Carrier
Container
General Cargo
Miscellaneous
Port Type
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Metric Tonnes
NOX
5,125
0
5,125
148
0
148
0
276
276
19,373
227
19,600
33,990
0
33,990
7,402
22
7,424
179
0
PM10
421
0
421
13
0
13
0
27
27
1,570
19
1,589
2,733
0
2,733
630
2
631
16
0
PM2.5
384
0
384
12
0
12
0
25
25
1,431
17
1,448
2,494
0
2,494
576
2
578
15
0
HC
185
0
185
5
0
5
0
10
10
633
7
640
1,282
0
1,282
251
1
252
6
0
CO
577
0
577
12
0
12
0
23
23
1,732
18
1,750
2,833
0
2,833
684
2
686
35
0
S02
3,676
0
3,676
141
0
141
0
210
210
14,945
147
15,092
22,628
0
22,628
6,208
15
6,223
128
0
C02
198,637
0
198,637
6,364
0
6,364
0
13,273
13,273
767,825
9,807
777,632
1,288,596
0
1,288,596
302,338
969
303,307
8,209
0
2-14
-------
Ship Type
Passenger
Refrigerated
Cargo
Roll-On/Roll-Off
Tanker
Ocean Going Tug
Integrated Tug-
Barge
All
Port Type
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Total
Deep Water
Great Lakes
Grand Total
Metric Tonnes
NOX
179
19,165
0
19,165
3,027
0
3,027
3,391
0
3,391
29,758
22
29,780
48
0
48
0
3
3
121,606
549
122,155
PM10
16
1,819
0
1,819
247
0
247
281
0
281
2,796
2
2,798
5
0
5
0
0
0
10,530
50
10,580
PM2.5
15
1,668
0
1,668
226
0
226
259
0
259
2,562
2
2,564
4
0
4
0
0
0
9,631
46
9,677
HC
6
578
0
578
98
0
98
113
0
113
994
1
995
2
0
2
0
0
0
4,148
19
4,167
CO
35
1,470
0
1,470
313
0
313
278
0
278
2,695
2
2,697
6
0
6
0
0
0
10,635
45
10,680
SO2
128
14,184
0
14,184
1,968
0
1,968
2,193
0
2,193
27,802
15
27,817
34
0
34
0
2
2
93,908
389
94,297
CO2
8,209
893,157
0
893,157
130,060
0
130,060
139,396
0
139,396
1,259,107
1,012
1,260,119
2,182
0
2,182
0
149
149
4,995,871
25,210
5,021,082
2.3.3 Interport Emissions
This section presents our nationwide analysis of the methodology and inputs used to
estimate interport emissions from main propulsion and auxiliary engines used by Category 3 ocean-
going vessels for the 2002 calendar year. The modeling domain for vessels operating in the ocean
extends from the U.S. coastline to a 200 nautical mile boundary. For ships operating in the Great
Lakes, it extends out to the international boundary with Canada. The emission results are divided
into nine geographic regions of the U.S. (including Alaska and Hawaii), and then totaled to provide
a national inventory.
The interport emissions described in this section represent total interport emissions prior to
any adjustments made to incorporate near-port inventories. The approach used to replace the near-
port portion of the interport emissions is provided in Section 2.3.4.
2.3.3.1 Interport Methodology
The interport emissions were estimated using the Waterway Network Ship Traffic, Energy,
and Environmental Model (STEEM).3'4 STEEM was developed by the University of Delaware as a
comprehensive approach to quantify and geographically represent interport ship traffic, emissions,
and energy consumption from large vessels calling on U.S. ports or transiting the U.S. coastline to
other destinations, and shipping activity in Canada and Mexico. The model estimates emissions
from main propulsion and auxiliary marine engines used on Category 3 vessels that engage in
2-15
-------
foreign commerce using historical North American shipping activity, ship attributes (i.e.,
characteristics), and activity-based emission factor information. These inputs are assembled using a
CIS platform that also contains an empirically derived network of shipping lanes. It includes the
emissions for all ship operational modes from cruise in unconstrained shipping lanes to
maneuvering in a port. The model, however, excludes hotelling operations while the vessel is
docked or anchored, and very low speed maneuvering close to a dock. For that reason, STEEM is
referred to as an "interport" model, to easily distinguish it from the near ports analysis.
STEEM uses advanced ArcGIS tools and develops emission inventories in the following
way. The model begins by building a spatially-defined waterway network based on empirical
shipping location information from two global ship reporting databases. The first is the
International Comprehensive Ocean-Atmosphere Data Set (ICOADS), which contains reports on
marine surface and atmospheric conditions from the Voluntary Observing Ships (VOS) fleet.25
There are approximately 4,000 vessels worldwide in the VOS system. The ICOADS project is
sponsored by the National Oceanic and Atmospheric Administration and National Science
Foundation's National Center for Atmospheric Research (NCAR). The second database is the
Automated Mutual-Assistance Vessel Rescue (AMVER) system.26 The AMVER data set is based
on a ship search and rescue reporting network sponsored by the U.S. Coast Guard. The AMVER
system is also voluntary, but is generally limited to ships over 1,000 gross tons on voyages of 24
hours or longer. About 8,600 vessels reported to AMVER in 2004.
The latitude and longitude coordinates for the ship reports in the above databases are used to
statistically create and spatially define the direction and width of each shipping lane in the waterway
network. Each statistical lane (route and segment) is given a unique identification number for
computational purposes. For the current analysis, STEEM used 20 years of ICOADS data (1983-
2002) and about one year of AMVER data (part of 2004 and part of 2005) (Figure 2-2).
Nautical Miles
Figure 2-2 AMVER and ICOADS data
Every major ocean and Great Lake port is also spatially located in the waterway network
using ArcGIS software. For the U.S., the latitude and longitude for each port is taken from the
USAGE report on vessel entrances and clearances.11 Each port also has a unique identification
number for computational purposes.
2-16
-------
As illustrated in Figure 2-3, the waterway network represented by STEEM resembles a
highway network on land. It is composed of ports, which are origins and destinations of shipping
routes: junctions where shipping routes intersect, and segments that are shipping lanes between two
connected junctions. Each segment can have only two junctions or ports, and ship traffic flow can
enter and leave a segment only through a junction or at a port. The figure represents only a sample
of the many routes contained in the model.
| | 0 • 5,000
^11 5.000 - 50.000
^H > 50,000
Nautical Mil-23
Figure 2-3 Illustration of STEEM Modeling Domain and Spatial Distribution of Shipping Lanes
The STEEM interport model also employs a number of databases to identify the movements
for each vessel (e.g., trips), individual ship attributes (e.g., vessel size and horsepower), and related
emission factor information (e.g., emission rates) that are subsequently used in the inventory
calculations.
To allocate ships to the statistical lanes, STEEM uses ArcGIS Network Analyst tools along
with specific information on each individual ship movement to solve the most probable path on the
network between each pair of ports (i.e., a trip) for a certain ship size. This is assumed to represent
the least-energy path, which in most cases is the shortest distance unless prevented by weather or
sea conditions, water depth, channel width, navigational regulations, or other constraints that are
beyond the model's capability to forecast.
After identifying the shipping route and resulting distance associated with each unique trip,
the emissions are simply calculated for each operational mode using the following generalized
equation along with information from the ship attributes and emission factor databases:
Equation 2-5
Emissions per trip = distance (nautical miles) / speed (nautical miles/hour) x horsepower (kW) x
fractional load factor x emission factor (g/kW-hour)
In STEEM, emissions are calculated separately for distances representing cruise and
maneuvering operational modes. Maneuvering occurs at slower speeds and load factors than
during cruise conditions. In STEEM, maneuvering is assumed to occur for the first and last
2-17
-------
20 kilometers of each trip when a ship is entering or leaving a port. A ship is assumed to
move at maneuvering speed for an entire trip if the distance is less than 20 kilometers.
Finally, the emissions along each shipping route (i.e., segment) for all trips are proportioned
among the respective cells that are represented by the gridded modeling domain. For this work,
emissions estimates were produced at a cell resolution of 4 kilometers by 4 kilometers, which is
appropriate for most atmospheric air quality models. The results for each cell are then summed, as
appropriate, to produce emission inventories for the various geographic regions of interest in this
analysis.
2.3.3.2 Inputs for Interport Emission Calculations
The STEEM model includes detailed information about ship routes and destinations in order
to provide spatially allocated emissions of ships in transit. The shipping lanes and directions were
empirically derived from ship positioning data in several datasets. The International
Comprehensive Ocean-Atmosphere Data Set (ICOADS) contains reports on marine surface and
atmospheric conditions from the Voluntary Observing Ships (VOS) fleet.27 STEEM also uses a
dataset derived from the Automated Mutual-Assistance Vessel Rescue (AMVER) system,28 which
is based on a ship search and rescue reporting network sponsored by the U.S. Coast Guard. Traffic
along each of these lanes is derived from USAGE entrance and clearance data for 2002,29 together
with Lloyd's Register-Fairplay Ltd's data for ship characteristics. Information for number of calls,
ship characteristics, propulsion engine power, and cruise speed were obtained from these data.
The emission factors and load factors used as inputs to STEEM are very similar to those
used for the ports analysis. Additional adjustments were made to interport emission results for
PMio and S02 in order to reflect recent U.S. Government review of available engine test data and
fuel sulfur levels. Details of the STEEM emission inputs and adjustments are located in Appendix
2E.
2.3.3.3 Interport Domestic Traffic
As previously noted, STEEM includes the emissions associated with ships that are engaged
in foreign commerce. As a result, U.S. flag vessels carrying domestic cargo (Jones Act ships) are
not included. The STEEM interport analysis also roughly estimated the emissions associated with
these ships that are engaged solely in domestic commerce.1'4 Specifically, the interport analysis
estimated that the large ocean-going vessels carrying only domestic cargo excluded from STEEM
represent approximately 2-3 percent of the total U.S. emissions.
In section 2.3.2.3.9 in the estimation of port inventories, the estimate of excluded installed
power was roughly 6.5 percent. It is not inconsistent that the STEEM estimate of excluded
emissions is lower than the excluded power estimated from calls to U.S. ports, since the STEEM
model includes ships that are transiting without stopping at U.S. ports. Since most of the Jones Act
ships tend to travel closer along the coast line, most of the Jones Act ship traffic is expected to fall
within the proposed EGA. Therefore, the results presented in this chapter are expected to
underpredict the benefits of the proposed EGA.
2-18
-------
2.3.4 Combining the Near Port and Interport Inventories
The national and regional inventories in this study are a combination of the results from the
near ports analysis described in Section 2.3.2 and the STEEM interport modeling described in
Section 2.3.3. The two inventories are quite different in form. As previously presented, the
STEEM modeling domain spans the Atlantic and Pacific Oceans in the northern hemisphere. The
model characterizes emissions from vessels while traveling between ports. That includes when a
vessel is maneuvering a distance of 20 kilometers to enter or exit a port, cruising near a port as it
traverses the area, or moving in a shipping lane across the open sea. For the U.S., STEEM includes
the emissions associated with 251 ports. The results are spatially reported in a gridded format that
is resolved to a cell dimension of 4 kilometers by 4 kilometers.
The near port results, however, are much more geographically limited and are not reported
in a gridded format. The analysis includes the emissions associated with ship movements when
entering or exiting each of 117 major U.S. ports. For deep water ports that include when a vessel is
hotelling and maneuvering in the port, operating in the RSZ that varies in length for each port, and
cruising 25 nautical miles between the end of the RSZ and an unconstrained shipping lane. For
Great Lakes ports that includes hotelling and maneuvering, three nautical miles of RSZ operation,
and cruising 7 nautical miles between the end of the RSZ and open water. The results are reported
for each port and mode of operation.
To precisely replace only the portion of the STEEM interport inventory that is represented in
the near port inventory results, it is necessary to spatially allocate the emissions in a format that is
compatible with the STEEM 4 kilometers by 4 kilometers gridded output. Once that has been
accomplished, the two inventories can be blended together. Both of these processes are described
below. This work was conducted by ENVIRON International as a subcontractor under the U.S.
Government contract with ICF.2
2.3.4.1 Spatial Location of the Near Port Inventories
The hotelling, maneuvering, RSZ, and cruise emissions from the near port inventories were
spatially located by their respective latitude and longitude coordinates using ArcGIS software. For
this study, shapefiles were created that depicted the emission locations as described above.
Additional shapefiles were also obtained to locate other geographic features such as the coastline
and rivers of the U.S. These shapefiles and the STEEM output can be layered upon each other,
viewed in ArcMap, and analyzed together. The following sections provide a more detailed
description of how the shapefiles representing the ports, RSZ lanes, and cruise lanes were
developed.
2.3.4.1.1 Ports
Each port, and thus the designated location for hotelling and maneuvering emissions, is
modeled as a single latitude/longitude coordinate point using the port center as defined by USAGE
in the Principal Ports of the United States dataset. The hotelling and maneuvering emissions
represented by the latitude/longitude coordinate for each port were subsequently assigned to a single
cell in the gridded inventory where that point was located. It should be noted that modeling a port
as a point will over specify the location of the emissions associated with that port if it occupies an
2-19
-------
area greater than one grid cell, or 4 kilometers by 4 kilometers. The coordinates of all of the 117
ports used in this work are shown in Appendix 2A.
2.3.4.2 Reduced Speed Zone Operation
The RSZ routes associated with each of the 117 ports were modeled as lines. Line
shapefiles were constructed using the RSZ distance information described in Section 2.3.2 and the
USAGE National Waterway Network (NWN) geographic database of navigable waterways in and
around the U.S.14 The coordinates of RSZ endpoints for all of the 117 ports used in this work are
shown in Appendix 2C.
The RSZ emissions were distributed evenly along the length of the line. The
latitude/longitude coordinates for each point along the line were subsequently used to assign the
emissions to a grid cell based on the proportion of the line segment that occurred in the respective
cell.
2.3.4.2.1 Cruise Operations
The cruise mode links that extend 25 nautical miles for deep water ports or 7 nautical miles
for Great Lake ports from the end of the RSZ end point were also modeled with line shapefiles.
These links were spatially described for each port following the direction of the shipping lane
evident in the STEEM data. Again, as with RSZ emissions, the latitude/longitude coordinates for
each point along the line were subsequently used to assign the emissions to a grid cell based on the
proportion of the line segment that occurred in the respective cell.
2.3.4.3 Combining the Near Port and STEEM Emission Inventories in Port Areas
After spatially defining the geographic location of the near port emissions, but before
actually inserting them into the gridded STEEM inventory, it was necessary to determine if all of
the STEEM emissions within an affected cell should be replaced, or if some of the emissions should
be retained. In this latter case, ships would be traversing the area near a port, but not actually
entering or exiting the port.
The percentage of STEEM emissions that are attributable to a port, and should be removed
and replaced, were approximated by dividing the STEEM emissions in the isolated portion of the
route that lead only to the port, with the STEEM emissions in the major shipping lane.
The actual merging of the two inventories was performed by creating a number of databases
that identified the fraction of the near port inventory for each pollutant species and operating mode
that should be added to the grid cells for each port. A similar database was also created that
identified how much of the original STEEM emissions should be reduced to account for ship
movements associated directly with a port, while preserving those that represented transient vessel
traffic. These databases were subsequently used to calculate the new emission results for each
affected cell in the original STEEM gridded inventory, resulting in the combined inventory results
for this study.
In a few cases, the outer edges of the port inventories fell outside the international boundary;
that portion outside the U.S. boundary was removed. As a result, the port totals presented in the
2-20
-------
next section are slightly less than those reported in Section 2.3.2.4. The removed portion represents
less than 2 percent of the total port emissions.
Since STEEM includes emissions associated with 251 ports, the 117 ports do not cover all
the ports identified by the shipping lane paths evident in the STEEM data. In the remaining ports,
the STEEM model output was used.
2.3.5 2002 Baseline Inventories
The modeling domain of the new combined emission inventory described above is the same
as the original STEEM domain, i.e., the Atlantic and Pacific Oceans, the Gulf of Mexico, the Great
Lakes, Alaska, and Hawaii. Inventories for the nine geographic regions of the U.S. specified in
Section 2.2 were created using ArcGIS software to intersect the regional shapefiles with the 4
kilometers by 4 kilometers gridded domain. Any grid cell split by a regional boundary was
considered to be within a region if over 50 percent of its area was within the region. The emissions
from the cells within a region were then summed. The final emission inventories for 2002 are shown
in Table 2-11 for each of the nine geographic regions and the nation. The geographic scope of these
regions was previously displayed in Figure 2-1.
Table 2-11 2002 Regional and National Emissions from Category 3 Vessel Main and Auxiliary Engines
Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
Total Metric Tonnes
Metric Tonnes
NOX
18,051
60,019
219,560
172,897
22,600
31,799
26,037
104,155
15,019
670,135
PM10
1,425
4,689
17,501
14,043
1,775
2,498
2,154
8,094
1,179
53,358
PM25a
1,311
4,313
16,101
12,920
1,633
2,297
1,982
7,447
1,085
49,089
HC
597
1,989
7,277
5,757
749
1,053
938
3,464
498
22,324
CO
1,410
4,685
17,231
14,169
1,765
2,484
2,090
8,437
1,174
53,444
SO2
10,618
34,786
145,024
104,852
13,182
18,546
15,295
60,443
8,766
411,511
CO2
657,647
2,143,720
8,131,553
6,342,139
818,571
1,151,725
990,342
3,796,572
541,336
24,573,605
a Estimated from PMi0 using a multiplicative adjustment factor of 0.92.
The relative contributions of the near port and interport emission inventories to the total U.S.
ship emissions are presented in Table 2-12 and Table 2-13. As expected, based on the geographic
scope of the two types of inventories, the interport and near port inventories are about 80 percent
and 20 percent of the total, respectively.
2-21
-------
Table 2-12 2002 Contribution of Near Port and Interport Emissions to the Total C3 Inventory
Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
Total Metric Tonnes
Metric Tonnes
NOX
Port
833
0
48,313
33,637
2,916
0
14,015
20,079
491
120,285
Interport
17,218
60,019
171,247
139,260
19,684
31,799
12,022
84,076
14,528
549,852
Total
18,051
60,019
219,560
172,897
22,600
31,799
26,037
104,155
15,019
670,137
PM10
Port
80
0
4,126
3,169
251
0
1,216
1,525
44
10, 413
Interport
1,345
4,689
13,375
10,874
1,524
2,498
938
6,569
1,135
42,945
Total
1,425
4,689
17,501
14,043
1,775
2,498
2,154
8,094
1,179
53,358
PM25a
Port
74
0
3,796
2,916
231
0
1,119
1,403
41
9,580
Interport
1,237
4,313
12,305
10,004
1,402
2,297
863
6,044
1,044
39,510
Total
1,311
4,313
16,101
12,920
1,633
2,297
1,982
7,447
1,085
49,089
a Estimated from PMi0 using a multiplicative adjustment factor of 0.92.
Table 2-13 2002 Contribution of Near Port and Interport Emissions to the Total C3 Inventory (Cont'd)
Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
Total Metric Tonnes
Metric Tonnes
HC
Port
27
0
1,603
1,142
96
0
540
678
17
4,103
Interport
570
1,989
5,674
4,615
653
1,053
398
2,786
481
18,219
Total
597
1,989
7,277
5,757
749
1,053
938
3,464
498
22,322
CO
Port
66
0
3,864
3,305
230
0
1,152
1,876
40
10,533
Interport
1,344
4,685
13,367
10,864
1,535
2,484
938
6,561
1,134
42,912
Total
1,410
4,685
17,231
14,169
1,765
2,484
2,090
8,437
1,174
53,445
SO2
Port
641
0
45,952
24,187
1,891
0
8,329
11,715
346
93,062
Interport
9,977
34,786
99,072
80,665
11,291
18,546
6,966
48,728
8,420
318,450
Total
10,618
34, 786
145,024
104,852
13,182
18,546
15,295
60,443
8,766
411,512
As noted previously, these inventories exclude a portion of traffic from U.S. flag ships
carrying domestic cargo. Estimates range from roughly 2 to 7 percent of installed power, which
indicates that the inventories may be underestimated by 2 to 7 percent.
2.4 Development of 2020 Inventories
2.4.1 Outline of Methodology
The emissions from Category 3 ocean-going vessels (main propulsion and auxiliary engines)
are projected to 2020 by applying certain growth factors to the 2002 emission inventories to account
for the expected change in ship traffic over these time periods due to growth in trade.
The remaining sections describe the derivation of the growth adjustment factors for each of
the modeling regions described in Section 2.2. Emission control program related adjustments to the
2-22
-------
2020 inventories are then described. A baseline inventory and an inventory within the proposed
EGA are then presented.
2.4.2 Growth Factors by Geographic Region
The growth factors that are used to estimate future year emission inventories are based on
the expected demand for marine bunker fuels that is associated with shipping goods, i.e.,
commodities, into and out of the U.S. This section describes the growth factors that are used to
project the emissions to 2020 for each of the nine geographic regions evaluated in this analysis.
The use of bunker fuel as a surrogate for estimating future emissions is appropriate because the
quantity of fuel consumed by C3 engines is highly correlated with the amount of combustion
products, i.e., pollutants that are emitted from those vessels. The term bunker fuel in this report also
includes marine distillate oil and marine gas oil that are used in some auxiliary power engines.
The remainder of this section first summarizes the development of growth rates by RTI
International (RTI) for five geographic regions of the U.S., as performed under contract to the U.S.
government.5'6 This is followed by the derivation of the growth factors that are used in this study
for the nine geographic regions of interest.
2.4.2.1 Summary of Regional Growth Rate Development
RTI developed fuel consumption growth rates for five geographic regions of the U.S. These
regions are the East Coast, Gulf Coast, North Pacific, South Pacific, and Great Lakes. The amount
of bunker fuel required in any region and year is based on the demand for transporting various types
of cargo by Category 3 vessels. This transportation demand is in turn driven by the demand for
commodities that are produced in one location and consumed in another, as predicted by an
econometric model. The flow of commodities is matched with typical vessels for trade routes
(characterized according to cargo capacity, engine horsepower, age, specific fuel consumption, and
engine load factors). Typical voyage parameters are then assigned to the trade routes that include
average ship speed, round trip mileage, tons of cargo shipped, and days in port. Fuel consumption
for each trade route and commodity type thus depends on commodity projections, ship
characteristics, and voyage characteristics. Figure 2-4 illustrates the approach to developing
baseline projections of marine fuel consumption.
As a means of comparison, the IMO Secretary General's Informal Cross
Government/Industry Scientific Group of Experts presented a growth rate that ranged from 3.3% to
3.7%.30 RTFs overall U.S. growth rate was projected at 3.4%, which is consistent with that range.
2-23
-------
Ship Analysis: by Vessel Type and Size Category
Inputs Outputs
Deadweight for all Vessels of
Given Type & Sizea
Horsepower, Year of Build
for all Vessels of Given
Type & Size8
Specific Fuel Consumption
(g/SHP-HR) by Year of Build"
Engine Load Factors0
Average Cargo /7\
Carried (Tons) n^-y
Average Daily Fuel
i« Consumption
(Tons/Day)
Average Daily Fuel
Consumption (Tons/Day) J/'R^
- Main, Aux. Engine at Sea A )
- Aux. Engine in Port
Trade Analysis: by Commodity and Trade Route
Inputs
Average Ship Speed0
Round Trip Mileaged
Tons of Cargo Shipped6
Average Cargo CarriedA~A
per Ship Voyage l^^,
Outputs
Days at Sea and in
, Port, per Voyage
Total Days at »f~c\
i Sea and in Port V, J
Total Estimated Bunker Fuel Demand
f N
Average Daily Fuel Consumption
(Tons/Day) x Total Days at Sea _ Bunker Fuel
- Main, Aux. Engine at Sea (^\ and in Port (^\ Demand
- Aux. Engine in Port V. } \ }
\^ J
Driven by changes in engine efficiency. Driven growth in
commodity flows.
a - Clarksons Ship Register Database
b - Engine Manufacturers' Data, Technical Papers
c - Corbett and Wang (2005) "Emission Inventory Review: SECA Inventory Progress Discussion"
d - Combined trade routes and heavy leg analysis
e - Global Insight Inc. (Gil) Trade Flow Projections
Figure 2-4 Illustration of Method for Estimating Bunker Fuel Demand
2.4.2.2 Trade Analysis
The trade flows between geographic regions of the world, as illustrated by the middle
portion of Figure 2-4 were defined for the following eight general types of commodities:
- liquid bulk - crude oil
- liquid bulk - refined petroleum products
- liquid bulk - residual petroleum products
- liquid bulk - chemicals (organic and inorganic)
- liquid bulk -gas (including LNG and LPG)
dry bulk (e.g., grain, coal, steel, ores and scrap)
- general cargo (e.g., lumber/forest products)
2-24
-------
containerized cargo
The analysis specifically evaluated trade flows between 21 regions of the world. Table 2-14
shows the countries associated with each region.
Table 2-14 Aggregate Regions and Associated Countries
Aggregate Regions
U.S. Atlantic Coast
U.S. Great Lakes
U.S. Gulf Coast
E. Canada3
W. Canada3
U.S. Pacific North
U.S. Pacific South
Greater Caribbean
South America
Africa - West
Africa-North/East-
Mediterranean
Africa-East/South
Europe-North
Europe-South
Europe-East
Caspian Region
Russia/FSU
Middle East Gulf
Australia/NZ
Japan
Pacific-High Growth
China
Rest of Asia
Base Countries / Regions
U.S. Atlantic Coast
U.S. Great Lakes
U.S. Gulf Coast
Canada3
Canada3
U.S. Pacific North
U.S. Pacific South
Colombia, Mexico, Venezuela, Caribbean Basin, Central America
Argentina, Brazil, Chile, Peru, Other East Coast of S. America, Other
West Coast of S. America
Western Africa
Mediterranean Northern Africa, Egypt, Israel,
Kenya, Other Eastern Africa, South Africa, Other Southern Africa
Austria, Belgium, Denmark, Finland, France, Germany, Ireland,
Netherlands, Norway, Sweden, Switzerland, United Kingdom
Greece, Italy, Portugal, Spain, Turkey, Other Europe
Bulgaria, Czech Republic, Hungary, Poland, Romania, Slovak Republic
Southeast CIS
The Baltic States, Russia Federation, Other Western CIS
Jordan, Saudi Arabia, UAE, Other Persian Gulf
Australia, New Zealand
Japan
Hong Kong S.A.R., Indonesia, Malaysia, Philippines, Singapore, South
Korea, Taiwan, Thailand
China
Viet Nam, India, Pakistan, Other Indian Subcontinent
3 Canada is treated as a single destination in the GI model. Shares of Canadian imports from
and exports to regions of the world in 2004 are used to divide Canada trade into shipments
to/from Eastern Canada ports and shipments to/from Western Canada ports.31
The overall forecast of demand for shipping services and bunker fuel was determined for
each of the areas using information on commodity flows from Global Insight's (GI) World Trade
Service. Specifically, GI provided a specialized forecast that reports the flow of each commodity
type for the period 1995-2024, based on a proprietary econometric model. The general structure of
the GI model for calculating trade flows assumes a country's imports from another country are
driven by the importing country's demand forces (given that the exporting country possesses
enough supply capacity), and affected by exporting the country's export price and importing
country's import cost for the commodity. The model then estimates demand forces, country-specific
exporting capacities, export prices, and import costs.
2-25
-------
The GI model included detailed annual region-to-region trade flows for eight composite
commodities from 1995 to 2024, in addition to the total trade represented by the commodities.
Table 2-15 illustrates the projections for 2012 and 2020, along with baseline data for 2005. In 2005,
dry bulk accounted for 41 percent of the total trade volume, crude oil accounted for 28 percent, and
containers accounted for 12 percent. Dry bulk and crude oil shipments are expected to grow more
slowly over the forecast period than container shipments. By 2020, dry bulk represents 39 percent
of the total, crude oil is 26 percent, and containers rise to 17 percent.
Table 2-15 Illustration of World Trade Estimates for Composite Commodities, 2005,2012, and 2020
Commodity Type
Dry Bulk
Crude Oil
Container
Refined Petroleum
General Cargo
Residual Petroleum and Other Liquids
Chemicals
Natural Gas
Total International Cargo Demand
Cargo (millions of tons)
2005
2,473
1,703
714
416
281
190
122
79
5,979
2012
3,051
2,011
1,048
471
363
213
175
91
7,426
2020
3,453
2,243
1,517
510
452
223
228
105
8,737
2.4.2.3 Ship Analysis by Vessel Type and Size
Different types of vessels are required to transport the different commodities to the various
regions of the world. Profiles of these ships were developed to identify the various vessel types and
size categories that are assigned to transport commodities of each type along each route. These
profiles include attributes such as ship size, engine horsepower, engine load factors, age, and engine
fuel efficiency. This information was subsequently used to estimate average daily fuel consumption
for each typical ship type and size category.
The eight GI commodity categories were mapped to the type of vessel that would be used to
transport that type of cargo using information from Clarkson's Snipping Database.32 These
assignments are shown in Table 2-16.
2-26
-------
Table 2-16 Assignment of Commodities to Vessel Types
Commodity Ship Category Vessel Type
Liquid bulk - crude oil
Liquid bulk - refined
petroleum products
Liquid bulk - residual
petroleum products
Liquid bulk - chemicals
(organic and inorganic)
Liquid bulk - natural gas
(including LNG and LPG)
Dry bulk (e.g. grain, coal,
steel, ores and scrap)
General cargo (including
neobulk, lumber/forest
products)
Containerizable cargo
Crude Oil Tankers
Product Tankers
Product Tankers
Chemical Tankers
Gas Carriers
Dry Bulk Carriers
General Cargo
Container Ships
Tanker
Product Carrier
Product Carrier
Chemical & Oil Carrier
LNG Carrier, LPG Carrier, Chemical & LPG Carrier,
Ethylene/LPG, Ethylene/LPG/Chemical,
LNG/Ethylene/LPG, LNG/Regasification, LPG/Chemical,
LPG/Oil, Oil & Liquid Gas Carrier
Bulk Carrier
General Cargo Liner, Reefer, General Cargo Tramp, Reefer
Fish Carrier, Ro-Ro, Reefer/Container, Ro-Ro
Freight/Passenger, Reefer/Fleet Replen., Ro-Ro/Container,
Reefer/General Cargo, Ro-Ro/Lo-Lo, Reefer/Pallets
Carrier, Reefer/Pass./Ro-Ro, Reefer/Ro-Ro Cargo
Fully Cellular Container
Each of the vessel types were classified by their cargo carrying capacity or deadweight tons
(DWT). The size categories were identified based on both industry definitions and natural size
breaks within the data. Table 2-17 summarizes the size categories that were used in the analysis and
provides other information on the general attributes of the vessels from Clarkson's Shipping
Database. The vessel size descriptions are also used to define shipping routes based on physical
limitations that are represented by canals or straits through which ships can pass. Very large crude
oil tankers are the largest by DWT rating, and the biggest container ships (Suezmax) are also very
large.
2-27
-------
Table 2-17 Fleet Characteristics
Ship
Type
Container
General
Cargo
Dry Bulk
Crude Oil
Tanker
Chemical
Tanker
Petroleum
Product
Tanker
Natural
Gas
Carrier
Other
Total
Size by
DWT
Suezmax
PostPanamax
Panamax
Intermediate
Feeder
All
Capesize
Panamax
Handymax
Handy
VLCC
Suezmax
AFRAmax
Panamax
Handymax
Coastal
All
AFRAmax
Panamax
Handy
Coastal
VLGC
LGC
Midsize
All
--
Maximum
Size
(DWT)
83,000
56,500
42,100
14,000
0
Maximum
Size
(DWT)
140,000
83,000
56,500
42,100
14,000
All
79,000
54,000
40,000
0
180,000
120,000
75,000
43,000
27,000
0
0
79,000
54,000
40,000
0
180,000
120,000
75,000
43,000
27,000
All
68,000
40,000
27,000
0
60,000
35,000
0
0
68,000
40,000
27,000
0
60,000
35,000
All
--
--
Number
of Ships
101
465
375
1,507
1,100
3,214
715
1,287
991
2,155
470
268
511
164
100
377
2,391
226
352
236
349
157
140
863
7,675
26,189
Total
DWT
(millions)
9.83
30.96
18.04
39.8
8.84
26.65
114.22
90.17
46.5
58.09
136.75
40.63
51.83
10.32
3.45
3.85
38.8
19.94
16.92
7.9
3.15
11.57
6.88
4.79
88.51
888.4
Total
Horse
Power
(millions)
8.56
29.3
15.04
32.38
7.91
27.07
13.81
16.71
10.69
19.58
15.29
5.82
8.58
2.17
1.13
1.98
15.54
3.6
4.19
2.56
1.54
5.63
2.55
3.74
53.6
308.96
Total
Kilowatts
(millions)
6.38
21.85
11.21
24.14
5.90
20.18
10.30
12.46
7.97
14.60
11.40
4.34
6.40
1.62
0.84
1.48
11.59
2.68
3.12
1.91
1.15
4.20
1.90
2.79
39.96
230.36
The average fuel consumption for each vessel type and size category was estimated in a
multi-step process using individual vessel data on engine characteristics. Clarkson's Shipping
Database Register provides each ship's total installed horsepower (HP), type of propulsion (diesel
or steam), and year of build. These characteristics are then matched to information on typical
specific fuel consumption (SFC), which is expressed in terms of grams of bunker fuel burned per
horsepower-hour (g/HP-hr, which is equivalent to 1.341 g/kW-hr).
The SFC values are based on historical data from Wartsila Sulzer, a popular manufacturer of
diesel engines for marine vessels. RTI added an additional 10 percent to the reported "test bed" or
"catalogue" numbers to account for the guaranteed tolerance level and an in-service SFC
2-28
-------
differential. Overall, the 10 percent estimate is consistent with other analyses that show some
variation between the "test bed" SFC values reported in the manufacturer product catalogues and
those observed in actual service. This difference is explained by the fact that old, used engines
consume more fuel than brand new engines and in-service fuels may be different than the test bed
fuels.33
Figure 2-5 shows SFC values that were used in the model regarding the evolution of specific
fuel oil consumption rates for diesel engines over time. Engine efficiency in terms of SFC has
improved over time, most noticeably in the early 1980s in response to rising fuel prices. However,
there is a tradeoff between improving fuel efficiency and reducing emissions. Conversations with
engine manufacturers indicate that it is reasonable to assume SFC will remain constant for the
projection period of this study, particularly as they focus on meeting NOx emission standard as
required by MARPOL Annex VI, or other potential pollution control requirements. Post-2000 SFC
values are constant at approximately 135 g/hp-hr (180 g/kW-hr).
200
180
160
140
120
100
80
60
40
20
0
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020
Figure 2-5 Diesel Engine Specific Fuel Consumption
RTI assumed a fixed SFC of 220 g/HP-hr (295 g/kW-hr) for steam engines operating on
bunker fuel.
Using the above information, the average daily fuel consumption (AFC), expressed in metric
tons of fuel at full engine load, for each vessel type and size category is found using the following
equation:
Equation 2-6
Fleet AFCV,, = —^[SFCVS xHPVS x 10~6tonnes/g]
2-29
-------
Where:
Fleet AFC = Average daily fuel consumption in metric tonnes at full engine load
- v = Vessel type
- s = Vessel size category
N = Number of vessels in the fleet
SFC = Specific fuel consumption in grams of bunker fuel burned per horsepower-
hour in use(g/HP-hr)
HP = Total installed engine power, in horsepower (HP)
106 tonnes/g = Conversion from grams to metric tonnes
As previously noted, AFC values calculated in the above equation are based on total
horsepower; therefore, they must be scaled down to reflect typical operation using less than 100
percent of the horsepower rating, i.e., actual engine load. Table 2-18 shows the engine load factors
that were used to estimate the typical average daily fuel consumption (tons/day) for the main
propulsion engine and the auxiliary engines when operated at sea and in port. 4
Table 2-18 Main and Auxiliary Engine Load Factors
Vessel Type
Container Vessels
General Cargo Carriers
Dry Bulk Carriers
Crude Oil Tankers
Chemical Tankers
Petroleum Product Tankers
Natural Gas Carrier
Other
Main
Engine
Load Factor
(%)
80
80
75
75
75
75
75
70
Auxiliary Engine as
Percent of Main
Engine
22.0
19.1
22.2
21.1
21.1
21.1
21.1
20.0
Auxiliary Engine as
Percent of Main Engine at
Sea
11.0
9.5
11.1
10.6
10.6
10.6
10.6
10.0
The RTI analysis also assumed that the shipping fleet changes over time as older vessels are
scrapped and replaced with newer ships. Specifically, vessels over 25 years of age are retired and
replaced by new ships of the most up-to-date configuration. This assumption leads to the following
change in fleet characteristics over the projection period:
• New ships have engines rated at the current SFC, so even though there are no further
improvements in specific fuel consumption, the fuel efficiency of the fleet as a whole
will improve over time through retirement and replacement.
• New ships will weigh as much as the average ship built in 2005, so the total cargo
capacity of the fleet will increase over time as smaller ships retire and are replaced.
• Container ships will increase in size over time on the trade routes between Asia to either
North America or Europe.
2.4.2.4 Trade Analysis by Commodity Type and Trade Route
Determining the total number of days at sea and in port requires information on the relative
amount of each commodity that is carried by the different ship type size categories on each of the
trade routes. For example, to serve the large crude oil trade from the Middle East Gulf region to the
2-30
-------
Gulf Coast of the U.S., 98 percent of the deadweight tonnage is carried on very large oil tankers,
while the remaining 2 percent is carried on smaller Suezmax vessels. After the vessel type size
distribution was found, voyage parameters were estimated. Specifically, these are days at sea and in
port for each voyage (based on ports called, distance between ports, and ship speed), and the
number of voyages (based on cargo volume projected by GI and the DTW from Clarkson's
Shipping Database). The length of each voyage and number of voyages were used to estimate the
total number of days at sea and at port, which is a parameter used later to calculate total fuel
consumption for each vessel type and size category over each route and for each commodity type.
(More information on determining the round trip distance for each voyage that is associated with
cargo demand for the U.S. is provided in Section 2.4.2.5.)
speed:
The days at sea were calculated by dividing the round trip distance by the average vessel
Equation 2-7
round trip distance route
Days at Sea Per Voyagevsroute =
speedvx x 24 hrs
Where:
v = Vessel type
s = Vessel size category
route = Unique trip itinerary
round trip route distance = Trip length in nautical miles
speed = Vessel speed in knots or nautical miles per hour
24 hrs = Number of hours in one day
Table 2-19 presents the speeds by vessel type that were used in the analysis.34 These values
are the same for all size categories, and are assumed to remain constant over the forecast period.
Table 2-19 Vessel Speed by Type
Vessel Type
Crude Oil Tankers
Petroleum Product Tankers
Chemical Tankers
Natural Gas Carriers
Dry Bulk Carriers
General Cargo Vessels
Container Vessels
Other
Speed (knots)
13.2
13.2
13.2
13.2
14.1
12.3
19.9
12.7
The number of voyages along each route for each trade was estimated for each vessel type v
and size category s serving a given route by dividing the tons of cargo moved by the amount of
cargo (DTW) per voyage:
2-31
-------
Equation 2-8
, T , r,T total metric tonnes of cargo moved
Number of Voyagesv tmde =
fleet average DWTv s x utilization rate
Where:
v = Vessel type
s = Vessel size category
trade = Commodity type
Fleet average DWT = Median dead weight tonnage carrying capacity in metric tons
Utilization rate = Fraction of total ship DWT capacity used
The cargo per voyage is based on the fleet average ship size from the vessel profile analysis.
For most cargo, a utilization rate of 0.9 is assumed to be constant throughout the forecast period.
Lowering this factor would increase the estimated number of voyages required to move the
forecasted cargo volumes, which would lead to an increase in estimated fuel demand.
In addition to calculating the average days at sea per voyage, the average days in port per
voyage was also estimated by assuming that most types of cargo vessels spend four days in port per
voyage. RTI notes, however, that this can vary somewhat by commodity and port.
2.4.2.5 Worldwide Estimates of Fuel Demand
This section describes how the information from the vessel and trade analyses were used to
calculate the total annual fuel demand associated with international cargo trade. Specifically, for
each year;; of the analysis, the total bunker fuel demand is the sum of the fuel consumed on each
route of each trade (commodity). The fuel consumed on each route of each trade is in turn the sum
of the fuel consumed for each route and trade for that year by propulsion main engines and auxiliary
engines when operated at sea and in port. These steps are illustrated by the following equations:
Equation 2-9
pp _ y y pp
1 ^y ~ ^ ^ L ^trade.route.year
trade route
= I, I, [AFCtradeirouteiyatseax DaysatSeatradeiroute]y+AFCtradeirouteiyatportxDaysatPorttradeirouteiy]
trade route
Where:
FC = Fuel consumed in metric tonnes
y = calendar year
trade = Commodity type
route = Unique trip itinerary
AFC = Average daily fuel consumption in metric tonnes
yatsea = Calendar year main and auxiliary engines are operated at sea
yatport = Calendar year main and auxiliary engines are operated in port
2-32
-------
Equations 2-10
AFCtrade,route,yatsea = £ r (Percent of trade along route) v s [Fleet AFCVS x (MELF+AE at sea LF)]
AFCtrade,route,yatPort = £ r (Percent of trade along route) vs [Fleet AFCVS x AE import LF]
DaysatSeatraderout = 2 (Percent of trade along route) vs|~Days at sea per voyagevsx Number of voyagesi
Days at Port traderoutey = 2 (P ercent of tradealong route) vs [Days at port per voyage x Number of voyages]
Where:
- AFC = Average daily fuel consumption in metric tones
trade = Commodity type
- route = Unique trip itinerary
- yatsea = Calendar year main and auxiliary engines are operated at sea
- yatport = Calendar year main and auxiliary engines are operated in port
- y = calendar year
- v = Vessel type
- s = Vessel size category
- t = Trade
- r = Route
Fleet AFC = Average daily fuel consumption in metric tonnes at full engine load
- MELF = main engine load factor, unitless
- AE at sea LF = auxiliary engine at-sea load factor, unitless
- AE in port LF = auxiliary engine in-port load factor, unitless
The inputs for these last four equations are all derived from the vessel analysis in Section
2.4.2.3 and the trade analysis in Section 2.4.2.2.
2.4.2.6 Worldwide Bunker Fuel Consumption
Based on the methodology outlined above, estimates of global fuel consumption over time
were computed, and growth rates determined from these projections.
2-33
-------
H Container 01 General Cargo D Dry Bulk 0 Crude Oil
D Chemicals D Petroleum • Natural Gas D Other
• Fishing Vessels • Passenger Ships D Military Vessels
Figure 2-6 Worldwide Bunker Fuel Consumption
Figure 2-6 shows estimated world-wide bunker fuel consumption by vessel type. Figure 2-7
shows the annual growth rates by vessel-type/cargo that are used in the projections shown in Figure
2-6. Total annual growth is generally between 2.5 percent and 3.5 percent over the time period
between 2006 and 2020 and generally declines over time, resulting in an average annual growth of
around 2.6 percent.
2-34
-------
10%
-O- Total -•- Container —*— General Cargo -•- Dry Bulk
-*- Crude Oil -X-Chemicals -*- Petroleum -•-Natural Gas
"~ Other Fishing Vessels Passenger Ships Military Vessels
Figure 2-7 Annual Growth Rate in World-Wide Bunker Fuel Use by Commodity Type
2.4.2.7 Fuel Demand Used to Import and Export Cargo for the United States
The methodology described above provides an estimate of fuel consumption for
international cargo worldwide. RTI also estimated the subset of fuel demand for cargo imported to
and exported from five regions of the U.S. The five regions are:
• North Pacific
• South Pacific
• Gulf
• East Coast
• Great Lakes
For this analysis, the same equations were used, but were limited to routes that carried cargo
between specific cities in Asia, Europe and Middle East to the various ports in the specific regions
of the U.S.
The trip distances for non-container vessel types were developed from information from
Worldscale Association and Maritime Chain.35'36 The data from Worldscale is considered to be the
industry standard for measuring port-to-port distances, particularly for tanker traffic. The reported
distances account for common routes through channels, canals, or straits. This distance information
was supplemented by data from Maritime Chain, a web service that provides port-to-port distances
along with some information about which channels, canals, or straits must be passed on the voyage.
2-35
-------
Voyage distances for container vessels are based on information from Containerization
International Yearbook (CIY)37 and calculations by RTI. That reference provides voyage
information for all major container services. Based on the frequency of the service, number of
vessels assigned to that service, and the number of days in operation per year, RTI estimated the
average length of voyages for the particular bilateral trade routes in the Global Insights trade
forecasts.
The distance information developed above was combined with the vessel speeds previously
shown in Table 2-19 to find the length of a voyage in days. Table 2-20 presents the day lengths for
non-containerized vessel types and Table 2-21 shows the same information for container vessels.
Table 2-20 Day Length for Voyages for Non-Container Cargo Ship (approximate average)
Global Insights Trade Regions
Africa East-South
Africa North-Mediterranean
Africa West
Australia-New Zealand
Canada East
Canada West
Caspian Region
China
Europe Eastern
Europe Western-North
Europe Western-South
Greater Caribbean
Japan
Middle East Gulf
Pacific High Growth
Rest of Asia
Russia-FSU
Rest of South America
Days per Voyage
US South
Pacific
68
49
56
48
37
11
95
41
61
53
54
26
35
77
52
68
64
51
US North
Pacific
75
56
63
47
46
5
89
36
68
60
61
33
31
72
48
64
71
30
US East
Coast
57
37
36
65
7
40
41
73
38
24
30
16
65
56
67
66
38
41
US Great
Lakes
62
43
46
81
18
58
46
87
45
32
37
29
81
65
76
64
46
46
US Gulf
54
47
43
63
19
39
48
69
46
34
37
17
62
83
88
73
48
44
Table 2-21 Day Length for Voyages for Container-Ship Trade Routes
Origin - Destination Regions
Asia - North America (Pacific)
Europe - North America (Atlantic)
Mediterranean - North America
Australia/New Zealand - North America
South America - North America
Africa South - North America (Atlantic)
Africa West - North America (Atlantic)
Days per
Voyage
37
37
41
61
48
54
43
2-36
-------
Origin - Destination Regions
Asia - North America (Atlantic)
Europe - North America (Pacific)
Africa South - North America (Pacific)
Africa West - North America (Pacific)
Caspian Region - North America (Atlantic)
Caspian Region - North America (Pacific)
Middle East/Gulf Region - North America (Atlantic)
Middle East/Gulf Region - North America (Pacific)
Days per
Voyage
68
64
68
38
42
38
63
80
2.4.2.8 Bunker Fuel Consumption for the United States
Figure 2-8 and Figure 2-9 present the estimates of fuel use for delivering trade goods to and
from the U.S. The results in Figure 2-8 show estimated historical bunker fuel use in year 2001 of
around 47 million tonnes (note: while this fuel is used to carry trade goods to and from the U.S., it is
not necessarily all purchased in the U.S. and is not all burned in U.S. waters). This amount grows
to over 90 million tonnes by 2020 with the most growth occurring on trade routes from the East
Coast and the "South Pacific" region of the West Coast.
H US North Pacific IB US Great Lakes D US Gulf H US East Coast 1 US South Pacific
Figure 2-8 Bunker Fuel Used to Import and Export Cargo by Region of the United States
2-37
-------
Figure 2-9 shows the estimated annual growth rates for the fuel consumption that are used in
the projections shown in Figure 2-8. Overall, the average annual growth rate in marine bunkers
associated with future U.S. trade flows is 3.4 percent between 2005 and 2020.
10%
e
s
a
eu
-o-United States
•*• US Great Lakes
• US South Pacific
• US Gulf
•US North Pacific
• US East Coast
Figure 2-9 Annual Growth Rates for Bunker Fuel Used to Import and Export Cargo by Region of the
United States
2.4.2.9 2020 Growth Factors for Nine Geographic Regions
The results of the RTI analysis described above are used to develop the growth factors that
are necessary to project the 2002 base year emissions inventory to 2020. The next two sections
describe how the five RTI regions were associated with the nine regions analyzed in this report, and
how the specific growth rates for each of the nine regions were developed.
2.4.2.9.1 Mapping the RTI Regional Results to the Nine Region Analysis
The nine geographic regions analyzed in this study were designed to be consistent with the
five RTI regional modeling domains. More specifically, four of the nine geographic areas in this
study, i.e., Alaska East, Alaska West, Hawaii East, and Hawaii West are actually subsets of two
broader regional areas that were analyzed by RTI, i.e., the North Pacific for both Alaska regions and
South Pacific for Hawaii. Therefore, the growth rate information from the related larger region was
assumed to be representative for that state.
2-38
-------
The nine geographic regions represented in the emission inventory study are presented in
Figure 2-1. The association of the RTI regions to the emission inventory regions is shown in Table
2-22.
Table 2-22 Association of the RTI Regions to the Nine Emission Inventory
Regions
Consumption Region
North Pacific
North Pacific
North Pacific
South Pacific
South Pacific
South Pacific
Gulf
East Coast
Great Lakes
Corresponding Emission
Inventory Region
North Pacific (NP)
Alaska East (AE)
Alaska West (AW)
South Pacific (SP)
Hawaii East (HE)
Hawaii West (HW)
Gulf Coast (GC)
East Coast (EC)
Great Lakes (GL)
2.4.2.9.2 Growth Factors for the Emission Inventory Analysis
Emission inventories for 2020 are estimated by multiplying the 2002 baseline inventory for
each region by a corresponding growth factor that was developed from the RTI regional results.
Specifically, the average annual growth rate from 2002-2020 was calculated for each of the five
regions. Each regional growth rate was then compounded over the inventory projection time period
for 2020, i.e., 18 years. The resulting multiplicative growth factors for each emission inventory
region and the associated RTI average annual growth rates are presented in Table 2-23 for 2020.
2-39
-------
Table 2-23 Regional Emission Inventory Growth Factors for 2020
Emission
Inventory Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
2002-2020 Average
Annualized Growth Rate (%)
3.3
3.3
4.5
2.9
5.0
5.0
3.3
5.0
1.7
Multiplicative Growth
Factor Relative to 2002
1.79
1.79
2.21
1.67
2.41
2.41
1.79
2.41
1.35
2.4.3 Emission Controls in 2020 Baseline and Control Scenarios
This section describes the control programs present in the 2020 baseline and control
scenarios. Section 2.4.4 describes the process of incorporating these programs into the 2020
emission inventories.
The baseline scenario includes the International Marine Organization's (IMO) Tier I NOx
standard for marine diesel engines that became effective in 2000, as well as the Tier II standard that
will become effective in 2011. Also included in the baseline inventories is the NOx retrofit
program for pre-controlled engines proposed by IMO.
Although the 0.1% fuel sulfur requirement goes into place for all vessels operating in EGAs
beginning in 2015, the use of 2020 as the analytic year will still provide a representative scenario
for the impact of the 0.1% fuel sulfur requirement on human health and the environment. This is
because the fuel requirements of the EGA go into effect all at once; there is no phase-in. So the
impacts of the fuel requirement in 2020 are expected to be the same as in 2015, with a small
increase due to growth. With regard to the NOx impacts, while 2020 will include five years of
turnover to the Tier III standards, the long service lives of engines on ocean-going vessels mean that
these impacts will be small and affect less than 25% of the total fleet, assuming an average 20-year
service life. These NOx reductions would not inflate the benefits of the program by very much, if
any. Note that the global fuel sulfur standard does not go into effect until 2020. We did not include
this in the 2020 analysis, to provide a better estimate of benefits in the early (pre-2020) years of the
program
The effects of these controls are reflected in the 2020 emission inventories by applying
appropriate adjustment factors that reflect the percentage of the vessel fleet in those years that are
estimated to comply with the controls. Adjustment factors are ratios of 2020 to 2002 calendar year
(CY) emission factors (EFs). Adjustment factors are derived separately by engine type for
propulsion and auxiliary engines. The adjustment factors for propulsion engines are applied to the
propulsion portion of the port inventory and the interport portion of the inventory. The adjustment
factors for auxiliary engines are applied to the auxiliary portion of the port inventory.
The control scenario includes an Emission Control Area (EGA) within a distance of 200
nautical miles (nm) from shore. Outside this distance, baseline controls were applied (i.e., the Tier I
2-40
-------
and Tier II NOx standards, the NOx retrofit program, and current fuel sulfur content levels). The
EGA NOx controls include the baseline controls above, plus Tier II NOx standards. Fuel sulfur
content is also assumed to be controlled to 1,000 ppm within the EGA. Note that gas and steam
turbine engines are not subject to any of the NOX standards; however, these engines are not a large
part of the inventory.
The retrofit program for Tier 0 (pre-control) engines was modeled as 11 percent control
from Tier 0 for 80 percent of 1990 thru 1999 model year (MY) engines greater than 90 liters per
cylinder (L/cyl) starting in 2011. The retrofit program was also modeled with a five year phase-in.
The current Tier I controls, which also are modeled as achieving an 11 percent reduction from Tier
0, apply to the 2000 thru 2010 MY engines. In 2011 thru 2015, Tier II controls are applied. Tier II
controls are modeled as a 2.5 g/kW-hr reduction from Tier I. In the EGA area only, for 2016 MY
engines and beyond, Tier III controls are applied. Tier III controls are modeled as achieving an 80
percent reduction from Tier I levels. Control of fuel sulfur content within the EGA area affects S02
and PM emissions.
EGA controls were applied to the 48 state region as well as Alaska East (AE) and Hawaii
East (HE). Alaska West (AW) and Hawaii West (HW) are baseline cases only.
2.4.4 2020 Emission Factors
This section describes the emission factors that are used in the 2020 scenarios. HC and CO
emission factors are assumed to remain unchanged from the 2002 scenario. NOX and fuel sulfur
controls are anticipated to lower NOx, SOz and PM emission factors. The switch to lower sulfur
distillate fuel use is also assumed to lower COz emissions slightly.
The NOx emission factors (EFs) by engine/ship type and tier are provided in Table 2-24.
Tier 0 refers to pre-control. There are separate entries for Tier 0/1/2 base and Tier 0/1/2 control,
since the control engines would be using distillate fuel, and there are small NOx emission
reductions assumed when switching from residual to distillate fuel.17 The NOx control EFs by tier
were derived using the assumptions described in section 2.4.3.
Table 2-24 Modeled NOX Emission Factors by Tier
Engine/
Ship Type
Main
SSD
MSD
ST
GT
Aux
Pass
Other
NOX EF (g/kW-hr)
Baseline
TierO
18.1
14
2.1
6.1
14.6
14.5
TO
retrofit
16.1
12.5
n/a
n/a
n/aa
n/aa
Tier I
16.1
12.5
n/a
n/a
13.0
12.9
Tier
II
13.6
10.0
n/a
n/a
10.5
10.4
Control Areas
TierO
17
13.2
2
5.7
14.6
14.5
TO
retrofit
15.1
11.7
n/a
n/a
n/aa
n/aa
Tier I
15.1
11.7
n/a
n/a
13.0
12.9
Tier II
12.6
9.2
n/a
n/a
10.5
10.4
Tier
III
3
2.3
n/a
n/a
2.6
2.6
a The retrofit program applies to engines over 90 L/cyl; auxiliary engines are smaller than
this cutpoint and would therefore not be subject to the program.
2-41
-------
The NOx EFs by tier were then used with the vessel age distributions (Table 2-25 & Table
2-26) to generate calendar year NOx EFs by engine/ship type for the base and control areas included
in the scenarios. These calendar year NOx EFs are provided in Table 2-27 below. Since the age
distributions are different for vessels in the Great Lakes, NOX EFs were determined separately for
the Great Lakes.
Table 2-25 Vessel Age Distribution for Deep Sea Ports by Engine Type
Age Group
(years old)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35+
Propulsion Engine Type a (Fraction of Total)
MSD
0.00570
0.07693
0.10202
0.08456
0.08590
0.06427
0.06024
0.07867
0.06730
0.04181
0.04106
0.03100
0.04527
0.03583
0.03519
0.02921
0.00089
0.01326
0.00847
0.00805
0.00566
0.00495
0.00503
0.00676
0.00539
0.01175
0.00803
0.00522
0.00294
0.00285
0.00254
0.00084
0.00023
0.00117
0.00132
0.01967
SSD
0.02667
0.07741
0.07512
0.07195
0.05504
0.05563
0.04042
0.07266
0.05763
0.04871
0.04777
0.03828
0.03888
0.02787
0.02824
0.01466
0.01660
0.01582
0.02414
0.01982
0.02258
0.02945
0.01883
0.01080
0.01091
0.01099
0.01045
0.00835
0.00788
0.00370
0.00106
0.00113
0.00367
0.00582
0.00092
0.00013
GT
0.00000
0.07189
0.14045
0.05608
0.67963
0.04165
0.00000
0.00626
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00034
0.00370
0.00000
0.00000
0.00000
0.00000
0.00000
ST
0.00447
0.12194
0.16464
0.05321
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.04873
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00875
0.00883
0.00883
0.18029
0.11065
0.01395
0.08657
0.02907
0.05126
0.00605
0.07105
0.00000
0.00000
0.03172
All
Auxiliary
Engines
0.01958
0.07670
0.08426
0.07489
0.07831
0.05685
0.04455
0.07150
0.05764
0.04475
0.04364
0.03538
0.04160
0.02909
0.02935
0.01869
0.01189
0.01462
0.01966
0.01550
0.01756
0.02260
0.01467
0.00943
0.00900
0.01224
0.01130
0.00738
0.00659
0.00349
0.00193
0.00096
0.00322
0.00419
0.00098
0.00598
a MSD is medium speed diesel, SSD is slow speed diesel, GT is gas turbine, ST is steam
turbine.
2-42
-------
Table 2-26 Vessel Age Distribution for Great Lake Ports by Engine Type
Age Group
(years old)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35+
Propulsion Engine Type" (Fraction of Total)
MSB
0.01610
0.02097
0.01370
0.02695
0.01571
0.04584
0.01494
0.01327
0.00099
0.00027
0.01085
0.00553
0.00739
0.02289
0.00000
0.00275
0.00069
0.00000
0.00342
0.00219
0.00867
0.00000
0.03375
0.04270
0.08161
0.02935
0.18511
0.01870
0.13815
0.05487
0.00000
0.03986
0.03654
0.03358
0.00295
0.06974
SSD
0.03913
0.03489
0.04644
0.03040
0.04547
0.01498
0.02180
0.01857
0.04842
0.03376
0.01177
0.01183
0.00546
0.02557
0.00286
0.00510
0.00073
0.00104
0.01967
0.01220
0.06140
0.05638
0.02108
0.02051
0.01010
0.05217
0.00522
0.00389
0.01438
0.01160
0.00114
0.00000
0.00282
0.00000
0.00123
0.30796
ST
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
1.00000
All Auxiliary
Engines
0.02399
0.02243
0.02544
0.02511
0.02497
0.02442
0.01528
0.01391
0.02107
0.01454
0.01076
0.00782
0.00626
0.02242
0.00121
0.00361
0.00078
0.00041
0.01059
0.00645
0.03034
0.02503
0.02279
0.02606
0.03744
0.03480
0.07701
0.01083
0.06181
0.02697
0.00047
0.01611
0.01631
0.01358
0.00165
0.31734
a MSD is medium speed diesel, SSD is slow speed diesel, GT is gas turbine, ST is steam
turbine.
b Fleet average weighted by installed power (ship port calls x main propulsion engine
power).
2-43
-------
Table 2-27 Modeled NOX Emission Factors by Calendar Year and Control Type
Engine/
Ship
Type
Main
SSD
MSD
ST
GT
Aux
Pass
Other
CY NOX EF (g/kW-hr)
2002
18.1
14
2.1
6.1
14.6
14.5
2020 Base
DSP
14.7
10.9
2.1
6.1
11.7
11.5
GL
15.9
13.1
2.1
n/a
13.6
13.4
2020 ECA Control
DSP
10.8
7.7
2.0
5.7
8.6
8.6
GL
13.1
11.8
2.0
n/a
12.0
12.0
DSP = Deep water ports and areas other than the Great Lakes
GL = Great Lakes
The PM and S02 EFs are a function of fuel sulfur level. For the baseline portions of the
inventory, there are two residual fuel sulfur levels modeled: 25,000 ppm for the West Coast and
27,000 ppm for the rest of the U.S. The baseline distillate fuel sulfur level assumed for all areas is
15,000 ppm. As discussed in section 2.3.2.3.5, for the baseline, main engines use residual fuel and
auxiliary engines use a mix of residual and distillate fuel. For the control areas, there is one level of
distillate fuel sulfur assumed to be used by all engines: 1,000 ppm for the ECA control areas.
Table 2-28 provides the PMio EFs by engine/ship type and fuel sulfur level. For modeling
purposes, PM2.s is assumed to be 92 percent of PMi0. The PM EFs are adjusted to reflect the
appropriate fuel sulfur levels using Equation 2-2.
Table 2-29 provides the modeled S02 EFs. S02 emission reductions are directly
proportional to reductions in fuel sulfur content.
C02 is directly proportional to fuel consumed. Table 2-30 provides the modeled C02 and
brake specific fuel consumption (BSFC) EFs. Due to the higher energy content of distillate fuel on
a mass basis, the switch to distillate fuel for the control areas results in a small reduction to BSFC
and, correspondingly, C02 emissions.17
2-44
-------
Table 2-28 Modeled PM10 Emission Factors
Engine/
Ship Type
Main
SSD
MSD
ST
GT
Aux
Pass
Other
PM10 EF (g/kW-hr)
Baseline
Other than West Coast
27,000 ppm S
1.40
1.40
1.50
1.50
1.40
1.20
West Coast"
25,000 ppm S
1.40
1.40
1.40
1.40
1.30
1.10
Control Areas
ECA
1,000 ppm S
0.19
0.19
0.17
0.17
0.19
0.19
a For the base cases, the West Coast fuel is assumed to be used in the following
regions: Alaska East (AE), Alaska West (AW), Hawaii East (HE), Hawaii West
(HW), North Pacific (NP), and South Pacific (SP).
Table 2-29 Modeled SO2 Emission Factors*
Engine/
Ship Type
Main
SSD
MSD
ST
GT
Aux
Pass
Other
SO2 EF (g/kW-hr)
Baseline
Other than West Coast
27,000 ppm S
10.29
11.09
16.10
16.10
10.70
9.66
West Coast3
25,000 ppm S
9.53
10.26
14.91
14.91
9.93
9.07
Control Areas
ECA Control
1,000 ppm S
0.36
0.39
0.57
0.57
0.39
0.39
a For the base cases, the West Coast fuel is assumed to be used in the following
regions: Alaska East (AE), Alaska West (AW), Hawaii East (HE), Hawaii West
(HW), North Pacific (NP), and South Pacific (SP).
Table 2-30 Modeled Fuel Consumption and CO2 Emission Factors
Engine/
Ship Type
Main
SSD
MSD
ST
GT
Aux
Pass
Other
EF (g/kW-hr)
Baseline
BSFC
195
210
305
305
210
210
C02
620
668
970
970
668
668
Control Areas
BSFC
185
200
290
290
200
200
C02
589
637
923
923
636
636
2-45
-------
2.4.5 Calculation of 2020 Near Port and Interport Inventories
Based on the emission factors described in Section 2.4.4, appropriate adjustments were
applied to the NOX, PM (PMio and PM2.5), S02, and C02 inventory of each 2020 scenario. This
section describes the development and application of the adjustment factors to the port and interport
inventories, and the methodology for combining the port and interport portions.
2.4.5.1 Port Methodology
2.4.5.1.1 Non-California Ports
For the non-California ports, 2002 emissions for each port are summed by engine/ship type.
Propulsion and auxiliary emissions are summed separately, since the EF adjustment factors differ.
The appropriate regional growth factor, as provided in Table 2-23, is then applied, along with EF
adjustment factors by engine/ship type. The EF adjustment factors are a ratio of the control EF to
the 2002 EF. Table 2-31 through Table 2-35 provide the EF adjustment factors for each pollutant
and control area. The ports will be subject to EGA controls in the control scenario. These tables are
also used as input for the California ports and interport control inventory development, discussed in
subsequent sections.
Table 2-31 NOX EF Adjustment Factors by Engine/Ship Type and Control Type"
Engine/ Ship
Type
Main
SSD
MSD
ST
GT
Aux
Pass
Other
2020 Base
DSP
0.8130
0.7804
1.0000
1.0000
0.7985
0.7972
GL
0.8783
0.9366
1.0000
n/a
0.9296
0.9292
2020 ECA
Control
DSP
0.5967
0.5515
0.9524
0.9344
0.5869
0.5940
GL
0.7219
0.8423
0.9524
n/a
0.8196
0.8295
a NOX adjustment factors are a ratio of future base or control EFs to 2002 EFs
DSP = deep water ports and areas other than the Great Lakes; GL = Great Lakes
2-46
-------
Table 2-32 PM10 EF Adjustment Factors by Engine/Ship Type and Control Type"
Engine/ Ship
Type
Main
SSD
MSD
ST
GT
Aux
Pass
Other
2020 Base
Other
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
we
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
2020 ECA Control
Other
0.1352
0.1328
0.1108
0.1108
0.1328
0.1550
we
0.1352
0.1328
0.1187
0.1187
0.1430
0.1691
PMio adjustment factors are a ratio of the control EFs to the 2002 EFs.
PM is not adjusted for the future baseline because fuel sulfur levels are
only assumed to change within the ECA.
Other = Other than West Coast
WC = Ports/areas within the West Coast. This includes the regions of
Alaska, Hawaii, North Pacific, and South Pacific.
Table 2-33 PM2.5 EF Adjustment Factors by Engine/Ship Type and Control Type3
Engine/ Ship
Type
Main
SSD
MSD
ST
GT
Aux
Pass
Other
2020 Base
Other
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
WC
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
2020 ECA Control
Other
0.1339
0.1316
0.1092
0.1092
0.1316
0.1555
WC
0.1339
0.1316
0.1176
0.1176
0.1426
0.1711
a PM2.5 adjustment factors are a ratio of the control EFs to the 2002 EFs.
PM is not adjusted for the future baseline because fuel sulfur levels are
only assumed to change within the ECA. The PM2.5 adjustment factors are
slightly different from those for PMi0 due to rounding.
Other = Other than West Coast
WC = Ports/areas within the West Coast. This includes the regions of
Alaska, Hawaii, North Pacific, and South Pacific.
2-47
-------
Table 2-34 SO2 EF Adjustment Factors by Engine/Ship Type and Control Type"
Engine/ Ship
Type
Main
SSD
MSD
ST
GT
Aux
Pass
Other
2020 Base
Other
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
we
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
2020 ECA Control
Other
0.0351
0.0353
0.0352
0.0352
0.0365
0.0405
we
0.0380
0.0381
0.0380
0.0380
0.0394
0.0431
a S02 adjustment factors are a ratio of the control EFs to the 2002
EFs. S02 is not adjusted for the future baseline because fuel sulfur
levels are only assumed to change within the ECA.
Other = Other than West Coast
WC = Ports/areas within the West Coast. This includes the regions
of Alaska, Hawaii, North Pacific, and South Pacific.
Table 2-35 CO2 EF Adjustment Factors by Engine/Ship Type and Control Type"
Engine/ Ship
Type
Main
SSD
MSD
ST
GT
Aux
Pass
Other
2020 Base
Other
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
WC
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
2020 ECA Control
Other
0.9488
0.9531
0.9509
0.9509
0.9525
0.9525
WC
0.9488
0.9531
0.9509
0.9509
0.9593
0.9683
a C02 adjustment factors are a ratio of the control EFs to the 2002
EFs. C02 is not adjusted for the future baseline because fuel
consumption (BSFC) is only assumed to change within the ECA.
Other = Other than West Coast
WC = Ports/areas within the West Coast. This includes the regions
of Alaska, Hawaii, North Pacific, and South Pacific.
2.4.5.1.2 California Ports
For the California ports, 2002 emissions for each port are summed by ship type. Propulsion
and auxiliary emissions are summed separately, since the EF adjustment factors differ. The EF
adjustment factors by engine/ship type, provided in the previous section, are consolidated by ship
type, using the CARB assumption that engines on all ships except passenger ships are 95 percent
slow speed diesel (SSD) engines and 5 percent medium speed diesel engines (MSD) based upon a
2005 ARE survey.0 All passenger ships were assumed to be medium speed diesel engines with
electric drive propulsion (MSD-ED). Steam turbines (ST) and gas-turbines (GT) are not included in
' California Air Resources Board, 2005 Oceangoing Ship Survey, Summary of Results, September 2005.
2-48
-------
the CARB inventory. The EF adjustment factors by ship type are then applied, along with ship-
specific growth factors supplied by CARB. The ship-specific growth factors relative to 2002 are
provided in Table 2-36 below.
Table 2-36 Growth Factors by Ship Type for California Ports Relative to 2002
Ship Type
Auto
Bulk
Container
General
Passenger
Reefer
RoRo
Tanker
Calendar Year
2002
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
2020
1.5010
0.2918
2.5861
0.7331
7.5764
1.0339
1.5010
2.0979
2.4.5.2 Interport Methodology
The interport portion of the inventory is not segregated by engine or ship type. As a result,
regional EF adjustment factors were developed based on the assumed mix of main (propulsion)
engine types in each region. The mix of main engine types by region was developed using the ship
call data and is presented in Table 2-37. Main engines are considered a good surrogate for interport
emissions, since the majority of emissions while underway are due to the main engines. The EF
adjustment factors by main engine type in Section 2.4.5.1were used together with the mix of main
engine types by region to develop the EF regional adjustment factors for each control area. The
resulting EF regional adjustment factors for each pollutant and control area are provided in Table
2-38 through Table 2-42 below. These EF regional adjustment factors, together with the regional
growth factors in Table 2-23, were applied to calculate the future inventories for each control area.
Table 2-37 Installed Power by Main Engine Type and Region
Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
2020 Installed Power (%)
MSB
19.1%
19.1%
25.6%
13.7%
66.2%
66.2%
5.1%
29.2%
48%
SSD
18.4%
18.4%
72.5%
85.5%
18.5%
18.5%
83.5%
70.8%
44%
GT
0.3%
0.3%
0.9%
0.0%
7.4%
7.4%
1.6%
0.0%
0%
ST
62.2%
62.2%
1.0%
0.8%
8.0%
8.0%
9.7%
0.0%
8%
Total
100%
100%
100%
100%
100%
100%
100%
100%
100%
2-49
-------
Table 2-38 NOX EF Adjustment Factors by Region and Control Type"
U.S. Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
OutofRegionb
2002
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
2020
Base
0.9237
0.9237
0.8082
0.8102
0.8202
0.8202
0.8325
0.8036
0.8131
0.8095
ECA Control
0.8104
n/a
0.5917
0.5935
0.6201
n/a
0.6343
0.5837
0.7989
n/a
aNOx adjustment factors are a ratio of future base or control EFs to
2002 EFs. These regional adjustment factors are used to adjust the
interport portion of the 2002 inventory.
b Out of Region refers to areas outside the 200 nm US modeling
boundary, but within the air quality modeling domain. The out of
region adjustment factors are derived by weighting the regional
adjustment factors by the main propulsion power in each region.
Table 2-39 PM10 EF Adjustment Factors by Region and Control Type3
U.S. Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
Out of Regionb
2002
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
2020
Base
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
ECA Control
0.1244
n/a
0.1341
0.1347
0.1311
n/a
0.1332
0.1345
0.1320
n/a
a PMio adjustment factors are a ratio of future base or control EFs to
2002 EFs. These regional adjustment factors are used to adjust the
interport portion of the 2002 inventory.
b Out of Region refers to areas outside the 200 nm US modeling
boundary, but within the air quality modeling domain. The out of
region adjustment factors are derived by weighting the regional
adjustment factors by the main propulsion power in each region.
2-50
-------
Table 2-40 PM2.5 EF Adjustment Factors by Region and Control Type"
U.S. Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
Out of Regionb
2002
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
2020
Base
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
ECA Control
0.1233
n/a
0.1329
0.1334
0.1299
n/a
0.1320
0.1332
0.1307
n/a
a PM2.5 adjustment factors are a ratio of future base or control EFs
to 2002 EFs. These regional adjustment factors are used to adjust
the interport portion of the 2002 inventory.
b Out of Region refers to areas outside the 200 nm US modeling
boundary, but within the air quality modeling domain. The out of
region adjustment factors are derived by weighting the regional
adjustment factors by the main propulsion power in each region.
Table 2-41 SO2 EF Adjustment Factors by Region and Control Type3
U.S. Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
Out of Regionb
2002
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
2020
Base
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
ECA Control
0.0380
n/a
0.0352
0.0352
0.0381
n/a
0.0380
0.0380
0.0352
n/a
a S02 adjustment factors are a ratio of future base or control EFs to
2002 EFs. These regional adjustment factors are used to adjust the
interport portion of the 2002 inventory.
b Out of Region refers to areas outside the 200 nm US modeling
boundary, but within the air quality modeling domain. The out of
region adjustment factors are derived by weighting the regional
adjustment factors by the main propulsion power in each region are
derived by weighting the regional adjustment factors by the main
propulsion power in each region.
2-51
-------
Table 2-42 CO2 EF Adjustment Factors by Region and Control Type"
U.S. Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
Out of Region15
2002
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
2020
Base
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
ECA Control
0.9509
n/a
0.9499
0.9494
0.9519
n/a
0.9493
0.9501
0.9510
n/a
C02 adjustment factors are a ratio of future base or control EFs
to 2002 EFs. These regional adjustment factors are used to adjust
the interport portion of the 2002 inventory.
Out of Region refers to areas outside the 200 nm US modeling
boundary, but within the air quality modeling domain. The out of
region adjustment factors are derived by weighting the regional
adjustment factors by the main propulsion power in each region.
2.4.5.3 Estimating and Combining the Near Port and Interport Control Inventories
To produce future year control scenarios, the interport inventories were scaled by a growth
factor to 2020, as previously described. An ECA boundary line was drawn so that each point on it
was at a 200 nm distance from the nearest point on land. Adjustment factors, as described in
section 2.4.4, were then applied to interport emissions within the ECA boundary.
To create control scenarios in the near port inventories, growth and control factors were
applied to the 2002 near port inventories (described in sections 2.4.2 and 2.4.4). The near port
inventories were then converted into a gridded format (section 2.3.4). Using this grid, STEEM
values were removed from near port cells and near port emissions were used as replacement values.
In cases where the emissions near ports were only partially attributable to port traffic, the STEEM
inventory was reduced rather than removed.
Interport and near port emissions were then aggregated to form regional totals.
2.4.6 2020 Baseline and Control Inventories and Fuel Consumption
The baseline emission inventories for 2020 are presented in Table 2-43.
2-52
-------
Table 2-43 2020 Baseline Inventory
U.S. Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
Total U.S. Metric Tonnes
Metric Tonnes per Year
NOx
27,982
89,826
391,995
232,114
42,935
60,409
38,051
208,294
18,768
1,110,375
PM10
2,561
8,118
39,003
23,403
4,185
5,888
3,916
20,148
1,613
108,835
PM2.5a
2,356
7,469
35,882
21,531
3,850
5,417
3,603
18,536
1,484
100,128
HC
1,073
3,444
16,216
9,590
1,765
2,483
1,706
8,585
681.914
45,544
CO
2,534
8,112
38,382
23,628
4,161
5,855
3,799
20,686
1,607
108,762
S02
19,084
60,227
323,038
174,751
31,075
43,722
27,807
149,751
11,993
841,447
C02
1,182,047
3,711,596
18,121,202
10,567,512
1,930,172
2,715,741
1,800,743
9,490,502
740,624
50,260,140
' Estimated from PMi0 using a multiplicative conversion factor of 0.92.
The EGA control case inventories for each of the nine geographic regions and the U.S.
domain total are presented in Table 2-44. The regional and total inventories include all emissions
within the 200 nm US modeling domain. Controls are applied to all regions included in the
proposed EGA.
Table 2-44 Category 3 Vessel Inventories for 2020 Proposed ECA Control Case"
U.S. Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
Total U.S. Metric Tonnes
Metric Tonnes per Year
NOx
25,978
89,826
289,671
170,861
32,952
60,409
29,105
150,461
16,420
865,684
PM10
322
8,118
5,286
3,201
551
5,888
539
2,753
207
26,864
PM2.5a
296
7,469
4,863
2,945
507
5,417
496
2,533
190
24,715
HC
1,073
3,444
16,216
9,590
1,765
2,483
1,706
8,585
681
45,544
CO
2,534
8,112
38,382
23,628
4,161
5,855
3,799
20,686
1,607
108,762
S02
728
60,227
11,514
6,255
1,187
43,722
1,076
5,786
420
130,914
C02
1,124,652
3,711,596
17,233,800
10,034,946
1,838,832
2,715,741
1,715,210
9,009,986
704,390
48,089,152
a This scenario assumes ECA controls apply within 200 nautical miles of all U.S. regions. Alaska
West and Hawaii West are not subject to ECA controls.
The fuel consumption by fuel type in the baseline and ECA cases is also presented in
Table 2-45.
2-53
-------
Table 2-45 Fuel Consumption by Category 3 Vessels in Baseline and ECA Scenarios.
U.S. Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
Total U.S. Metric
Tonnes
Baseline
Metric Tonnes Fuel
Distillate
3,386
0
202,139
96,428
10,529
0
28,532
83,576
1,269
425,860
Residual
367,977
1,166,068
5,490,981
3,223,557
595,871
853,202
537,206
2,898,045
231,412
15,364,319
Total
371,363
1,166,068
5,693,120
3,319,985
606,400
853,202
565,738
2,981,622
232,681
15,790,179
With ECA
Metric Tonnes Fuel
Distillate
353,331
0
5,414,326
3,152,669
577,704
0
538,866
2,830,658
221,297
13,088,852
Residual
0
1,166,068
0
0
0
853,202
0
0
0
2,019,270
Total
353,331
1,166,068
5,414,326
3,152,669
577,704
853,202
538,866
2,830,658
221,297
15,108,122
2.5 Projected Emission Reductions
The projected reduction (tonnes) for the 2020 control case relative to the 2020 baseline is
presented in Table 2-46. Reductions by region, for the total U.S., and for the total 48-states, are
provided by pollutant in each table.
Table 2-46 Reductions for 2020 Proposed ECA Control Case3
U.S. Region
Alaska East (AE)
Alaska West (AW)
East Coast (EC)
Gulf Coast (GC)
Hawaii East (HE)
Hawaii West (HW)
North Pacific (NP)
South Pacific (SP)
Great Lakes (GL)
Total U.S. Metric
Tonnes
Metric Tonnes per Year
NOX
2,004
0
102,324
61,253
9,983
0
8,946
57,833
2,348
244,690
PM10
2,239
0
33,717
20,202
3,634
0
3,377
17,395
1,406
81,971
PM25a
2,060
0
31,019
18,586
3,343
0
3,107
16,003
1,294
75,413
HC
0
0
0
0
0
0
0
0
0
0
CO
0
0
0
0
0
0
0
0
0
0
S02
18,356
0
311,524
168,496
29,888
0
26,731
143,965
11,573
710,534
C02
57,395
0
887,402
532,566
91,340
0
85,533
480,516
36,234
2,170,987
a The emission reductions are relative to the 2020 baseline.
2.6 Conclusion
An emission inventory for ships in the U.S. was developed based on the latest state of the art
models and inputs, using a "bottom-up" methodology. The inventory includes emissions for 117
ports, as well as emissions for ships while underway in U.S. waters. The analysis clearly
2-54
-------
demonstrates that emissions from ships in the proposed EGA are contributing to U.S. air pollution.
The inventory data were used as an input for the air quality modeling analysis.
2-55
-------
Appendices
Appendix 2A: Port Coordinates
Table 2A-1 Port Coordinates
Port Name
Albany, NY
Alpena, Ml
Anacortes, WA
Anchorage, AK
Ashtabula, OH
Baltimore, MD
Barbers Point, Oahu, HI
Baton Rouge, LA
Beaumont, TX
Boston, MA
Bridgeport, CT
Brownsville, TX
Brunswick, GA
Buffalo, NY
Burns Waterway Harbor, IN
Calcite, Ml
Camden-Gloucester, NJ
Carquinez, CA
Catalina, CA
Charleston, SC
Chester, PA
Chicago, IL
Cleveland, OH
Conneaut, OH
Coos Bay, OR
Corpus Christi, TX
Detroit, Ml
Duluth-Superior, MN and Wl
El Segundo, CA
Erie, PA
Escanaba, Ml
Eureka, CA
Everett, WA
Fairport Harbor, OH
Fall River, MA
Freeport, TX
Galveston, TX
Gary, IN
Georgetown, SC
Grays Harbor, WA
Gulfport, MS
Hilo, HI
USAGE
Code
C0505
L3617
C4730
C4820
L3219
C0700
C4458
C2252
C2395
C0149
C0311
C2420
C0780
L3230
L3739
L3620
C0551
CCA01
CCA02
C0773
C0297
L3749
L3217
L3220
C4660
C2423
L3321
L3924
CCA03
L3221
L3795
CCA04
C4725
L3218
C0189
C2408
C2417
L3736
C0772
C4702
C2083
C4400
Port Coordinates
Longitude
-73.7482
-83.4223
-122.6
-149.895
-80.7917
-76.5171
-158.109
-91.1993
-94.0881
-71.0523
-73.1789
-97.3981
-81.4999
-78.8953
-87.1552
-83.7756
-75.1043
-122.123
-118.496
-79.9216
-75.3222
-87.638
-81.6719
-80.5486
-124.21
-97.3979
-83.1096
-92.0964
-118.425
-80.0679
-87.025
-124.186
-122.229
-81.2941
-71.1588
-95.3304
-94.8127
-87.3251
-79.2896
-124.122
-89.0853
-155.076
Latitude
42.64271
45.0556
48.49617
61.23778
41.91873
39.20899
21.29723
30.42292
30.08716
42.35094
41.172
25.9522
31.15856
42.8783
41.64325
45.39293
39.94305
38.03556
33.43943
32.78878
39.85423
41.88662
41.47852
41.96671
43.36351
27.81277
42.26909
46.77836
33.91354
42.15154
45.73351
40.79528
47.98476
41.76666
41.72166
28.9384
29.31049
41.61202
33.36682
46.91167
30.35216
19.72861
2-56
-------
Port Name
Honolulu, HI
Hopewell, VA
Houston, TX
Indiana Harbor, IN
Jacksonville, FL
Kahului, Maui, HI
Kalama, WA
Lake Charles, LA
Long Beach, CA
Longview, WA
Lorain, OH
Los Angeles, CA
Manistee, Ml
Marblehead, OH
Marcus Hook, PA
Matagorda Ship Channel, TX
Miami, FL
Milwaukee, Wl
Mobile, AL
Morehead City, NC
Muskegon, Ml
Nawiliwili, Kauai, HI
New Bedford, MA
New Castle, DE
New Haven, CT
New Orleans, LA
New York, NY and NJ
Newport News, VA
Nikishka, AK
Oakland, CA
Olympia, WA
Other Puget Sound, WA
Palm Beach, FL
Panama City, FL
Pascagoula, MS
Paulsboro, NJ
Penn Manor, PA
Pensacola, FL
Philadelphia, PA
Plaquemines, LA, Port of
Port Angeles, WA
Port Arthur, TX
Port Canaveral, FL
Port Dolomite, Ml
Port Everglades, FL
Port Hueneme, CA
Port Inland, Ml
USAGE
Code
C4420
C0738
C2012
L3738
C2017
C4410
C4626
C2254
C4110
C4622
L3216
C4120
L3720
L3212
C5251
C2410
C2164
L3756
C2005
C0764
L3725
C4430
C0187
C0299
C1507
C2251
C0398
C0736
C4831
C4345
C4718
C4754
C2162
C2016
C2004
C5252
C0298
C2007
C0552
C2255
C4708
C2416
C2160
L3627
C2163
C4150
L3803
Port Coordinates
Longitude
-157.872
-77.2763
-95.2677
-87.4455
-81.6201
-156.473
-122.863
-93.2221
-118.21
-122.914
-82.1951
-118.241
-86.3443
-82.7091
-75.4042
-96.5641
-80.1832
-87.8997
-88.0411
-76.6947
-86.3501
-159.353
-70.9162
-75.5616
-72.9047
-90.0853
-74.0384
-76.4582
-151.314
-122.308
-122.909
-122.72
-80.0527
-84.1993
-88.5588
-75.2266
-74.7408
-87.2579
-75.2022
-89.6875
-123.453
-93.9607
-80.6082
-84.3128
-80.1178
-119.208
-85.8628
Latitude
21.31111
37.32231
29.72538
41.67586
30.34804
20.89861
46.02048
30.22358
33.73957
46.14222
41.48248
33.77728
44.25082
41.52962
39.81544
28.5954
25.78354
42.98824
30.72527
34.71669
43.19492
21.96111
41.63641
39.65668
41.29883
29.91414
40.67395
36.98522
60.74793
37.82152
47.06827
48.84099
26.76904
30.19009
30.34802
39.82689
40.13598
30.40785
39.91882
29.48
48.1305
29.83142
28.41409
45.99139
26.09339
34.14824
45.95508
2-57
-------
Port Name
Port Manatee, FL
Portland, ME
Portland, OR
Presque Isle, Ml
Providence, Rl
Redwood City, CA
Richmond, CA
Richmond, VA
Sacramento, CA
San Diego, CA
San Francisco, CA
Sandusky, OH
Savannah, GA
Searsport, ME
Seattle, WA
South Louisiana, LA, Port of
St. Clair, Ml
Stockton, CA
Stoneport, Ml
Tacoma, WA
Tampa, FL
Texas City, TX
Toledo, OH
Two Harbors, MN
Valdez, AK
Vancouver, WA
Wilmington, DE
Wilmington, NC
USAGE
Code
C2023
C0128
C4644
L3845
C0191
CCA05
C4350
C0737
CCA06
C4100
C4335
L3213
C0776
C0112
C4722
C2253
L3509
C4270
L3619
C4720
C2021
C2404
L3204
L3926
C4816
C4636
C0554
C0766
Port Coordinates
Longitude
-82.5613
-70.2513
-122.665
-87.3852
-71.3984
-122.21
-122.374
-77.4194
-121.544
-117.178
-122.399
-82.7123
-81.0954
-68.925
-122.359
-90.6179
-82.4941
-121.316
-83.4703
-122.452
-82.5224
-94.9181
-83.5075
-91.6626
-146.346
-122.681
-75.507
-77.954
Latitude
27.63376
43.64951
45.47881
46.57737
41.81178
37.51306
37.92424
37.45701
38.56167
32.70821
37.80667
41.47022
32.08471
44.45285
47.58771
30.03345
42.82663
37.9527
45.28073
47.28966
27.78534
29.36307
41.66294
47.00428
61.12473
45.62244
39.71589
34.23928
2-58
-------
Appendix 2B: Port Methodology and Equations
Near port emissions for each port are calculated for four modes of operation: 1) hotelling, 2)
maneuvering, 3) reduced speed zone (RSZ), and 4) cruise. Hotelling, or dwelling, occurs while the
vessel is docked or anchored near a dock, and only the auxiliary engine (s) are being used to provide
power to meet the ship's energy needs. Maneuvering occurs within a very short distance of the
docks. The RSZ varies from port to port, though generally the RSZ would begin and end when the
pilots board or disembark, and typically occurs when the near port shipping lanes reach
unconstrained ocean shipping lanes. The cruise mode emissions in the near ports analysis extend 25
nautical miles beyond the end of the RSZ lanes for deep water ports and 7 nautical miles for Great
Lake ports.
Emissions are calculated separately for propulsion and auxiliary engines. The basic
equation used is as follows:
Equation 2B-1
Emissions moAe[eng] = (calls) x (P[eng]) x (hrs I callmode) x (LFmode[eng]) x (EF[eng]) x (Adj) x (1(T6 tonnes I g)
Where:
Emissionsmode [eng] = Metric tonnes emitted by mode and engine type
Calls = Round-trip visits (i.e., one entrance and one clearance is considered a call)
P[eng] = Total engine power by engine type, in kilowatts
hrs/callmode = Hours per call by mode
LFmode [eng] = Load factor by mode and engine type (unitless)
EF[eng] = Emission factor by engine type for the pollutant of interest, in g/kW-hr
(these vary as a function of engine type and fuel used, rather than activity mode)
Adj = Low load adjustment factor, unitless (used when the load factor is below 0.20)
10 = Conversion factor from grams to metric tonnes
Main engine load factors are calculated directly from the propeller curve based upon the
cube of actual speed divided by maximum speed (at 100% maximum continuous rating [MCR]). In
addition, cruise mode activity is based on cruise distance and speed inputs. The following sections
provide the specific equations used to calculate propulsion and auxiliary emissions for each activity
mode.
Cruise
Cruise emissions are calculated for both propulsion (main) and auxiliary engines. The basic
equation used to calculate cruise mode emissions for the main engines is:
Equation 2B-2
Emissions cmise[mam] = (calls) x (P[mam]) x (hrs I call cruj x (LFcmise[mam]) x (EF[mam]) x (10~6 tonnes I g)
Where:
EmissionScruise [main] = Metric tonnes emitted from main engines in cruise mode
Calls = Round-trip visits (i.e., one entrance and one clearance is considered a call)
2-59
-------
P[main] = Total main engine power, in kilowatts
hrs/callcruise = Hours per call for cruise mode
LFcrUise [main] = Load factor for main engines in cruise mode (unitless)
EF[maln] = Emission factor for main engines for the pollutant of interest, in g/kW-hr (these vary
as a function of engine type and fuel used, rather than activity mode)
106 = Conversion factor from grams to metric tonnes
In addition, the time in cruise is calculated as follows:
Equation 2B-3
Hrs I callcmise = Cruise Distance [nmiles ]/Cruise Speed [knots ] x 2 trips I call
Where:
Cruise distance = one way distance (25 nautical miles for deep sea ports, and 7 nautical miles
for Great Lake ports)
Cruise speed = vessel service speed, in knots
2 trips/call = Used to calculate round trip cruise distance
Main engine load factors are calculated directly from the propeller curve based upon the
cube of actual speed divided by maximum speed (at 100% maximum continuous rating [MCR]):
Equation 2B-4
LoadFactor cmise [main] = (Cruise Speed [knots ] I Maximum Speed [knots ])3
Since cruise speed is estimated at 94 percent of maximum speed38, the load factor for main
engines at cruise is 0.83.
Substituting Equation 2B-3 for time in cruise into Equation 2B-2, and using the load factor
of 0.83, the equation used to calculate cruise mode emissions for the main engines becomes the
following:
Equation 2B-5 Cruise Mode Emissions for Main Engines
Emissionscrui,ain] = (calls) x (P,main]) x (CruiseDistance/CruiseSpeed) x (2 trips/call) x 0.83x (EF,in]) x (10~6 tonne
Where:
Emissionscruise [main] = Metric tonnes emitted from main engines in cruise mode
calls = Round-trip visits (i.e., one entrance and one clearance is considered a call)
P[mam] = Total main engine power, in kilowatts
Cruise distance = one way distance (25 nautical miles for deep sea ports, and 7 nautical miles
for Great Lake ports)
Cruise speed = vessel service speed, in knots
2 trips/call = Used to calculate round trip cruise distance
0.83 = Load factor for main engines in cruise mode, unitless
EF [main] = Emission factor for main engines for the pollutant of interest, in g/kW-hr (these vary
as a function of engine type and fuel used, rather than activity mode)
106 = Conversion factor from grams to metric tonnes
2-60
-------
The equation used to calculate cruise mode emissions for the auxiliary engines is:
Equation 2B-6 Cruise Mode Emissions for Auxiliary Engines
Emissions cruise[am:] = (calls) x (P[aux]) x (Cruise Distance/Cruise Speed) x (2trips/call) x (LFcruise[aux]) x (EF[aux]) x (1(T6 tonnes I
Where:
x] = Metric tonnes emitted from auxiliary engines in cruise mode
calls = Round-trip visits (i.e., one entrance and one clearance is considered a call)
P[aux] = Total auxiliary engine power, in kilowatts
Cruise distance = one way distance (25 nautical miles for deep sea ports, and 7 nautical miles
for Great Lake ports)
Cruise speed = vessel service speed, in knots
2 trips/call = Used to calculate round trip cruise distance
LFcmise [aux] = Load factor for auxiliary engines in cruise mode, unitless (these vary by ship type
and activity mode)
EF[aux] = Emission factor for auxiliary engines for the pollutant of interest, in g/kW-hr (these
vary as a function of engine type and fuel used, rather than activity mode)
106 = Conversion factor from grams to metric tonnes
The inputs of calls, cruise distance, and vessel speed are the same for main and auxiliary
engines. Relative to the main engines, auxiliary engines have separate inputs for engine power,
load factor, and emission factors. The activity-related inputs, such as engine power, vessel speed,
and calls, can be unique to each ship calling on a port, if ship-specific information is available. For
this analysis, these inputs were developed by port for bins that varied by ship type, engine type, and
dead weight tonnage (DWT) range.
Reduced Speed Zone
RSZ emissions are calculated for both propulsion (main) and auxiliary engines. The basic
equation used to calculate RSZ mode emissions for the main engines is:
Equation 2B-7
Emission$sz[mam] = (call§ x (P[mairf) x (hrsl' callRSZ] x (LFRSZ[mair}) x (EF[mairf) x (Adj) x (10~6 tonnes/ g)
Where:
EmissionsRszimain] = Metric tonnes emitted from main engines in RSZ mode
calls = Round-trip visits (i.e., one entrance and one clearance is considered a call)
P[mam] = Total main engine power, in kilowatts
hrs/callRSz = Hours per call for RSZ mode
LFRSZ [main] = Load factor for main engines in RSZ mode, unitless
EF[maln] = Emission factor for main engines for the pollutant of interest, in g/kW-hr (these vary
as a function of engine type and fuel used, rather than activity mode)
Adj = Low load adjustment factor, unitless (used when the load factor is below 0.20)
106 = Conversion factor from grams to metric tonnes
2-61
-------
In addition, the time in RSZ mode is calculated as follows:
Equation 2B-8
Hrs Icall RSZ = RSZ Distance [nmiles ]/ RSZ Speed [knots } x 2 trips I call
Load factor during the RSZ mode is calculated as follows:
Equation 2B-9
LoadFactorRSZ [main] = (RSZ Speed I Maximum Speed)
In addition:
Equation 2B-10
Maximum Speed = Cruise Speed 10.9 4
Where:
0.94 = Fraction of cruise speed to maximum speed
Substituting Equation 2B-10 into Equation 2B-9, the equation to calculate load factor becomes:
Equation 2B-11
LoadFactorRSZ[main} = (RSZ Speed x 0.94 /Cruise Speed)
Where:
0.94 = Fraction of cruise speed to maximum speed
Load factors below 2 percent were set to 2 percent as a minimum.
Substituting Equation 2B-8 for time in mode and Equation 2B-11 for load factor into
Equation 2B-7 , the expression used to calculate RSZ mode emissions for the main engines
becomes:
Equation 2B-12 RSZ Mode Emissions for Main Engines
Emissions R8Z[aux} = (calls) x (P^aux]) x (RSZ Distance/ RSZ Speed) x (2 trips I call)
x (RSZ Speed x 0.94 /Cruise Speed )3 x (EF[aux]) x (Adj)x (1(T6 tonnes /g)
Where:
EmissionsRszimain] = Metric tonnes emitted from main engines in RSZ mode
calls = Round-trip visits (i.e., one entrance and one clearance is considered a call)
P[main] = Total main engine power, in kilowatts
RSZ distance = one way distance, in nautical miles (specific to each port)
RSZ speed = speed, in knots (specific to each port)
2 trips/call = Used to calculate round trip RSZ distance
Cruise speed = vessel service speed, in knots
EF[main] = Emission factor for main engines for the pollutant of interest, in g/kW-hr (these vary
as a function of engine type and fuel used, rather than activity mode)
Adj = Low load adjustment factor, unitless (used when the load factor is below 0.20)
2-62
-------
106 = Conversion factor from grams to tons
0.94 = Fraction of cruise speed to maximum speed
Emission factors are considered to be relatively constant down to about 20 percent load.
Below that threshold, emission factors tend to increase significantly as the load decreases. During
the RSZ mode, load factors can fall below 20 percent. Low load multiplicative adjustment factors
were developed and applied when the load falls below 20 percent (0.20). If the load factor is 0.20
or greater, the low load adjustment factor is set to 1.0.
The equation used to calculate RSZ mode emissions for the auxiliary engines is:
Equation 2B-13 RSZ Mode Emissions for Auxiliary Engines
Emissions RSZ[aux] = (calls) x (P^aux^) x (RSZ Distance/RSZ Speed) x (2 trips/call) x (LFRSZ^aw.^) x (EF^aux^) x (1CT6 tonnes I g)
Where:
EmissionsRsziaux] = Metric tonnes emitted from auxiliary engines in RSZ mode
calls = Round-trip visits (i.e., one entrance and one clearance is considered a call)
P[aux] = Total auxiliary engine power, in kilowatts
RSZ distance = one way distance, in nautical miles (specific to each port)
RSZ speed = speed, in knots (specific to each port)
2 trips/call = Used to calculate round trip cruise distance
LFRSZ [aux] = Load factor for auxiliary engines in RSZ mode, unitless (these vary by ship type
and activity mode)
EF[aux] = Emission factor for auxiliary engines for the pollutant of interest, in g/kW-hr (these
vary as a function of engine type and fuel used, rather than activity mode)
106 = Conversion factor from grams to metric tonnes
Unlike main engines, there is no need for a low load adjustment factor for auxiliary engines,
because of the way they are generally operated. When only low loads are needed, one or more
engines are shut off, allowing the remaining engines to maintain operation at a more efficient level.
The inputs of calls, RSZ distance, and RSZ speed are the same for main and auxiliary
engines. Relative to the main engines, auxiliary engines have separate inputs for engine power,
load factor, and emission factors. The RSZ distances vary by port rather than vessel or engine type.
Some RSZ speeds vary by ship type, while others vary by DWT. Mostly, however, RSZ speed is
constant for all ships entering the harbor area. All Great Lake ports have reduced speed zone
distances of three nautical miles occurring at halfway between cruise speed and maneuvering speed.
Maneuvering
Maneuvering emissions are calculated for both propulsion (main) and auxiliary engines.
The basic equation used to calculate maneuvering mode emissions for the main engines is:
Equation 2B-14
Emissionsman[mam] = (calls) x (P[mam]) x (hrs/callmj x (LFman[mam]) x (EF[mam]) x (Adj) x (1(T6 tonne s I g)
Where:
2-63
-------
Emissionsman[main] = Metric tonnes emitted from main engines in maneuvering mode
calls = Round-trip visits (i.e., one entrance and one clearance is considered a call)
P[mam] = Total main engine power, in kilowatts
hrs/callman = Hours per call for maneuvering mode
LFman [main] = Load factor for main engines in maneuvering mode, unitless
EF[maln] = Emission factor for main engines for the pollutant of interest, in g/kW-hr (these vary
as a function of engine type and fuel used, rather than activity mode)
Adj = Low load adjustment factor, unitless (used when the load factor is below 0.20)
10 = Conversion factor from grams to metric tonnes
Maneuvering time-in-mode is estimated based on the distance a ship travels from the
breakwater or port entrance to the pier/wharf/dock (PWD). Maneuvering times also include shifts
from one PWD to another or from one port within a greater port area to another. Average
maneuvering speeds vary from 3 to 8 knots depending on direction and ship type. For consistency,
maneuvering speeds were assumed to be the dead slow setting of approximately 5.8 knots.
Load factor during maneuvering is calculated as follows:
Equation 2B-15
LoadFactorman[main} = (Man Speed[knots] IMaximum Speed[knots])
In addition:
Equation 2B-16
Maximum Speed = Cruise Speed[knots] 10.94
Where:
0.94 = Fraction of cruise speed to maximum speed
Also, the maneuvering speed is 5.8 knots. Substituting Equation 2B-16 into Equation 2B-15, and
using a maneuvering speed of 5.8 knots, the equation to calculate load factor becomes:
Equation 2B-17
LoadFactorman{m} = (5.45 / Cruise Speed)3
Load factors below 2 percent were set to 2 percent as a minimum.
Substituting Equation 2B-17 for load factor into Equation 2B-14, the expression used to
calculate maneuvering mode emissions for the main engines becomes:
Equation 2B-18 Maneuvering Mode Emissions for Main Engines
{main}>
Emissions man{main} = (calls) x (P[main}) x (hrs I call man) x (5 A5/Cruise Speed)3 x (EF[main}) x (Adj) x (1(T6 tonnes I g)
Where:
Emissionsmanimain] = Metric tonnes emitted from main engines in maneuvering mode
calls = Round-trip visits (i.e., one entrance and one clearance is considered a call)
P[main] = Total main engine power, in kilowatts
2-64
-------
hrs/callman = Hours per call for maneuvering mode
Cruise speed = Vessel service speed, in knots
EF[main] = Emission factor for main engines for the pollutant of interest, in g/kW-hr (these vary
as a function of engine type and fuel used, rather than activity mode)
Adj = Low load adjustment factor, unitless (used when the load factor is below 0.20)
10 = Conversion factor from grams to metric tonnes
Since the load factor during maneuvering usually falls below 20 percent, low load
adjustment factors are also applied accordingly. Maneuvering times are not readily available for all
117 ports. For this analysis, maneuvering times and load factors available for a subset of the ports
were used to calculate maneuvering emissions for the remaining ports. This is discussed in more
detail in section 2.3.2.3.8.
The equation used to calculate maneuvering mode emissions for the auxiliary engines is:
Equation 2B-19 Maneuvering Mode Emissions for Auxiliary Engines
Emissionsman[aux] = (calls) x (P[aux]) x (hrs/callmj x (LFman[aux]) x (EF[aux]) x (10~6 tonnes Ig)
Where:
Emissionsmaniaux] = Metric tonnes emitted from auxiliary engines in maneuvering mode
calls = Round-trip visits (i.e., one entrance and one clearance is considered a call)
P[aux] = Total auxiliary engine power, in kilowatts
hrs/callman = Hours per call for maneuvering mode
LFman [aux] = Load factor for auxiliary engines in maneuvering mode, unitless (these vary by ship
type and activity mode)
EF[aux] = Emission factor for auxiliary engines for the pollutant of interest, in g/kW-hr (these
vary as a function of engine type and fuel used, rather than activity mode)
106 = Conversion factor from grams to metric tonnes
Low load adjustment factors are not applied for auxiliary engines.
Hotelling
Hotelling emissions are calculated for auxiliary engines only, as main engines are not
operational during this mode. The equation used to calculate hotelling mode emissions for the
auxiliary engines is:
Equation 2B-20 Hotelling Mode Emissions for Auxiliary Engines
Emissions hotel[aux] = (calls) x (P[aux]) x (hrs / call hotel) x (LFhotel[aux]) x (EF[aux])x (10~6 tonnes I g)
Where:
Emissionshotei[aux] = Metric tonnes emitted from auxiliary engines in hotelling mode
calls = Round-trip visits (i.e., one entrance and one clearance is considered a call)
P[aux] = Total auxiliary engine power, in kilowatts
hrs/callhotei = Hours per call for hotelling mode
LFhotei [aux] = Load factor for auxiliary engines in hotelling mode, unitless (these vary by ship
type and activity mode)
2-65
-------
EF[aux] = Emission factor for auxiliary engines for the pollutant of interest, in g/kW-hr (these
vary as a function of engine type and fuel used, rather than activity mode)
106 = Conversion factor from grams to metric tonnes
Hotelling times are not readily available for all 117 ports. For this analysis, hotelling times
available for a subset of the ports were used to calculate hotelling emissions for the remaining ports.
2-66
-------
Appendix 2C: Port Reduced Speed Zone (RSZ) Information
Table 2C-1 Port RSZ Information
Port Name
Albany, NY
Alpena, Ml
Anacortes, WA
Anchorage, AK
Ashtabula, OH
Baltimore, MD
Barbers Point, Oahu, HI
Baton Rouge, LA
Beaumont, TX
Boston, MA
Bridgeport, CT
Brownsville, TX
Brunswick, GA
Buffalo, NY
Burns Waterway Harbor, IN
Calcite, Ml
Camden-Gloucester, NJ
Carquinez, CA
Catalina, CA
Charleston, SC
Chester, PA
Chicago, IL
Cleveland, OH
Conneaut, OH
Coos Bay, OR
Corpus Christi, TX
Detroit, Ml
Duluth-Superior, MN and Wl
El Segundo, CA
Erie, PA
Escanaba, Ml
Eureka, CA
Everett, WA
Fairport Harbor, OH
Fall River, MA
Freeport, TX
Galveston, TX
Gary, IN
Georgetown, SC
RSZ
Speed
(knts)
c
e
a
14.5
e
c
10
10
7
10
10
8.8
13
e
e
e
c
12
12
12
c
e
e
e
6.5
d
e
e
12
e
e
12
a
e
9
c
c
e
12
RSZ
distance
(naut mi)
142.5
3
108.3
143.6
3
157.1
5.1
219.8
53.5
14.3
2
18.7
38.8
3
3
3
94
39
11.9
17.3
78.2
3
3
3
13
30.1
3
3
23.3
3
3
9
123.3
3
22.7
2.6
9.3
3
17.6
Final RSZ End Point(s)
Longitude
-73.8929
-83.2037
-124.771
-152.309
-80.8097
-75.8067
-158.132
-89.4248
-89.137
-93.7552
-70.7832
-73.1863
-97.0921
-80.9345
-81.1357
-79.0996
-87.1032
-83.5383
-75.0095
-122.632
-118.465
-79.6452
-75.0095
-87.4141
-81.765
-80.5639
-124.359
-96.8753
-83.1384
-91.8536
-118.926
-118.465
-80.115
-86.9224
-124.347
-124.771
-81.3917
-71.3334
-95.2949
-94.6611
-87.2824
-79.0779
Latitude
40.47993
44.99298
48.49074
59.5608
42.08549
36.8468
21.21756
28.91161
28.98883
29.55417
42.37881
41.13906
26.06129
31.29955
30.68935
42.81683
41.80625
45.39496
38.79004
37.76094
33.63641
32.62557
38.79004
41.86971
41.63079
42.13361
43.35977
27.74433
42.10308
46.78916
33.91252
33.63641
42.3151
45.58297
40.75925
48.49074
41.91401
41.41708
28.93323
29.3247
41.77658
33.1924
2-67
-------
Port Name
Grays Harbor, WA
Gulfport, MS
Hilo, HI
Honolulu, HI
Hopewell, VA
Houston, TX
Indiana Harbor, IN
Jacksonville, FL
Kahului, Maui, HI
Kalama, WA
Lake Charles, LA
Long Beach, CA
Longview, WA
Lorain, OH
Los Angeles, CA
Manistee, Ml
Marblehead, OH
Marcus Hook, PA
Matagorda Ship Channel, TX
Miami, FL
Milwaukee, Wl
Mobile, AL
Morehead City, NC
Muskegon, Ml
Nawiliwili, Kauai, HI
New Bedford, MA
New Castle, DE
New Haven, CT
New Orleans, LA
New York, NY and NJ
Newport News, VA
Nikishka, AK
Oakland, CA
Olympia, WA
Other Puget Sound, WA
Palm Beach, FL
Panama City, FL
Pascagoula, MS
Paulsboro, NJ
Penn Manor, PA
Pensacola, FL
Philadelphia, PA
RSZ
Speed
(knts)
a
10
10
10
10
c
e
10
10
b
6
12
b
e
12
e
e
c
7.3
12
e
11
10
e
10
9
c
10
10
c
14
14.5
12
a
a
3
10
10
c
c
12
c
RSZ
distance
(naut mi)
4.9
17.4
7.1
10
91.8
49.6
3
18.6
7.5
68.2
38
18.1
67.3
3
20.6
3
3
94.7
24
3.8
3
36.1
2.2
3
7.3
22.4
60.5
2.1
104.2
15.7
24.3
90.7
18.4
185.9
106
3.1
10
17.5
83.5
114.5
12.7
88.1
Final RSZ End Point(s)
Longitude
-124.24
-88.9263
-154.985
-157.956
-157.785
-75.8067
-94.6611
-87.4007
-81.3649
-156.44
-124.137
-93.3389
-118.465
-118.13
-124.137
-82.2701
-118.465
-118.13
-86.3819
-82.7293
-75.0095
-96.2287
-80.1201
-87.6718
-88.0644
-76.6679
-86.5377
-159.266
-71.1013
-75.0095
-72.9121
-89.4248
-89.137
-73.8929
-75.8067
-152.309
-122.632
-124.771
-124.771
-79.9973
-84.1797
-88.4804
-75.0095
-75.0095
-87.298
-75.0095
Latitude
46.89509
30.11401
19.76978
21.17658
21.23827
36.8468
29.3247
41.8401
30.39769
21.01066
46.22011
29.73094
33.63641
33.45211
46.22011
41.64023
33.63641
33.45211
44.41573
41.69638
38.79004
28.33472
25.75787
42.97343
30.1457
34.68999
43.29151
21.87705
41.38499
38.79004
41.26588
28.91161
28.98883
40.47993
36.8468
59.5608
37.76094
48.49074
48.49074
26.77129
30.0818
30.09597
38.79004
38.79004
30.27777
38.79004
2-68
-------
Port Name
Plaquemines, LA, Port of
Port Angeles, WA
Port Arthur, TX
Port Canaveral, FL
Port Dolomite, Ml
Port Everglades, FL
Port Hueneme, CA
Port Inland, Ml
Port Manatee, FL
Portland, ME
Portland, OR
Presque Isle, Ml
Providence, Rl
Redwood City, CA
Richmond, CA
Richmond, VA
Sacramento, CA
San Diego, CA
San Francisco, CA
Sandusky, OH
Savannah, GA
Searsport, ME
Seattle, WA
South Louisiana, LA, Port of
St. Clair, Ml
Stockton, CA
Stoneport, Ml
Tacoma, WA
Tampa, FL
Texas City, TX
Toledo, OH
Two Harbors, MN
Valdez, AK
Vancouver, WA
Wilmington, DE
Wilmington, NC
RSZ
Speed
(knts)
10
a
7
10
e
7.5
12
e
9
10
b
e
9
12
12
10
12
12
12
e
13
9
a
10
e
12
e
a
9
c
e
e
10
b
c
10
RSZ
distance
(naut mi)
52.4
65
21
4.4
3
2.1
2.8
3
27.4
11.4
105.1
3
24.9
36
22.6
106.4
90.5
11.7
14.4
3
45.5
22.2
133.3
142.8
3
86.9
3
150.5
30
15.1
3
3
27.2
95.7
65.3
27.6
Final RSZ End Point(s)
Longitude
-89.4248
-89.137
-124.771
-93.7552
-80.5328
-84.2445
-80.082
-119.238
-85.6524
-83.0364
-70.1077
-124.137
-87.082
-71.3334
-122.632
-122.632
-75.8067
-122.632
-117.315
-122.632
-82.5251
-78.0498
-68.7645
-124.771
-89.4248
-89.137
-82.5838
-122.632
-83.2355
-124.771
-83.0364
-94.6611
-83.3034
-91.4414
-146.881
-124.137
-75.0095
-80.325
Latitude
28.91161
28.98883
48.49074
29.55417
28.41439
45.83181
26.08627
34.10859
45.87553
27.59078
43.54224
46.22011
46.5804
41.41708
37.76094
37.76094
36.8468
37.76094
32.62184
37.76094
41.56193
33.83598
44.1179
48.49074
28.91161
28.98883
42.55923
37.76094
45.25919
48.49074
27.59078
29.3247
41.7323
46.93391
60.86513
46.22011
38.79004
31.84669
a Cruise speed through Strait of Juan de Fuca, then varies by ship type for remaining journey
b Inbound on Columbia River at 6.5 knots, outbound at 12 knots
c Speed varies by ship type similar to typical like port
d Speed varies by ship DWTs
e All Great Lake ports have reduced speed zone distances of 3 nautical miles with speeds halfway
between service speed and maneuvering speed.
2-69
-------
Appendix 2D: Use of Detailed Typical Port Data for Other Inputs
There is currently not enough information to readily calculate time-in-mode (hours/call) for
all 117 ports during the maneuvering and hotelling modes of operation. As a result, it was
necessary to review and select available detailed emission inventories that have been estimated for
selected ports to date. These ports are referred to as typical ports. The typical port information for
maneuvering and hotelling time-in-mode (as well as maneuvering load factors for the propulsion
engines) was then used for the typical ports and also assigned to the other modeled ports. A
modeled port is the port in which emissions are to be estimated. The methodology that was used to
select the typical ports and match these ports to the other modeled ports is briefly described in this
appendix, and more fully described in the ICF documentation.39
2.6.1 Selection of Typical Ports
In 1999, the U.S. Government published two guidance documents40'41 to calculate marine
vessel activity at ports. These documents contained detailed port inventories of eight deep sea
ports, two Great Lake ports and two inland river ports. The detailed inventories were developed by
obtaining ship call data from Marine Exchanges/Port Authorities (MEPA) at the various ports for
1996 and matching the various ship calls to data from Lloyds Maritime Information Services to
provide ship characteristics. The ports for which detailed inventories were developed are shown in
Table 2D-1 for deep sea ports and Table 2D-2 for Great Lake ports along with the level of detail of
shifts for each port. Most ports provided the ship name, Lloyd's number, the vessel type, the date
and time the vessel entered and left the port, and the vessel flag. Inland river ports were developed
from US Army Corps of Engineers (USAGE) Waterborne Commerce Statistics Center data.
Table 2D-1 Deep Sea MEPA Vessel Movement and Shifting Details
MEPA Area and Ports
Lower Mississippi River
including the ports of New
Orleans, South Louisiana,
Plaquemines, and Baton Rouge
Consolidated Port of New York
and New Jersey and other ports
on the Hudson and Elizabeth
Rivers
Delaware River Ports including
the ports of Philadelphia,
Camden, Wilmington and others
Puget Sound Area Ports including
the ports of Seattle, Tacoma,
Olympia, Bellingham, Anacortes,
and Grays Harbor
The Port of Corpus Christi, TX
MEPA Data Includes
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.
All PWDs or anchorages for shifting are named. Shifting arrival
and departure times are not given. Hotelling time is based upon the
entrance and clearance times and dates, subtracting out
maneuvering times. Maneuvering times were calculated based
upon the distance the ship traveled at a given maneuvering speed.
All PWDs or anchorages for shifting are named. Shifting arrival
and departure times are not given. Hotelling time is based upon the
entrance and clearance times and dates, subtracting out
maneuvering times. Maneuvering times were calculated based
upon the distance the ship traveled at a given maneuvering speed.
All PWDs or anchorages for shifting are named. Arrival and
departure dates and times are noted for all movements, allowing
calculation of maneuvering and hotelling both for individual shifts
and the overall call on port.
Only has information on destination PWD and date and time in
and out of the port area. No shifting details.
2-70
-------
MEPA Area and Ports
The Port of Coos Bay, OR
Patapsco River Ports including
the port of Baltimore Harbor, MD
The Port of Tampa, FL
MEPA Data Includes
Only has information on destination PWD and date and time in
and out of the port area. No shifting details.
All PWDs or anchorages for shifting are named. Shifting arrival
and departure times are not given. Hotelling time is based upon the
entrance and clearance times and dates, subtracting out
maneuvering times. Maneuvering times were calculated based
upon the distance the ship traveled at a given maneuvering speed.
All PWDs or anchorages for shifting are named. Arrival and
departure dates and times are noted for all movements, allowing
calculation of maneuvering and hotelling both for individual shifts
and the overall call.
Table 2D-2 Great Lake MEPA movements and shifts
MEPA Area and Ports
Port of Cleveland, OH
Port of Burns Harbor, IN
MEPA Data Includes
Information on the first and last PWD for the vessel (gives
information for at most one shift per vessel) . No information
intermediate PWDs..
on
No shifting details, No PWDs listed..
Since 1999, several new detailed emissions inventories have been developed and were
reviewed for use as additional or replacement typical ports: These included:
• Port of Los Angeles38'42
• Puget Sound Ports43
• Port of New York/New Jersey44
• Port of Houston/Galveston45
• Port of Beaumont/Port Arthur46
• Port of Corpus Christi47
• Port of Portland48
• Ports of Cleveland, OH and Duluth-Superior, MN&WI49
Based on the review of these newer studies, some of the previous typical ports were replaced
with newer data and an additional typical port was added. Data developed for Cleveland and
Duluth-Superior for LADCO was used in lieu of the previous typical port data for Cleveland and
Burns Harbor because it provided more detailed information and better engine category definitions.
The Port of Houston/Galveston inventory provided enough data to add an additional typical port.
All three port inventories were adjusted to reflect the current methodology used in this study.
The information provided in the current inventory for Puget Sound Ports43 was used to
calculate RSZ speeds, load factors, and times for all Puget Sound ports. As described in Section
2-71
-------
2.6.3.2, an additional modeled port was also added to account for the considerable amount of Jones
Act tanker ship activity in the Puget Sound area that is not contained in the original inventory.
The newer Port of New York/New Jersey inventory provided a check against estimates
made using the 1996 data. All other new inventory information was found to lack sufficient detail
to prepare the detailed typical port inventories needed for this project.
The final list of nine deep sea and two Great Lake typical ports used in this analysis and their
data year is as follows:
• Lower Mississippi River Ports [1996]
• Consolidated Ports of New York and New Jersey and Hudson River [1996]
• Delaware River Ports [1996]
• Puget Sound Area Ports [1996]
• Corpus Christi, TX[ 1996]
• Houston/Galveston Area Ports [1997]
• Ports on the Patapsco River [1996]
• Port of Coos Bay, OR [1996]
• Port of Tampa, FL [1996]
• Port of Cleveland, OH on Lake Erie [2005]
• Duluth-Superior, MN & WI on Lake Michigan [2005]
The maneuvering and hotelling time-in-modes, as well as the maneuvering load factors for
these typical ports, were binned by ship type, engine type, and DWT type, using the same bins
described in the section entitled "Bins by Ship Type, Engine Type, and DWT Range."
2.6.2 Matching Typical Ports to Modeled Ports
The next step in the process was to match the ports to be modeled with the typical port
which was most like it. Three criteria were used for matching a given port to a typical port:
regional differences0, maximum vessel draft, and the ship types that call on a specific port. One
container port, for instance, may have much smaller bulk cargo and reefer ships number of calls on
that port than another. Using these three criteria and the eleven typical ports that are suitable for
port matching, the 89 deep sea ports and 28 Great Lake ports were matched to the typical ports. For
a typical port, the modeled and typical port is the same (i.e., the port simply represents itself). For
California ports, we used data provided by ARE as discussed in Section 2.6.3. The matched ports
for the deep sea ports are provided in Table 2D-3.
D The region in which a port was located was used to group top ports as it was considered a primary influence on the
characteristics (size and installed power) of the vessels calling at those ports.
2-72
-------
Table 2D-3 Matched Ports for the Deep Sea Ports
Modeled Port Name
Anacortes, WA
Barbers Point, HI
Everett, WA
Grays Harbor, WA
Honolulu, HI
Kalama, WA
Longview, WA
Olympia, WA
Port Angeles, WA
Portland, OR
Seattle, WA
Tacoma, WA
Vancouver, WA
Valdez, AK
Other Puget Sound
Anchorage, AK
Coos Bay, OR
Hilo, HI
Kahului, HI
Nawiliwili, HI
Nikishka, AK
Beaumont, TX
Freeport, TX
Galveston, TX
Houston, TX
Port Arthur, TX
Texas City, TX
Corpus Christi, TX
Lake Charles, LA
Mobile, AL
Brownsville, TX
Gulfport, MS
Manatee, FL
Matagorda Ship
Panama City, FL
Pascagoula, MS
Pensacola, FL
Tampa, FL
Everglades, FL
New Orleans, LA
Baton Rouge, LA
Typical Like Port
Puget Sound
Puget Sound
Puget Sound
Puget Sound
Puget Sound
Puget Sound
Puget Sound
Puget Sound
Puget Sound
Puget Sound
Puget Sound
Puget Sound
Puget Sound
Puget Sound
Puget Sound
Coos Bay
Coos Bay
Coos Bay
Coos Bay
Coos Bay
Coos Bay
Houston
Houston
Houston
Houston
Houston
Houston
Corpus Christi
Corpus Christi
Corpus Christi
Tampa
Tampa
Tampa
Tampa
Tampa
Tampa
Tampa
Tampa
Tampa
Lower Mississippi
Lower Mississippi
2-73
-------
Modeled Port Name
South Louisiana, LA
Plaquemines, LA
Albany, NY
New York/New Jersey
Portland, ME
Georgetown, SC
Hopewell, VA
Marcus Hook, PA
Morehead City, NC
Paulsboro, NJ
Chester, PA
Fall River, MA
New Castle, DE
Penn Manor, PA
Providence, RI
Brunswick, GA
Canaveral, FL
Charleston, SC
New Haven, CT
Palm Beach, FL
Bridgeport, CT
Camden, NJ
Philadelphia, PA
Wilmington, DE
Wilmington, NC
Richmond, VA
Jacksonville, FL
Miami, FL
Searsport, ME
Boston, MA
New Bedford/Fairhaven, MA
Baltimore, MD
Newport News, VA
Savannah, GA
Catalina, CA
Carquinez, CA
El Segundo, CA
Eureka, CA
Hueneme, CA
Long Beach, CA
Los Angeles, CA
Oakland, CA
Typical Like Port
Lower Mississippi
Lower Mississippi
New York/New Jersey
New York/New Jersey
New York/New Jersey
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Delaware River
Patapsco River
Patapsco River
Patapsco River
ARE Supplied
ARE Supplied
ARE Supplied
ARE Supplied
ARE Supplied
ARE Supplied
ARE Supplied
ARE Supplied
2-74
-------
Modeled Port Name
Redwood City, CA
Richmond, CA
Sacramento, CA
San Diego, CA
San Francisco, CA
Stockton, CA
Typical Like Port
ARE Supplied
ARE Supplied
ARE Supplied
ARE Supplied
ARE Supplied
ARE Supplied
Great Lake ports were matched to either Cleveland or Duluth as shown in Table 2D-4.
Table 2D-4 Great Lake Match Ports
Port Name
Alpena, MI
Buffalo, NY
Burns Waterway, IN
Calcite, MI
Cleveland, OH
Dolomite, MI
Erie, PA
Escanaba, MI
Fairport, OH
Gary, IN
Lorain, OH
Marblehead, OH
Milwaukee, WI
Muskegon, MI
Presque Isle, MI
St Clair, MI
Stoneport, MI
Two Harbors, MN
Ashtabula, OH
Chicago, IL
Conneaut, OH
Detroit, MI
Duluth-Superior, MN&WI
Indiana, IN
Inland Harbor, MI
Manistee, MI
Sandusky, OH
Toledo, OH
Typical Like Port
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Cleveland
Duluth-Superior
Duluth-Superior
Duluth-Superior
Duluth-Superior
Duluth-Superior
Duluth-Superior
Duluth-Superior
Duluth-Superior
Duluth-Superior
Duluth-Superior
2-75
-------
Once a modeled port was matched to a typical port, the maneuvering and hotelling time-in-
mode values, as well as the maneuvering load factors by bin for the typical ports, were used directly
for the modeled ports, with no adjustments.
2.6.2.1 Bin Mismatches
In some cases, the specific DWT range bin at the modeled port was not in the typical like
port data. In those cases, the next nearest DWT range bin was used for the calculations. In a few
cases, the engine type for a given ship type might not be in the typical like port data. In these cases,
the closest engine type at the typical like port was used. Also in a few cases, a specific ship type in
the modeled port data was not in the typical like port data. In this case, the nearest like ship type at
the typical port was chosen to calculate emissions at the modeled port.
2.6.3 Stand Alone Ports
In a few cases, the USAGE entrances and clearances data was not used to calculate
emissions at the modeled port. These include the California ports for which we received data from
ARE, the Port of Valdez, Alaska, and a conglomerate port within the Puget Sound area, as
described below.
2.6.3.1 California Ports
The California Air Resources Board (ARE) supplied inventories for 14 California ports for
2002. The data received from ARE for the California ports were modified to provide consistent PM
and SOz emissions to those calculated in this report. In addition, cruise and RSZ emissions were
calculated directly based upon average ship power provided in the ARE methodology document50
and number of calls, because ARE did not calculate cruise emissions, and transit (RSZ) emissions
were allocated to counties instead of ports. ARE provided transit distances for each port to
calculate the RSZ emissions. Ship propulsion and auxiliary engine power were calculated based
upon the methodology previously described for use in computing cruise and RSZ emissions. For
maneuvering and hotelling emissions, the ARE values were used and adjusted as discussed below.
The data supplied by ARE included domestic traffic as well as foreign cargo traffic.
For PM emission calculations, ARE used an emission factor of 1.5 g/kWh to calculate total
PM emissions and factors of 0.96 and 0.937 to convert total PM to PMio and PM2.s respectively.
Since an emission factor of 1.4 g/kWh was used in our calculations for PMio and an emission factor
of 1.3 g/kWh for PM2.5, ARE PMio and PM2.5 emissions were multiplied by factors of 0.972 and
0.925, respectively to get consistent PMio and PM2.5 emissions for propulsion engines.
For auxiliary engines, ARE used the same emission factors as above, while we used
and PM2.5 emission factors of 1.3 and 1.2 g/kWh, respectively for passenger ships and 1.1 and 1.0
g/kWh, respectively for all other ships. In the ARE inventory, all passenger ships are treated as
electric drive and all emissions are allocated to auxiliary engines. ARE auxiliary engine emissions
were thus multiplied by factors of 0.903 and 0.854 respectively for passenger ships and 0.764 and
0.71 1 respectively for other ships to provide consistent PM emission calculations.
2-76
-------
E
emissions were also different between the ARE and these analyses. ARE used a
composite propulsion engine S02 emission factor of 10.55 g/kWh while we used a composite S02
emission factor of 9.57 g/kWh. Thus, ARE SOz propulsion emissions were multiplied by a factor
of 0.907 to be consistent with our emission calculations. For auxiliary engines, ARE used SOz
emission factors of 1 1.48 and 9.34 g/kWh, respectively for passenger and other ships, while we use
emission factors of 9.93 and 9.07 g/kWh, respectively. Thus, ARE auxiliary SOz emissions were
multiplied by factors of 0.865 and 0.971, respectively for passenger and other ships to provide
consistent SOz emissions.
2.6.3.2 Port in Puget Sound
In the newest Puget Sound inventory43, it was found that a considerable amount of tanker
ships stop at Cherry Point, Ferndale, March Point and other areas which are not within the top 89
U.S. deep sea ports analyzed in this analysis. In addition, since they are ships carrying U.S. cargo
(oil from Alaska) from one U.S. port to another, they are not documented in the USAGE entrances
and clearances data. To compensate for this anomaly, an additional port was added which
encompassed these tanker ships stopping within the Puget Sound area but not at one of the Puget
Sound ports analyzed in this analysis. Ship calls in the 1996 typical port data to ports other than
those in the top 89 U.S. deep sea ports were analyzed separately. There were 363 ship calls by
tankers to those areas in 1996. In the inventory report for 2005, there were 468 calls. For 2002, it
was estimated there were 432 calls. The same ship types and ship characteristics were used as in
the 1996 data, but the number of calls was proportionally increased to 432 calls to represent these
ships. The location of the "Other Puget Sound" port was approximately at Cherry Point near
Aberdeen.
2.6.3.3 PortofValdez
In a recent Alaska port inventory,51 it was found that significant Category 3 domestic tanker
traffic enters and leaves the Port of Valdez on destination to West Coast ports. Since the USAGE
entrances and clearances data did not contain any tanker calls at Valdez in 2002, the recent Alaska
inventory data was used to calculate emissions at that port. In this case, the number of calls and
ship characteristics for 2002 were taken directly from the Alaska inventory and used in determining
emissions for the modeled port with the Puget Sound area typical port being used as the like port.
E Based upon ARE assuming 95 percent of the engines were SSD and 5 percent were MSD. The composite S02 EF of
9.57 g/kW-hr was calculated using this weighting, along with the SSD and MSD S02 EFs for the West Coast ports
reported in Table 2-4.
2-77
-------
Appendix 2E: Emission Inputs to STEEM
The STEEM waterway network model relies on a number of inputs to identify the
movements for each vessel, individual ship attributes, and related emission factor information.
Each of these databases is described separately below.
2.6.4 Shipping Movements
The shipping activity and routes database provides information on vessel movements or
trips. It is developed using port entrances and clearances information from the USAGE report for
the U.S. and the Lloyd's Maritime Intelligence Unit (LMIU) for Canada and Mexico.52 These
sources contain information for each vessel carrying foreign cargo at each major port or waterway
that, most importantly for this analysis, includes:
• Vessel name
• Last port of call (entrance record) or next port of call (clearance record)
The database then establishes unique identification numbers for each ship, each port pair,
and each resulting trip.
2.6.5 Ship Attributes
The ship attributes data set contains the important characteristics of each ship that are
necessary for the STEEM interport model to calculate the emissions associated with each trip. The
information in this data set is matched to each previously assigned ship identification number. The
following information comes from the USAGE entrances and clearances report for each ship
identification number:
• Ship type
• Gross registered tonnage (CRT)
• Net registered tonnage (NRT)
The ship attributes data set contains the following information from Lloyd's Register-
Fairplay for each ship identification number.
• Main propulsion engine installed power (horsepower)
• Service speed (cruise speed)
• Ship size (length, wide, and draft)
Sometimes data was lacking from the above references for ship speed. In these instances,
the missing information was developed for each of nine vessel types and the appropriate value was
applied to each individual ship of that type. Specifically, the missing ship speeds for each ship
category were obtained from the average speeds used in a Lloyd's Register study of the Baltic Sea
and from an Entec UK Limited study for the European Commission. 54 The resulting vessel
cruise speeds for ships with missing data are shown in Table 2E-1.
2-78
-------
Table 2E-1 Average Vessel Cruise Speed by Ship Type"
Ship Type Average Cruise Speed (knots)
Bulk Carrier
Container Ship
General Cargo
Passenger Ship
Refrigerated Cargo
Roll On-Roll Off
Tanker
Fishing
Miscellaneous
14.1
19.9
12.3
22.4
16.4
16.9
13.2
11.7
12.7
a Used only when ship specific data were missing from the
commercial database references.
The average speed during maneuvering is approximately 60 percent of a ship's cruise speed
based on using the propeller law described earlier and the engine load factor for maneuvering that is
presented later in this section.
As with vessel cruise speed, main engine installed power was sometimes lacking in the
Lloyd's Register-Fairplay data set. Here again, the missing information was developed for nine
different vessel types and the appropriate value was applied to each individual ship of that type
when the data were lacking. In this case, the missing main engine horsepower was estimated by
regressing the relationships between CRT and NRT, and between installed power and CRT for each
category. This operation is performed internally in the model and the result applied to each
individual ship, as appropriate.
The ship attributes database also contains information on the installed power of engines used
for auxiliary purposes. However, this information is usually lacking in the Lloyds data set, so an
alternative technique was employed to estimate the required values. In short, the STEEM model
uses a ratio of main engine horsepower to auxiliary engine horsepower that was determined for
eight different vessel types using information primarily from ICF International.55 (The ICF report
attributed these power values to a study for the Port of Los Angeles by Starcrest Consulting.38) The
auxiliary engine power for each individual vessel of a given ship type is then estimated by
multiplying the appropriate main power to auxiliary power ratio and the main engine horsepower
rating for that individual ship. The main and auxiliary power values and the resulting auxiliary
engine to main engine ratios are shown in Table 2E-2.
2-79
-------
Table 2E-2 Auxiliary Engine Power Ratios
Vessel Type
Bulk Carrier
Container Ship
General Cargo
Passenger Ship
Refrigerated Cargo
Roll On-Roll Off
Tanker
Miscellaneous
Average Main
Engine Power
(kW)
7,954
30,885
9,331
39,563
9,567
10,696 c
9,409
6,252
Average Auxiliary
Engine Power (kW)
1,169
5,746
1,777
39,563 a
3,900 b
2,156C
1,985
1,680
Auxiliary to Main
Engine Power
Ratio
0.147
0.186
0.190
1.000
0.136
0.202
0.211
0.269
a The ICF reference reported a value of 11,000 for auxiliary engines used on passenger
-i J J
vessels.
b The STEEM used auxiliary engine power as reported in the ARE methodology document.
c The STEEM purportedly used values for Roll On-Roll Off main and auxiliary engines that
represent a trip weighted average of the Auto Carrier and Cruise Ship power values from the
ICF reference.
Finally, the ship attributes database provides information on the load factors for main
engines during cruise and maneuvering operation, in addition to load factors for auxiliary marine
engines. Main engine load factors for cruise operation were taken from a study of international
shipping for all ship types, except passenger vessels.56 For this analysis, the STEEM model used a
propulsion engine load factor for passenger ship engines at cruise speed of 55 percent of the total
installed power. This is based on engine manufacturer data contained in two global shipping
studies.56'57 During maneuvering, it was assumed that all main engines, including those for
passenger ships, operate at 20 percent of the installed power. This is consistent with a study done
by Entec UK for the European Commission. The main engine load factors at cruise speed by ship
type are shown in Table 2E-3.
Auxiliary engine load factors, except for passenger ships, were obtained from the ICF
International study referenced above. These values are also shown in Table 2E-. For cruise mode,
neither port nor interport portions of the inventory were adjusted for low load operation, as the low
load adjustments are only applied to propulsion engines with load factors below 20%.
Table 2E-3 Main and Auxiliary Engine Load Factors at Cruise Speed by Ship Type
Ship Type
Bulk Carrier
Container Ship
General Cargo
Passenger Ship
Refrigerated Cargo
Roll On-Roll Off
Tanker
Miscellaneous
Average Main Engine
Load Factor (%)
75
80
80
55
80
80
75
70
Average Auxiliary Engine
Load Factor (%)
17
13
17
25
20
15
13
17
2-80
-------
2.6.6 Emission Factor Information
The emission factor data set contains emission rates for the various pollutants in terms of
grams of pollutant per kilowatt-hour (g/kW-hr) . The main engine emission factors are shown in
Table 2E-4. The speed specific factors for NOx, HC, and SOz were taken from several recent
analyses of ship emissions in the U.S., Canada, and Europe. 50'55'56' 58 The PM factor was based
on discussions with the California Air Resources Board (ARE) staff. The fuel specific CO emission
factor was taken from a report by ENVIRON International.59 The STEEM study used the composite
emission factors shown in the table because the voyage data used in the model do not explicitly
identify main engine speed ratings, i.e., slow or medium, or the auxiliary engine fuel type, i.e.,
marine distillate or residual marine. The composite factor for each pollutant is determined by
weighting individual emission factors by vessel engine population data from a 2005 survey of
ocean-going vessels that was performed by ARE.6
Table 2E-4 Main Engine Emission Factors by Ship and Fuel Type
Engine Type
Slow Speed
Medium Speed
Composite EF
Main Engine Emission Factors (g/kW-hr)
Fuel Type
Residual
Marine
Residual
Marine
Residual
Marine
NOX
18.1
14
17.9
PM10
1.5
1.5
1.5
PM25a
1.4
1.4
1.4
HC
0.6
0.5
0.6
CO
1.4
1.1
1.4
SO2
10.5
11.5
10.6
a Estimated from PMi0 using a multiplicative adjustment factor of 0.92.
The emission factors for auxiliary engines are shown in Table 2E-5. The fuel specific main
emission factors for NOx and HC were taken from several recent analyses of ship emissions in the
U.S., Canada, and Europe, as referenced above for the main engine load factors. The PM factor for
marine distillate was taken from a report by ENVIRON International, which was also referenced
above. The PM factor for residual marine was based on discussions with the California Air
Resources Board (ARE) staff. The CO factors are from the Starcrest Consulting study of the Port
of Los Angeles.38 For SOz, the fuel specific emission factors were obtained from Entec and
Corbett and Koehler:56 The composite emission factors displayed in the table are discussed below.
Table 2E-5 Auxiliary Engine Emission Factors by Ship and Fuel Type
Engine Type
Medium Speed
Medium Speed
Composite EF
Auxiliary Engine Emission Factors (g/kW-hr)
Fuel Type
Marine
Distillate
Residual
Marine
Residual
Marine
NOX
13.9
14.7
14.5
PM10
0.3
1.5
1.2
PM25a
0.3
1.4
1.1
HC
0.4
0.4
0.4
CO
1.1
1.1
1.1
S02
4.3
12.3
**
a Estimated from PM10 using a multiplicative adjustment factor of 0.92.
b See Table 2E-6 for composite S02 emission factors by vessel type.
2-81
-------
As for main engines, the STEEM study used the composite emission factors for auxiliary
engines. For all pollutants other than SOz, underlying data used in the model do not explicitly
identify auxiliary engine voyages by fuel type, i.e., marine distillate or residual marine. Again, the
composite factor for those pollutants was determined by weighting individual emission factors by
vessel engine population data from a 2005 survey of ocean-going vessels that was performed by
ARE.
61
For S02, composite emission factors for auxiliary engines were calculated for each vessel
type. These composite factors were determined by taking the fuel specific emission factors from
Table 2E-5 and weighting them with an estimate of the amount of marine distillate and residual
marine that is used by these engines. The relative amount of each fuel type consumed was taken
from the 2005 ARE survey. The relative amounts of each fuel type for each vessel type and the
resulting SOz emission factors are shown in Table 2E-6.
Table 2E-6 Auxiliary Engine SO2 Composite Emission Factors by Vessel Type
Vessel Type
Bulk Carrier
Container Ship
General Cargo
Passenger Ship
Refrigerated Cargo
Roll On-Roll Off
Tanker
Miscellaneous
Residual Marine
(%)
71
71
71
92
71
71
71
0
Marine Distillate
(%)
29
29
29
8
29
29
29
100
Composite
Emission Factor
(g/kW-hr)
9.98
9.98
9.98
11.66
9.98
9.98
9.98
4.3
2.6.7 Adjustments to STEEM PM and SO2 Emission Inventories
The interport emission results contained in this study for PMio and SOz were taken from the
STEEM inventories and then adjusted to reflect the U.S. Government's recent review of available
engine test data and fuel sulfur levels for the near port analysis. In the near ports work, a PM
emission factor of 1.4 g/kW-hr was used for most main engines, e.g., slow speed diesel and medium
speed diesel engines, all of which are assumed to use residual marine. A slightly higher value was
used for steam turbine and gas turbine engines, and a slightly lower value was used for most
auxiliary engines. However, these engines represent only a small fraction of the total emissions
inventory. As shown in Section 2.6.6, the STEEM study used an emission factor of 1.5 g/kW-hr for
all main engines and a slightly lower value for auxiliary engines. Here again, the auxiliary engines
comprise only a small fraction of the total emissions from these ships. Therefore, for simplicity, the
interport PM inventories were adjusted by multiplying the STEEM results by the ratio of the two
primary emission factors, i.e., 1.4/1.5 or 0.933, to approximate the difference in fuel effects.
2-82
-------
Appendix 2F: Inventories Used for Air Quality Modeling
The emission inventories presented in this chapter are slightly different from the emissions
inventories used in the air quality modeling presented in Chapter 3. Specifically, the inventories
used in the air quality modeling reflect a slightly different boundary for the proposed EGA that was
based on a measurement error. Due to the nature of the measurement error, the corrections to the
EGA boundaries are not uniform, but are different by coastal area. As seen in Table 2F-1, the
changes are not expected to have a significant impact on the results of our analysis. The
measurement error affects only those portions that are farthest from shore.
The inventories used for air quality modeling also only contain Tier I NOx controls, as
opposed to the Tier I and Tier II controls contained in the final inventories.
A comparison of the air quality and final inventories by region for the 2020 baseline
scenarios is provided in Table 2F-1. Results are provided only for NOx, PM2.s, and S02, since the
air quality modeling is focused on ozone and PM2.5. As shown, the inventory provided for air
quality modeling generally understates the inventory reductions and air quality benefits produced by
the EGA.
Table 2F-1 Comparison of Air Quality Inventories vs Final Inventories for 2020 Baseline Case
U.S. Region
East Coast (EC)
Gulf Coast (GC)
North Pacific
(NP)
South Pacific
(SP)
Great Lakes
(GL)
Total 48-State
Metric Tonnes per Year
NOX
AQ
439,713
261,024
42,291
216,849
19,842
979,719
Final
391,995
232,114
38,051
208,294
18,768
889,222
% Diff
12%
12%
11%
4%
6%
10%
PM25
AQ
35,891
21,669
3,575
17,092
1,484
79,711
Final
35,882
21,531
3,603
18,536
1,484
81,037
% Diff
0%
1%
-1%
-8%
0%
-2%
SOX
AQ
323,108
175,862
27,580
138,102
11,993
676,645
Final
323,038
174,751
27,807
149,751
11,993
687,339
% Diff
0%
1%
-1%
-8%
0%
-2%
2-83
-------
1 ICF International (March 2009). Inventory Contribution of U.S. Flagged Vessels, prepared for the U.S.
Environmental Protection Agency, EPA Report Number EPA-420-R-09-005.
2 ICF International (October 2007). Commercial Marine Port Inventory Development, prepared for the U.S.
Environmental Protection Agency, EPA Report Number EPA-420-R-07-012c, Docket ID EPA-HQ-OAR-2007-0121-
0063.1.
3 Corbett, J. et al. (April 2007). Estimation, Validation and Forecasts of Regional Commercial Marine Vessel
Inventories, Final Report, prepared by University of Delaware for the California Air Resource Board, Contract Number
04-346, and the Commission for Environmental Cooperation in North America, Contract Number 113.111, Docket ID
EPA-HQ-OAR-2007-0121-0063.2.
4 Corbett, J. et al. (May 2006). Estimation, Validation and Forecasts of Regional Commercial Marine Vessel
Inventories, Tasks 1 and 2: Baseline Inventory and Ports Comparison, Final Report, prepared by University of
Delaware for the California Air Resource Board, Contract Number 04-346, and the Commission for Environmental
Cooperation in North America, Contract Number 113.111, May 2006, Docket ID EPA-HQ-OAR-2007-0121-0013.
5 RTI International (December 2006). Global Trade and Fuels Assessment - Future Trends and Effects of Designation
Requiring Clean Fuels in the Marine Sector: Task Order No. 1, Draft Report, prepared for the U.S. Environmental
Protection Agency, EPA Report Number EPA420-D-07-006, Docket ID EPA-HQ-OAR-2007-0121-0063.3.
6 RTI International (April 24, 2006). RTI Estimates of Growth in Bunker Fuel Consumption, Memorandum with
spreadsheet from Michael Gallaher and Martin Ross, RTI, to Barry Garelick and Russ Smith, U.S. Environmental
Protection Agency, Docket ID EPA-HQ-OAR-2007-0121-0063.4.
7 National Oceanic and Atmospheric Administration, Exclusive Economic Zone, Available online at
http://nauticalcharts.noaa.gov/csdl/eez.htm.
8 U.S. Department of Interior, North American Atlas-Political Boundaries, Available online at
http://www.nationalatlas.gov/mld/boundOm.html.
9 US Department of Transportation Maritime Administration, U.S. Water Transportation Statistical Snapshot, May
2008, available from www.marad.dot.gov
10 U.S. Army Corps of Engineers Navigation Data Center, Principal Ports of the United States, 2002, available at
http://www.iwr.usace.army.mil/ndc/db/pport/dbf/pport02.dbf.
11 U.S. Army Corps of Engineers Navigation Data Center, Vessel Entrances and Clearances, 2002, available at
http://www.iwr.usace.army.mil/ndc/db/entclrn/data/entrclrn02/
12 ICF International (October 2007). Commercial Marine Port Inventory Development, prepared for the U.S.
Environmental Protection Agency, EPA Report Number EPA-420-R-07-012c, Docket ID EPA-HQ-OAR-2007-0121-
0063.1.
13 Nexus Media Communications, The Motor Ship's Guide to Marine Diesel Engines 2005, available at
http://www.motorship.com/
14 U.S. Army Corps of Engineers, National Waterway Network, Available online at
http://www.iwr.usace.army.mil/ndc/data/datanwn.htm, Downloaded April 2006.
15 California Air Resources Board (September 2005). 2005 Oceangoing Ship Survey, Summary of Results.
2-84
-------
16 Starcrest Consulting Group (June 2004). Port-Wide Baseline Air Emissions Inventory, prepared for the Port of Los
Angeles
17 Entec UK Limited (2002). Quantification of Emissions from Ships Associated with Ship Movements between Ports in
the European Community, prepared for the European Commission, Docket ID EPA-HQ-OAR-2007-0121-0059.
18 U.S. Environmental Protection Agency (January 2009). Main Engine CO and HC Emission Factors in C3 Model and
Current Literature, Memorandum from Ari Kahan to Docket EPA-HQ-OAR-2007-0121.
19 U.S. Environmental Protection Agency (September 2007). Estimation of Particulate Matter Emission Factors for
Diesel Engines on Ocean-Going Vessels, Memorandum from Mike Samulski to Docket EPA-HQ-OAR-2007-0121,
Docket ID EPA-HQ-OAR-2007-0121-0060.
20 U.S. Environmental Protection Agency (September 2007). Estimation of Particulate Matter Emission Factors for
Diesel Engines on Ocean-Going Vessels, Memorandum from Mike Samulski to Docket EPA-HQ-OAR-2007-0121,
Docket ID EPA-HQ-OAR-2007-0121-0060.
21 Memo from Chris Lindhjem of ENVIRON, PMEmission Factors, December 5, 2005.
22 U.S. Environmental Protection Agency, Exhaust and Crankcase Emission Factors for Nonroad Engine Modeling -
Compression Ignition (April 2004). Appendix C, EPA- 420-P-04-009, available online at
http://www.epa.gov/otaq/models/nonrdmdl/nonrdmdl2004/420p04009.pdf, Docket ID EPA-HQ-OAR-2003-0190-0411.
23 Energy and Environmental Analysis Inc. (February 2000). Analysis of Commercial Marine Vessels Emissions and
Fuel Consumption Data, EPA420-R-00-002, available online at http://www.epa.gov/otaq/models/nonrdmdl/c-
marine/r00002.pdf.
24ICF International (March 2009). Inventory Contribution of U.S. Flagged Vessels, prepared for the U.S.
Environmental Protection Agency, EPA Report Number EPA-420-R-09-005.
25 Corbett, J. et al. (May 2006). Estimation, Validation and Forecasts of Regional Commercial Marine Vessel
Inventories, Tasks 1 and 2: Baseline Inventory and Ports Comparison, Final Report, prepared by University of
Delaware for the California Air Resource Board, Contract Number 04-346, and the Commission for Environmental
Cooperation in North America, Contract Number 113.111, May 2006, Docket ID EPA-HQ-OAR-2007-0121-0013.
26 Corbett, J. et al. (May 2006). Estimation, Validation and Forecasts of Regional Commercial Marine Vessel
Inventories, Tasks 1 and 2: Baseline Inventory and Ports Comparison, Final Report, prepared by University of
Delaware for the California Air Resource Board, Contract Number 04-346, and the Commission for Environmental
Cooperation in North America, Contract Number 113.111, May 2006, Docket ID EPA-HQ-OAR-2007-0121-0013.
27 Corbett, J. et al. (May 2006). Estimation, Validation and Forecasts of Regional Commercial Marine Vessel
Inventories, Tasks 1 and 2: Baseline Inventory and Ports Comparison, Final Report, prepared by University of
Delaware for the California Air Resource Board, Contract Number 04-346, and the Commission for Environmental
Cooperation in North America, Contract Number 113.111, May 2006, Docket ID EPA-HQ-OAR-2007-0121-0013.
28 Corbett, J. et al. (May 2006). Estimation, Validation and Forecasts of Regional Commercial Marine Vessel
Inventories, Tasks 1 and 2: Baseline Inventory and Ports Comparison, Final Report, prepared by University of
Delaware for the California Air Resource Board, Contract Number 04-346, and the Commission for Environmental
Cooperation in North America, Contract Number 113.111, May 2006, Docket ID EPA-HQ-OAR-2007-0121-0013.
29 U.S. Army Corps of Engineers Navigation Data Center (2002), Vessel Entrances and Clearances available at
http://www.iwr.usace.army.mil/ndc/db/entclrn/data/entrclrn02/
2-85
-------
30IMO. Revision of MARPOL Annex VI and the NOX technical code. Input from the four subgroups and individual
experts to the final report of the Informal Cross Government/Industry Scientific Group of Experts. BLG/INF.10
12/28/2007
31 Transport Canada; Transportation in Canada Annual Report 2004. 2004. (Tables 3-26 and 8-27).
http://www.tc.gc.ca/pol/en/report/anre2004/8F_e.htm.
32 RTI International (December 2006). Global Trade and Fuels Assessment - Future Trends and Effects of Designation
Requiring Clean Fuels in the Marine Sector: Task Order No. 1, Draft Report, prepared for the U.S. Environmental
Protection Agency, EPA Report Number EPA420-D-07-006, Docket ID EPA-HQ-OAR-2007-0121-0063.3.
33 RTI International (December 2006). Global Trade and Fuels Assessment - Future Trends and Effects of Designation
Requiring Clean Fuels in the Marine Sector: Task Order No. 1, Draft Report, prepared for the U.S. Environmental
Protection Agency, EPA Report Number EPA420-D-07-006, Docket ID EPA-HQ-OAR-2007-0121-0063.3.
34 Corbett, James and Chengfeng Wang (October 26, 2005). Emission Inventory Review SECA Inventory Progress
Discussion, p 11, memorandum to California Air Resources Board.
35 RTI International (December 2006). Global Trade and Fuels Assessment - Future Trends and Effects of Designation
Requiring Clean Fuels in the Marine Sector: Task Order No. 1, Draft Report, prepared for the U.S. Environmental
Protection Agency, EPA Report Number EPA420-D-07-006, Docket ID EPA-HQ-OAR-2007-0121-0063.3.
36 RTI International (December 2006). Global Trade and Fuels Assessment - Future Trends and Effects of Designation
Requiring Clean Fuels in the Marine Sector: Task Order No. 1, Draft Report, prepared for the U.S. Environmental
Protection Agency, EPA Report Number EPA420-D-07-006, Docket ID EPA-HQ-OAR-2007-0121-0063.3.
37 RTI International (December 2006). Global Trade and Fuels Assessment - Future Trends and Effects of Designation
Requiring Clean Fuels in the Marine Sector: Task Order No. 1, Draft Report, prepared for the U.S. Environmental
Protection Agency, EPA Report Number EPA420-D-07-006, Docket ID EPA-HQ-OAR-2007-0121-0063.3.
38 Starcrest Consulting Group (June 2004). Port-Wide Baseline Air Emissions Inventory, prepared for the Port of Los
Angeles
39ICF International, Commercial Marine Port Inventory Development, prepared for the U.S. Environmental Protection
Agency, EPA Report Number EPA-420-R-07-012c, October 2007, Docket ID EPA-HQ-OAR-2007-0121-0063.1.
40 ARCADIS Geraghty & Miller, Inc. (September 1999). Commercial Marine Activity for Deep Sea Ports in the United
States, prepared for the U.S. Environmental Protection Agency, EPA Report Number: EPA420-R-99-020, available
online at http://www.epa.gov/otaq/models/nonrdmdl/c-marine/r99020.pdf.
41 ARCADIS Geraghty & Miller, Inc. (September 1999). Commercial Marine Activity for Deep Sea Ports in the United
States, prepared for the U.S. Environmental Protection Agency, EPA Report Number: EPA420-R-99-020, available
online at http://www.epa.gov/otaq/models/nonrdmdl/c-marine/r99020.pdf.
42 Starcrest Consulting Group (January 2007). Draft Port of Los Angeles Air Emissions Inventory for Calendar Year
2005.
43 Starcrest Consulting Group (April 2007). Puget Sound Maritime Air Forum Maritime Air Emissions Inventory.
44 Starcrest Consulting Group, LLC (April 2003). The New York, Northern New Jersey, Long Island Nonattainment
Area Commercial Marine Vessel Emission Inventory, Vol 1 - Report, Prepared for the Port Authority of New York &
New Jersey, United States and the Army Corps of Engineers, New York District.
45 Starcrest Consulting Group, LLC (November 2000). Houston-Galveston Area Vessel Emissions Inventory, Prepared
for the Port of Houston Authority and the Texas Natural Resource Conservation Commission.
2-86
-------
46 Eastern Research Group and Starcrest Consulting Group, LLC (January 2004). Update To The Commercial Marine
Inventory For Texas To Review Emissions Factors, Consider A Ton-Mile El Method, And Revise Emissions For The
Beaumont-Port Arthur Non-Attainment Area Final Report, Submitted to the Houston Advanced Research Center.
47 Zuber M. Farooqui and Kuruvilla John (June 2004). Refinement of the Marine Emissions Inventory for the Corpus
Christ! Urban Airshed, Department of Environmental Engineering, Texas A&M University - Kingsville, Proceedings of
the 97th Annual A&WMA Conf. & Exhibition.
48ICF International (October 2007). Commercial Marine Port Inventory Development, prepared for the U.S.
Environmental Protection Agency, EPA Report Number EPA-420-R-07-012c, October 2007, Docket ID EPA-HQ-
OAR-2007-0121-0063.1.
49 ENVIRON International Corporation (March 2007). LADCO 2005 Commercial Marine Emissions.
50 California Air Resources Board (October 2005). Emissions Estimation Methodology for Ocean-Going Vessels.
51 E.H. Pechan & Associates Inc. (June 2005), Commercial Marine Inventories for Select Alaskan Ports, Final Report,
Prepared for the Alaska Department of Conservation.
52 Corbett, J. et al. (May 2006). Estimation, Validation and Forecasts of Regional Commercial Marine Vessel
Inventories, Tasks 1 and 2: Baseline Inventory and Ports Comparison, Final Report, prepared by University of
Delaware for the California Air Resource Board, Contract Number 04-346, and the Commission for Environmental
Cooperation in North America, Contract Number 113.111, May 2006, Docket ID EPA-HQ-OAR-2007-0121-0013.
53 Lloyd's Register and International Maritime Organization, Marine Exhaust Emission Quantification Study - Baltic
Sea, inMEPC45/INF.7. 1998.
54 Entec UK Limited (2002). Quantification of Emissions from Ships Associated with Ship Movements between Ports in
the European Community, prepared for the European Commission, Docket ID EPA-HQ-OAR-2007-0121-0059.
55 ICF International (January 5, 2006). Current Methodologies and Best Practices in Preparing Port Emission
Inventories, Final Report, prepared for the U.S. Environmental Protection Agency, available online at
http://www.epa.gov/sectors/ports/bp_portemissionsfinal.pdf.
56 Corbett, J.J. and H.W. Koehler (2003). Updated Emissions from Ocean Shipping, Journal of Geophysical Research,
108(020); p. 4650.
57 Corbett, J.J. and H.W. Koehler (2004). Considering Alternative Input Parameters in an Activity-Based Ship Fuel
Consumption and Emissions Model: Reply to Comment by Oyvind Endresen et al. on "Updated Emissions from Ocean
Shipping," Journal of Geophysical Research. 109(023303).
58 Levelton Consultants Ltd. (2006). Marine Emission Inventory Study Eastern Canada and Great Lakes - Interim
Report 4: Gridding Results, prepared for Transportation Development Centre, Transport Canada.
59 Entec UK Limited (2002). Quantification of Emissions from Ships Associated with Ship Movements between Ports in
the European Community, prepared for the European Commission, Docket ID EPA-HQ-OAR-2007-0121-0059.
60 California Air Resources Board (September 2005). 2005 Oceangoing Ship Survey, Summary of Results.
61 California Air Resources Board (September 2005). 2005 Oceangoing Ship Survey, Summary of Results.
2-87
-------
3 Impacts of Shipping Emissions on Air Quality,
Health and the Environment
Designation of this Emission Control Area will significantly reduce emissions of SOx,
NOx and PM2.5 and ambient levels of particulate matter and ground-level ozone in large
portions of the United States, which will result in substantial benefits to human health and the
environment. This chapter describes the pollutants which would be reduced due to the EGA
designation and their impacts on human health and ambient air quality as well as the impacts
of these pollutants on the environment. Appendix A to Chapter 3 describes the relevant
meteorological conditions within the proposed areas that contribute to at-sea emissions being
transported to populated areas and contributing to harmful human health and ecological
impacts. Appendix B to Chapter 3 presents the expected percent reduction in nitrogen and
sulfur deposition in 18 regions of the U.S. due to the proposed EGA.
3.1 Pollutants Reduced by the EGA and their Associated Health Impacts
3.1.1 Description of Pollutants
3.1.1.1 Particulate Matter
Particulate matter (PM) is a generic term for a broad class of chemically and
physically diverse substances. It can be principally characterized as discrete particles that
exist in the condensed (liquid or solid) phase spanning several orders of magnitude in size.
Since 1987, EPA has delineated that subset of inhalable particles small enough to penetrate to
the thoracic region (including the tracheobronchial and alveolar regions) of the respiratory
tract (referred to as thoracic particles). Current national ambient air quality standards
(NAAQS) use PM2.s as the indicator for fine particles (with PM2.s referring to particles with a
nominal mean aerodynamic diameter less than or equal to 2.5 pm), and use PMio as the
indicator for purposes of regulating the coarse fraction of PMio (referred to as thoracic coarse
particles or coarse-fraction particles; generally including particles with a nominal mean
aerodynamic diameter greater than 2.5 pm and less than or equal to 10 pm, or PMio-z.s).
Ultrafine particles are a subset of fine particles, generally less than 100 nanometers (0.1 nm)
in aerodynamic diameter.
Particles span many sizes and shapes and consist of hundreds of different chemicals.
Particles originate from sources and are also formed through atmospheric chemical reactions;
the former are often referred to as "primary" particles, and the latter as "secondary" particles.
In addition, there are also physical, non-chemical reaction mechanisms that contribute to
secondary particles. Particle pollution also varies by time of year and location and is affected
by several weather-related factors, such as temperature, clouds, humidity, and wind. A
further layer of complexity comes from a particle's ability to shift between solid/liquid and
gaseous phases, which is influenced by concentration, meteorology, and temperature.
Fine particles are produced primarily by combustion processes and by transformations
of gaseous emissions (e.g., NOx, SOx and VOCs) in the atmosphere. The chemical and
physical properties of PM2.s may vary greatly with time, region, meteorology, and source
3-1
-------
category. Thus, PM2.5 may include a complex mixture of different pollutants including
sulfates, nitrates, organic compounds, elemental carbon and metal compounds. These
particles can remain in the atmosphere for days to weeks and travel through the atmosphere
hundreds to thousands of kilometers.1
3.1.1.2 Ozone
Ground-level ozone pollution is formed by the reaction of VOCs and NOx in the
atmosphere in the presence of heat and sunlight. These pollutants, often referred to as ozone
precursors, are emitted by many types of pollution sources such as highway vehicles and
nonroad engines (including ships), power plants, chemical plants, refineries, makers of
consumer and commercial products, industrial facilities, and smaller area sources.
The science of ozone formation, transport, and accumulation is complex.2 Ground-
level ozone is produced and destroyed in a cyclical set of chemical reactions, many of which
are sensitive to temperature and sunlight. When ambient temperatures and sunlight levels
remain high for several days and the air is relatively stagnant, ozone and its precursors can
build up and result in more ozone than typically would occur on a single high-temperature
day. Ozone can be transported hundreds of miles downwind of precursor emissions, resulting
in elevated ozone levels even in areas with low VOC or NOx emissions.
The highest levels of ozone are produced when both VOC and NOx emissions are
present in significant quantities on clear summer days. Relatively small amounts of NOx
enable ozone to form rapidly when VOC levels are relatively high, but ozone production is
quickly limited by removal of the NOx. Under these conditions NOx reductions are highly
effective in reducing ozone while VOC reductions have little effect. Such conditions are
called "NOx-limited." Because the contribution of VOC emissions from biogenic (natural)
sources to local ambient ozone concentrations can be significant, even some areas where man-
made VOC emissions are relatively low can be NOx-limited.
Ozone concentrations in an area also can be lowered by the reaction of nitric oxide
(NO) with ozone, forming nitrogen dioxide (NOz); as the air moves downwind and the cycle
continues, the NOz forms additional ozone. The importance of this reaction depends, in part,
on the relative concentrations of NOx, VOC, and ozone, all of which change with time and
location. When NOx levels are relatively high and VOC levels relatively low, NOx forms
inorganic nitrates (i.e., particles) but relatively little ozone. Such conditions are called "VOC-
limited". Under these conditions, VOC reductions are effective in reducing ozone, but NOx
reductions can actually increase local ozone under certain circumstances. Even in VOC-
limited urban areas, NOx reductions are not expected to increase ozone levels if the NOx
reductions are sufficiently large.
Rural areas are usually NOx-limited, due to the relatively large amounts of biogenic
VOC emissions in such areas. Urban areas can be either VOC- or NOx-limited, or a mixture
of both, in which ozone levels exhibit moderate sensitivity to changes in either pollutant.
3-2
-------
3.1.1.3 NOxandSOx
Sulfur dioxide (802), a member of the sulfur oxide (SOx) family of gases, is formed
from burning fuels containing sulfur (e.g., coal or oil), extracting gasoline from oil, or
extracting metals from ore. Nitrogen dioxide (NOz) is a member of the nitrogen oxide (NOx)
family of gases. Most NOz is formed in the air through the oxidation of nitric oxide (NO)
emitted when fuel is burned at a high temperature.
SOz andNOz can dissolve in water vapor and further oxidize to form sulfuric and
nitric acid which reacts with ammonia to form sulfates and nitrates, both of which are
important components of ambient PM. The health effects of ambient PM are discussed in
Section 3.1.2.1. NOx along with non-methane hydrocarbons (NMHC) are the two major
precursors of ozone. The health effects of ozone are covered in Section 3.1.2.2.
3.1.1.4 Diesel Exhaust PM
Ship emissions contribute to ambient levels of air toxics known or suspected as human
or animal carcinogens, or that have noncancer health effects. The population experiences an
elevated risk of cancer and other noncancer health effects from exposure to air toxics.3 These
compounds include diesel PM.
Marine diesel engines emit diesel exhaust (DE), a complex mixture comprised of
carbon dioxide, oxygen, nitrogen, water vapor, carbon monoxide, nitrogen compounds, sulfur
compounds and numerous low molecular-weight hydrocarbons. A number of these gaseous
hydrocarbon components are individually known to be toxic including aldehydes, benzene
and 1,3-butadiene. The diesel particulate matter (DPM) present in diesel exhaust consists of
fine particles (< 2.5pm), including a subgroup with a large number of ultrafine particles (< 0.1
pm). These particles have a large surface area which makes them an excellent medium for
adsorbing organics, and their small size makes them highly respirable. Many of the organic
compounds present in the gases and on the particles, such as polycyclic organic matter
(POM), are individually known to have mutagenic and carcinogenic properties. Marine diesel
engine emissions consist of a higher fraction of hydrated sulfate (approximately 60-90%) due
to the higher sulfur levels of the fuel, organic carbon (approximately 15-30%), and metallic
ash (approximately 7-11%) than are typically found in land-based engines.4 In addition,
while toxic trace metals emitted by marine diesel engines represent a very small portion of the
national emissions of metals (less than one percent) and are a small portion of DPM
(generally much less than one percent of DPM), we note that several trace metals of potential
toxicological significance and persistence in the environment are emitted by diesel engines.5
These trace metals include chromium, manganese, mercury, and nickel. In addition, small
amounts of dioxins have been measured in highway engine diesel exhaust, some of which
may partition into the particulate phase. Dioxins are a major health concern but diesel engines
are a minor contributor to overall dioxin emissions.
Diesel exhaust varies significantly in chemical composition and particle sizes between
different engine types (heavy-duty, light-duty), engine operating conditions (idle, accelerate,
decelerate), and fuel formulations (high/low sulfur fuel). Also, there are emissions
differences between on-road and nonroad engines because the nonroad engines are generally
3-3
-------
of older technology. This is especially true for marine diesel engines.6 After being emitted in
the engine exhaust, diesel exhaust undergoes dilution as well as chemical and physical
changes in the atmosphere. The lifetime for some of the compounds present in diesel exhaust
ranges from hours to days.
3.1.2 Health Effects Associated with Exposure to Pollutants
3.1.2.1 PM Health Effects
This section provides a summary of the health effects associated with exposure to
ambient concentrations of PM.A The information in this section is based on the data and
conclusions in the PM Air Quality Criteria Document (PM AQCD) and PM Staff Paper
prepared by the U.S. Environmental Protection Agency (EPA).B'7'8 We also present
additional recent studies published after the cut-off date for the PM AQCD.9' Taken
together this information supports the conclusion that exposure to ambient concentrations of
PM are associated with adverse health effects. Information specifically related to health
effects associated with exposure to diesel exhaust PM is included in Section 3.1.2.5 of this
document.
3.1.2.1.1 Short-term Exposure Mortality and Morbidity Studies
As discussed in the PM AQCD, short-term exposure to PM2.5 is associated with
premature mortality from cardiopulmonary diseases, hospitalization and emergency
department visits for cardiopulmonary diseases,11 increased respiratory symptoms,
decreased lung function13 and physiological changes or biomarkers for cardiac changes.14 In
addition, the PM AQCD described a limited body of new evidence from epidemiologic
A Personal exposure includes contributions from many different types of particles, from many sources, and in
many different environments. Total personal exposure to PM includes both ambient and nonambient
components; and both components may contribute to adverse health effects.
B The PM NAAQS is currently under review and the EPA is considering all available science on PM health
effects, including information which has been published since 2004, in the development of the upcoming PM
Integrated Science Assessment Document (ISA). A first draft of the PM ISA was completed in December 2008
and was submitted for review by the Clean Air Scientific Advisory Committee (CASAC) of EPA's Science
Advisory Board. Comments from the general public have also been requested. For more information, see
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=201805.
c These additional studies are included in the 2006 Provisional Assessment of Recent Studies on Health Effects
of Particulate Matter Exposure. The provisional assessment did not and could not (given a very short timeframe)
undergo the extensive critical review by CASAC and the public, as did the PM AQCD. The provisional
assessment found that the "new" studies expand the scientific information and provide important insights on the
relationship between PM exposure and health effects of PM. The provisional assessment also found that "new"
studies generally strengthen the evidence that acute and chronic exposure to fine particles and acute exposure to
thoracic coarse particles are associated with health effects. Further, the provisional science assessment found
that the results reported in the studies did not dramatically diverge from previous findings, and taken in context
with the findings of the CD, the new information and findings did not materially change any of the broad
scientific conclusions regarding the health effects of PM exposure made in the CD. However, it is important to
note that this assessment was limited to screening, surveying, and preparing a provisional assessment of these
studies. For reasons outlined in Section I.C of the preamble for the final PM NAAQS rulemaking in 2006 (see
71 FR 61148-49, October 17, 2006), EPA based its decision on the science presented in the 2004 CD.
3-4
-------
studies for potential relationships between short term exposure to PM and health endpoints
such as low birth weight, preterm birth, and neonatal and infant mortality.15
Among the studies of effects associated with short-term exposure to PM2.5, several
specifically address the contribution of mobile sources to short-term PM2.5-related effects on
premature mortality. The results from these studies generally indicated that several
combustion-related fine particle source-types are likely associated with mortality, including
motor vehicle emissions as well as other sources.16 The analyses incorporate source
apportionment tools into short-term exposure studies and are briefly mentioned here.
Analyses incorporating source apportionment by factor analysis with daily time-series studies
of daily death rates indicated a relationship between mobile source PM2.s and
mortality.17'18'19'20 Another recent study in 14 U.S. cities examined the effect of PMio
exposures on daily hospital admissions for cardiovascular disease. This study found that the
effect of PMio was significantly greater in areas with a larger proportion of PMio coming from
motor vehicles, indicating that PMio from these sources may have a greater effect on the
toxicity of ambient PMio when compared with other sources.21 These studies provide
evidence that PM-related emissions, specifically from mobile sources, are associated with
adverse health effects.
3.1.2.1.2 Long-term Exposure Mortality and Morbidity Studies
Long-term exposure to ambient PM2.5 is associated with premature mortality from
cardiopulmonary diseases and lung cancer,22 and effects on the respiratory system such as
decreased lung function or the development of chronic respiratory disease.23 Of specific
importance, the PM AQCD also noted that the PM components of gasoline and diesel engine
exhaust represent one class of hypothesized likely important contributors to the observed
ambient PM-related increases in lung cancer incidence and mortality.24
The PM AQCD and PM Staff Paper emphasized the results of two long-term
epidemiologic studies, the Six Cities and American Cancer Society (ACS) prospective cohort
studies, based on several factors - the large air quality data set for PM in the Six Cities Study,
the fact that the study populations were similar to the general population, and the fact that
these studies have undergone extensive reanalysis.25'2 |27'2829' These studies indicate that
there are positive associations for all-cause, cardiopulmonary, and lung cancer mortality with
long-term exposure to PM2.5. One analysis of a subset of the ACS cohort data, which was
published after the PM AQCD was finalized but in time for the 2006 Provisional Assessment,
found a larger association than had previously been reported between long-term PM2.s
exposure and mortality in the Los Angeles area using a new exposure estimation method that
accounted for variations in concentration within the city.31
As discussed in the PM AQCD, the morbidity studies that combine the features of
cross-sectional and cohort studies provide the best evidence for chronic exposure effects.
Long-term studies evaluating the effect of ambient PM on children's development have
shown some evidence indicating effects of PM2.5 and/or PMio on reduced lung function
growth.32 In another recent publication included in the 2006 Provisional Assessment,
investigators in southern California reported the results of a cross-sectional study of outdoor
PM2.5 and a measure of atherosclerosis development in the Los Angeles basin.3 The study
3-5
-------
found significant associations between ambient residential PM2.5 and carotid intima-media
thickness (CIMT), an indicator of subclinical atherosclerosis, an underlying factor in
cardiovascular disease.
3.1.2.2 Ozone Health Effects
This section provides a summary of the health effects associated with ambient ozone.0
The information in this section is based on the data and conclusions in the ozone air quality
criteria document (ozone AQCD) and ozone staff paper prepared by the U.S. EPA.34' 5
Taken together this information supports the conclusion that ozone-related emissions are
associated with adverse health effects.
Ozone-related health effects include lung function decrements, respiratory symptoms,
aggravation of asthma, increased hospital and emergency room visits, increased asthma
medication usage, and a variety of other respiratory effects. Cellular-level effects, such as
inflammation of lungs, have been documented as well. In addition, there is suggestive
evidence of a contribution of ozone to cardiovascular-related morbidity and highly suggestive
evidence that short-term ozone exposure directly or indirectly contributes to non-accidental
and cardiopulmonary-related mortality, but additional research is needed to clarify the
underlying mechanisms causing these effects. In a recent report on the estimation of ozone-
related premature mortality published by the National Research Council (NRC), a panel of
experts and reviewers concluded that short-term exposure to ambient ozone is likely to
contribute to premature deaths and that ozone-related mortality should be included in
estimates of the health benefits of reducing ozone exposure.36 People who appear to be more
susceptible to effects associated with exposure to ozone include children, asthmatics and the
elderly. Those with greater exposures to ozone, for instance due to time spent outdoors (e.g.,
children and outdoor workers), are also of concern.
A large number of scientific studies have identified several key health effects
associated with exposure to levels of ozone found today in many areas of the United States.
Short-term (1 to 3 hours) and prolonged exposures (6 to 8 hours) to ambient ozone
concentrations have been linked to lung function decrements, respiratory symptoms, increased
hospital admissions and emergency room visits for respiratory problems.3 38'39'40'41'42
Repeated exposure to ozone can increase susceptibility to respiratory infection and lung
inflammation and can aggravate preexisting respiratory diseases, such as asthma.43'44' 5'46'47
Repeated exposure to sufficient concentrations of ozone can also cause inflammation of the
lung, impairment of lung defense mechanisms, and possibly irreversible changes in lung
structure, which over time could affect premature aging of the lungs and/or the development
of chronic respiratory illnesses, such as emphysema and chronic bronchitis.48'49'50'51
Children and adults who are outdoors and active during the summer months, such as
D Human exposure to ozone varies over time due to changes in ambient ozone concentration and because people
move between locations which have notable different ozone concentrations. Also, the amount of ozone
delivered to the lung is not only influenced by the ambient concentrations but also by the individuals breathing
route and rate.
3-6
-------
construction workers, are among those most at risk of elevated ozone exposures.52 Children
and outdoor workers tend to have higher ozone exposure because they typically are active
outside, working, playing and exercising, during times of day and seasons (e.g., the summer)
when ozone levels are highest.53 For example, summer camp studies in the Eastern United
States and Southeastern Canada have reported statistically significant reductions in lung
function in children who are active outdoors.54'55'56'57'5 '59'60'61 Further, children are more
at risk of experiencing health effects from ozone exposure than adults because their
respiratory systems are still developing. These individuals (as well as people with respiratory
illnesses, such as asthma, especially asthmatic children) can experience reduced lung function
and increased respiratory symptoms, such as chest pain and cough, when exposed to relatively
low ozone levels during prolonged periods of moderate exertion.62'63'64'65
3.1.2.3 SOX Health Effects
This section provides an overview of the health effects associated with S02.
Additional information on the health effects of S02 can be found in the U.S. Environmental
Protection Agency Integrated Science Assessment for Sulfur Oxides.66 Following an
extensive evaluation of health evidence from epidemiologic and laboratory studies, the U.S.
EPA has concluded that there is a causal relationship between respiratory health effects and
short-term exposure to S02. The immediate effect of S02 on the respiratory system in humans
is bronchoconstriction. This response is mediated by chemosensitive receptors in the
tracheobronchial tree. These receptors trigger reflexes at the central nervous system level
resulting in bronchoconstriction, mucus secretion, mucosal vasodilation, cough, and apnea
followed by rapid shallow breathing. In some cases, local nervous system reflexes also may
be involved. Asthmatics are more sensitive to the effects of S02 likely resulting from
preexisting inflammation associated with this disease. This inflammation may lead to
enhanced release of mediators, alterations in the autonomic nervous system and/or
sensitization of the chemosensitive receptors. These biological processes are likely to
underlie the bronchoconstriction and decreased lung function observed in response to S02
exposure. In laboratory studies involving controlled human exposures to S02, respiratory
effects have consistently been observed following 5-10 min exposures at S02 concentrations
> 0.2 ppm in asthmatics engaged in moderate to heavy levels of exercise. In these studies, 5-
30% of relatively healthy exercising asthmatics are shown to experience moderate or greater
decrements in lung function (> 100% increase in sRaw (specific airway resistance) or > 15%
decrease in FEVi (forced expiratory volume in 1 second)) with peak exposures to S02
concentrations of 0.2-0.3 ppm. At concentrations > 0.4 ppm, a greater percentage of
asthmatics (20-60%) experience S02-induced decrements in lung function, which are
frequently accompanied by respiratory symptoms. A clear concentration-response
relationship has been demonstrated in laboratory studies following exposures to S02 at
concentrations between 0.2 and 1.0 ppm, both in terms of increasing severity of effect and
percentage of asthmatics adversely affected.
In epidemiologic studies, respiratory effects have been observed in areas where the
mean 24-hour S02 levels range from 1 to 30 ppb, with maximum 1 to 24-hour average S02
values ranging from 12 to 75 ppb. Important new multicity studies and several other studies
have found an association between 24-hour average ambient S02 concentrations and
respiratory symptoms in children, particularly those with asthma. Furthermore, limited
3-7
-------
epidemiologic evidence indicates that atopic children and adults may be at increased risk for
S02-induced respiratory symptoms. Generally consistent associations also have been
observed between ambient S02 concentrations and emergency department visits and
hospitalizations for all respiratory causes, particularly among children and older adults (> 65
years), and for asthma. Intervention studies provide additional evidence that supports a causal
relationship between S02 exposure and respiratory health effects. Two notable studies
conducted in several cities in Germany and in Hong Kong reported that decreases in S02
concentrations were associated with improvements in respiratory symptoms, though the
possibility remained that these health improvements may be partially attributable to declining
concentrations of air pollutants other than S02, most notably PM or constituents of PM. A
limited subset of epidemiologic studies has examined potential confounding by copollutants
using multipollutant regression models. These analyses indicate that although copollutant
adjustment has varying degrees of influence on the S02 effect estimates, the effect of S02 on
respiratory health outcomes appears to be generally robust and independent of the effects of
gaseous and particulate copollutants, suggesting that the observed effects of S02 on
respiratory endpoints occur independent of the effects of other ambient air pollutants.
Consistent associations between short-term exposure to S02 and mortality have been
observed in epidemiologic studies, with larger effect estimates reported for respiratory
mortality than cardiovascular mortality. While this finding is consistent with the
demonstrated effects of S02 on respiratory morbidity, uncertainty remains with respect to the
interpretation of these associations due to potential confounding by various copollutants. The
U.S. EPA has therefore concluded that the overall evidence is suggestive of a causal
relationship between short-term exposure to S02 and mortality. Significant associations
between short-term exposure to S02 and emergency department visits and hospital admissions
for cardiovascular diseases have also been reported. However, these findings have been
inconsistent across studies and do not provide adequate evidence to infer a causal relationship
between S02 exposure and cardiovascular morbidity.
3.1.2.4 NOX Health Effects
This section provides an overview of the health effects associated with N02.
Additional information on the health effects of N02 can be found in the U.S. Environmental
Protection Agency Integrated Science Assessment (ISA) for Nitrogen Oxides.67 The U.S.
EPA has concluded that the findings of epidemiologic, controlled human exposure, and
animal toxicological studies provide evidence that is sufficient to infer a likely causal
relationship between respiratory effects and short-term N02 exposure.68 The ISA concludes
that the strongest evidence for such a relationship comes from epidemiologic studies of
respiratory effects including symptoms, emergency department visits, and hospital
admissions. 69 The effect estimates from U.S. and Canadian studies generally indicate that
ambient N02 is associated with a 2-20% increase in risks for emergency department visits and
hospital admissions. Risks associated with respiratory symptoms are generally higher.70
These epidemiologic studies are supported by evidence from experimental studies, in
particular by controlled human exposure studies that evaluate airway hyperresponsiveness in
asthmatic individuals.71 The ISA draws two broad conclusions regarding airway
responsiveness following N02 exposure.72 First, the ISA concludes that N02 exposure may
enhance the sensitivity to allergen-induced decrements in lung function and increase the
3-8
-------
allergen-induced airway inflammatory response at exposures as low as 0.26 ppm N02 for 30
minutes.73 Second, exposure to N02 has been found to enhance the inherent responsiveness
of the airway to subsequent nonspecific challenges in controlled human exposure studies.74
In general, small but significant increases in nonspecific airway responsiveness were observed
in the range of 0.2 to 0.3 ppm N02 for 30-minute exposures and at 0.1 ppm N02 for 60-
minute exposures in asthmatics. These conclusions are consistent with results from animal
toxicological studies which have detected 1) increased immume-mediated pulmonary
inflammation in rats exposed to house dust mite allergen following exposure to 5 ppm N02
for 3-hour and 2) increased responsiveness to non-specific challenges following sub-chronic
(6-12 weeks) exposure to 1 to 4 ppm N02.75 Enhanced airway responsiveness could have
important clinical implications for asthmatics since transient increases in airway
responsiveness following N02 exposure have the potential to increase symptoms and worsen
asthma control.76 Together, the epidemiologic and experimental data sets form a plausible,
consistent, and coherent description of a relationship between N02 exposures and an array of
adverse health effects that range from the onset of respiratory symptoms to hospital
admission.
Although the weight of evidence supporting a causal relationship is somewhat less
certain than that associated with respiratory morbidity, N02 has also been linked to other
health endpoints. For example, results from several large U.S. and European multi-city
studies and a meta-analysis study indicate positive associations between ambient N02
concentrations and the risk of all-cause (nonaccidental) mortality, with effect estimates
ranging from 0.5 to 3.6% excess risk in mortality per standardized increment (20 ppb for 24-
hour averaging time, 30 ppb for 1-hour averaging time).77 In general, the N02 effect
estimates were robust to adjustment for co-pollutants. In addition, generally positive
associations between short-term ambient N02 concentrations and hospital admissions or
emergency department visits for cardiovascular disease have been reported.78 A number of
epidemiologic studies have also examined the effects of long-term exposure to N02 and
reported positive associations with decrements in lung function and partially irreversible
decrements in lung function growth.79 Specifically, results from the California-based
Children's Health Study, which evaluated N02 exposures in children over an 8-year period,
demonstrated deficits in lung function growth.80 This effect has also been observed in
Mexico City, Mexico81 and in Oslo, Norway,82 with decrements ranging from 1 to 17.5 ml
per 20- ppb increase in annual N02 concentration. Animal toxicological studies may provide
biological plausibility for the chonic effects of N02 that have been observed in these
epidemiologic studies.83 The main biochemical targets of N02 exposure appear to be
antioxidants, membrane polyunsaturated fatty acids, and thiol groups. N02 effects include
changes in oxidant/antioxidant homeostasis and chemical alterations of lipids and proteins.
Lipid peroxidation has been observed at N02 exposures as low as 0.04 ppm for 9 months and
at exposures of 1.2 ppm for 1 week, suggesting lower effect thresholds with longer durations
of exposure. Other studies showed decreases in formation of key arachidonic acid
metabolites in mornings following N02 exposures of 0.5 ppm. N02 has been shown to
increase collagen synthesis rates at concentrations as low as 0.5 ppm. This could indicate
increased total lung collagen, which is associated with pulmonary fibrosis, or increased
collagen turnover, which is associated with remodeling of lung connective tissue.
Morphological effects following chonic N02 exposures have been identified in animal studies
3-9
-------
that link to these increases in collagen synthesis and may provide plausibility for the deficits
in lung function growth described in epidemiologic studies.84
3.1.2.5 Diesel Exhaust PM Health Effects
A large number of health studies have been conducted regarding diesel exhaust.
These include epidemiologic studies of lung cancer in groups of workers and animal studies
focusing on non-cancer effects. Diesel exhaust PM (including the associated organic
compounds which are generally high molecular weight hydrocarbons but not the more volatile
gaseous hydrocarbon compounds) is generally used as a surrogate exposure measure for
whole diesel exhaust.
Diesel exhaust has been found to be of concern by several groups worldwide including
the U.S. government. The IPCS (International Programme on Chemical Safety) has
established an environmental health criteria for diesel fuel and exhaust emissions. In this
criteria the IPCS recommends that for the protection of human health diesel exhaust emissions
should be controlled. The IPCS explicitly states that urgent efforts should be made to reduce
emissions, specifically of particulates, by changing exhaust train techniques, engine design
and fuel composition.85
3.1.2.5.1 Potential Cancer Effects of Exposure to Diesel Exhaust
The U.S. EPA's 2002 final "Health Assessment Document for Diesel Engine Exhaust"
(the EPA Diesel HAD) classified exposure to diesel exhaust as likely to be carcinogenic to
humans by inhalation at environmental exposures, in accordance with the revised draft
1996/1999 U.S. EPA cancer guidelines.86'87 In accordance with earlier U.S. EPA guidelines,
exposure to diesel exhaust would similarly be classified as probably carcinogenic to humans
(Group Bl).88'89 A number of other agencies (National Institute for Occupational Safety and
Health, the International Agency for Research on Cancer, the World Health Organization,
California EPA, and the U.S. Department of Health and Human Services) have made similar
classifications.90'91'92'93'94 The Health Effects Institute has prepared numerous studies and
reports on the potential carcinogenicity of exposure to diesel exhaust.95'96'97
More specifically, the U.S. EPA Diesel HAD states that the conclusions of the
document apply to diesel exhaust in use today including both onroad and nonroad engines
including marine diesel engines present on ships. The U.S. EPA Diesel HAD acknowledges
that the studies were done on engines with generally older technologies and that "there have
been changes in the physical and chemical composition of some DE [diesel exhaust]
emissions (onroad vehicle emissions) over time, though there is no definitive information to
show that the emission changes portend significant toxicological changes." In any case, the
diesel technology used for marine diesel engines typically lags that used for onroad engines
which have been subject to PM standards since 1998. Thus it is reasonable to assume that the
hazards identified from older technologies may be largely applicable to marine engines.
For the Diesel HAD, the U.S. EPA reviewed 22 epidemiologic studies on the subject
of the carcinogenicity of exposure to diesel exhaust in various occupations, finding increased
lung cancer risk, although not always statistically significant, in 8 out of 10 cohort studies and
3-10
-------
10 out of 12 case-control studies which covered several industries. Relative risk for lung
cancer, associated with exposure, ranged from 1.2 to 1.5, although a few studies show relative
risks as high as 2.6. Additionally, the Diesel HAD also relied on two independent meta-
analyses, which examined 23 and 30 occupational studies respectively, and found statistically
significant increases of 1.33 to 1.47 in smoking-adjusted relative lung cancer risk associated
with diesel exhaust. These meta-analyses demonstrate the effect of pooling many studies and
in this case show the positive relationship between diesel exhaust exposure and lung cancer
across a variety of diesel exhaust-exposed occupations.98'99'100
The U.S. EPA recently assessed air toxic emissions and their associated risk (the
National-Scale Air Toxics Assessment or NATA for 1996 and 1999), and concluded that
diesel exhaust ranks with other emissions that the national-scale assessment suggests pose the
greatest relative risk.101'102 This national assessment estimates average population inhalation
exposures to DPM for nonroad and on-highway sources. These are the sum of ambient levels
weighted by the amount of time people spend in each of the locations.
In summary, the likely hazard to humans together with the potential for significant
environmental risks leads us to conclude that diesel exhaust emissions from marine engines
present public health issues of concern.
3.1.2.5.2 Other Health Effects of Exposure to Diesel Exhaust
Noncancer health effects of acute and chronic exposure to diesel exhaust emissions are
also of concern. The Diesel HAD established an inhalation Reference Concentration (RfC)
specifically based on animal studies of diesel exhaust exposure. An RfC is defined by the
U.S. EPA as "an estimate of a continuous inhalation exposure to the human population,
including sensitive subgroups, with uncertainty spanning perhaps an order of magnitude,
which is likely to be without appreciable risks of deleterious noncancer effects during a
lifetime." The U.S. EPA derived the RfC from consideration of four well-conducted chronic
rat inhalation studies showing adverse pulmonary effects.103'104'105'106 The diesel RfC is
based on a "no observable adverse effect" level of 144 pg/m3 that is further reduced by
applying uncertainty factors of 3 for interspecies extrapolation and 10 for human variations in
sensitivity. The resulting RfC derived in the Diesel HAD is 5 pg/m3 for diesel exhaust, as
measured by DPM. This RfC does not consider allergenic effects such as those associated
with asthma or immunologic effects. There is growing evidence that exposure to diesel
exhaust can exacerbate these effects, but the exposure-response data is presently lacking to
derive an RfC. The Diesel HAD states, "With DPM [diesel particulate matter] being a
ubiquitous component of ambient PM, there is an uncertainty about the adequacy of the
existing DE [diesel exhaust] noncancer database to identify all of the pertinent DE-caused
noncancer health hazards" (p. 9-19).
While there have been relatively few human studies associated specifically with the
noncancer impact of exposure to DPM alone, DPM is a component of the ambient particles
studied in numerous epidemiologic studies. The conclusion that health effects associated with
ambient PM in general are relevant to DPM is supported by studies that specifically associate
observable human noncancer health effects with exposure to DPM. As described in the
Diesel HAD, these studies identified some of the same health effects reported for ambient
3-11
-------
PM, such as respiratory symptoms (cough, labored breathing, chest tightness, wheezing), and
chronic respiratory disease (cough, phlegm, chronic bronchitis and suggestive evidence for
decreases in pulmonary function). Symptoms of immunological effects such as wheezing and
increased allergenicity are also seen. Studies in rodents, especially rats, show the potential for
human inflammatory effects in the lung and consequential lung tissue damage from chronic
diesel exhaust inhalation exposure. The Diesel HAD concludes "that acute exposure to DE
[diesel exhaust] has been associated with irritation of the eye, nose, and throat, respiratory
symptoms (cough and phlegm), and neurophysiological symptoms such as headache,
lightheadedness, nausea, vomiting, and numbness or tingling of the extremities."107 There is
also evidence for an immunologic effect such as the exacerbation of allergenic responses to
known allergens and asthma-like symptoms.108'109'110
The Diesel HAD briefly summarizes health effects associated with ambient PM and
discusses the PM2.5 NAAQS. There is a much more extensive body of human data, which is
also mentioned earlier in the health effects discussion for PM2.s (Section 3.2.1.1 of this
document), showing a wide spectrum of adverse health effects associated with exposure to
ambient PM, of which diesel exhaust is an important component. The PM2.s NAAQS is
designed to provide protection from the non-cancer and premature mortality effects of PM2.s
as a whole.
3.1.2.5.3 Exposure to Diesel Exhaust PM
Exposure of people to diesel exhaust depends on their various activities, the time spent
in those activities, the locations where these activities occur, and the levels of diesel exhaust
pollutants in those locations. The major difference between ambient levels of diesel
particulate and exposure levels for diesel particulate is that exposure levels account for a
person moving from location to location, the proximity to the emission source, and whether
the exposure occurs in an enclosed environment.
Occupational exposures to diesel exhaust from mobile sources, including marine
diesel engines, can be several orders of magnitude greater than typical exposures in the non-
occupationally exposed population. Over the years, diesel particulate exposures have been
measured for a number of occupational groups resulting in a wide range of exposures from 2
to 1280 pg/m3 for a variety of occupations. As discussed in the Diesel HAD, the National
Institute of Occupational Safety and Health (NIOSH) has estimated a total of 1,400,000
workers are occupationally exposed to diesel exhaust from on-road and nonroad vehicles
including marine diesel engines.
3.1.2.5.3.1 Elevated Concentrations and Ambient Exposures in Mobile Source-
Impacted Areas
While occupational studies indicate that those working in closest proximity to diesel
exhaust experience the greatest health effects, recent studies are showing that human
populations living near large diesel emission sources such as major roadways, m rail yards,
12 and marine ports113 are also likely to experience greater exposure to PM and other
components of diesel exhaust than the overall population, putting them at a greater health risk.
3-12
-------
The percentage of total port emissions that come from ships varies by port. However, ships
contribute to the DPM concentrations at ports, and elsewhere, that influence exposures.
Regions immediately downwind of marine ports may experience elevated ambient
concentrations of directly-emitted PM2.s from diesel engines. Due to the nature of marine
ports, emissions from a large number of diesel engines are concentrated in a small area. A
recent study from the California Air Resources Board (CARB) evaluated air quality impacts
of diesel engine emissions within the Port of Long Beach and Los Angeles in California, one
of the largest ports in the U.S.114 The port study employed the ISCST3 dispersion model.
With local meteorological data used in the modeling , annual average concentrations of DPM
were substantially elevated over an area exceeding 200,000 acres. Because the Ports are
located near heavily-populated areas, the modeling indicated that over 700,000 people lived in
areas with at least 0.3 pg/m3 of port-related DPM in ambient air, about 360,000 people lived
in areas with at least 0.6 pg/m3 of DPM, and about 50,000 people lived in areas with at least
1.5 pg/m3 of ambient DPM emitted directly from the port. This port study highlights the
substantial contribution these facilities make to ambient concentrations of DPM in large,
densely populated areas.
Figure 3.1-1 provides an aerial shot of the Port of Long Beach and Los Angeles in California.
Figure 3.1-1 Aerial Shot - Port of LA and Long Beach, California
The U.S. EPA recently updated its initial screening-level analysis115'116 of selected
marine port areas to better understand the populations, including minority, low-income, and
3-13
-------
children, that are exposed to diesel participate matter (DPM) emissions from these facilities.E
The results of this study are discussed here and are also available in the public docket.117'118
This screening-level analysis focused on a representative selection of national marine
ports.F Of the 45 marine ports studied, the results indicate that at least 18 million people,
including a disproportionate number of low-income households, African-Americans, and
Hispanics, live in the vicinity of these facilities and are being exposed to ambient DPM levels
that are 2.0 pg/m3 and 0.2 pg/m3 above levels found in areas further from these facilities.
Considering only ocean-going marine engine DPM emissions, the results indicate that 6.5
million people are exposed to ambient DPM levels that are 2.0 pg/m3 and 0.2 pg/m3 above
levels found in areas further from these facilities. Because those populations exposed to DPM
emissions from marine ports are more likely to be low-income and minority residents, these
populations would benefit from the standards being proposed by the coordinated strategy. The
detailed findings of this study are available in the public docket.
With regard to children, this analysis shows that at least four million children live in
the vicinity of the marine ports studied and are also exposed to ambient DPM levels that are
2.0 pg/m3 and 0.2 pg/m3 above levels found in areas further from these facilities. Of the 6.5
million people exposed to DPM emissions from ocean-going vessel emissions, 1.7 million are
children. The age composition of the total affected population in the screening analysis
matches closely with the age composition of the overall US population. However, for some
individual facilities the young (0-4 years) appear to be over-represented in the affected
population compared to the overall US population. Detailed results for individual harbors are
presented in the Appendices of the memorandum in the docket.
As part of this study, a computer geographic information system was used to identify
the locations and boundaries of a the harbor areas, and determine the size and demographic
characteristics of the populations living near these facilities. These facilities are listed in
Table 3.1-1. Figures 3.1-2 and 3.1-3 provide examples of digitized footprints of the marine
harbor areas included in this study.
E This type of screening-level analysis is an inexact tool and not appropriate for regulatory decision-making; it is
useful in beginning to understand potential impacts and for illustrative purposes.
F The Agency selected a representative sample from the top 150 U.S. ports including coastal, inland, and Great
Lake ports.
3-14
-------
Table 3.1-1 Marine Harbor Areas
Baltimore, MD
Boston, MA
Charleston, SC
Chicago, IL
Cincinnati, OH
Cleveland, OH
Corpus Christ! , TX
Detroit, MI
Duluth-Superior, MN
Freeport, TX
Gary, IN
Helena, AR
Houston, TX
Lake Charles, LA
Long Beach, CA
Los Angeles, CA
Louisville, KY
Miami, FL
Mobile, AL
Mount Vernon, IN
Nashville, TN
New Orleans, LA
New York, NY
Oakland, CA
Panama City, FL
Paulsboro, NJ
Philadelphia, PA
Pittsburgh, PA
Port Arthur, TX
Port Everglades, FL
Port of Baton Rouge, LA
Port of Plaquemines, LA
Portland, ME
Portland, OR
Richmond, CA
Savannah, GA
Seattle, WA
South Louisiana, LA
St. Louis, MO
Tacoma, WA
Tampa, FL
Texas City, TX
Tulsa - Port of Catoosa, OK
Two Harbors, MN
Wilmington, NC
Figure 3.1-2 Digitized footprint of New York, NY harbor area.
3-15
-------
Figure 3.1-3 Digitized footprint of Portland, OR harbor area.
In order to better understand the populations that live in the vicinity of marine harbor
areas and their potential exposures to ambient DPM, concentration isopleths surrounding the
45 marine port areas were created and digitized for all emission sources at the marine port and
for ocean-going vessel Category 3 engine emissions only. The concentration isopleths of
interest were selected to correspond to two DPM concentrations above urban background, 2.0
pg/m3 and 0.2 pg/m3. The isopleths were estimated using the AERMOD air dispersion
model. Figures 3.1-4 and 3.1-5 provide examples of concentration isopleths surrounding the
New York, NY harbor area for all emission sources and for ocean-going vessel Category 3
only engine emissions, respectively.
3-16
-------
fc
PortoTN*«YolK. NY
Figure 3.1-4 Concentration isopleths of New York, NY harbor area resulting from all emission sources.
3-17
-------
8
Port or N«f« Tort*. NY
Figure 3.1-5 Concentration isopleths of New York, NY harbor area resulting from only Category 3
vessels.
The size and characteristics of populations and households that reside within the area
encompassed by the two DPM concentration isopleths were determined for each isopleth and
the demographic compositions were assessed, including age, income level, and race/ethnicity.
In summary, the screening-level analysis found that for the 45 U.S. marine ports
studied, al least 18 million people live in the vicinity of these facilities and are exposed to
ambient DPM levels from all port emission sources that are 2.0 pg/m3 and 0.2 pg/m3 above
those found in areas further from these facilities. If only Category 3 engine DPM emissions
are considered, then the number of people exposed is 6.5 million.
3.1.2.6 Alaska and Hawaii Health Effects
The U.S. air quality maps below do not show Alaska and Hawaii. This is because the
domain of the CMAQ model does not include these states. However ship emission
inventories for Alaska and Hawaii were developed and are included in the totals presented in
Section 7. Based on the inventory there are substantial ship emissions in the proposed EGA
areas around Alaska and Hawaii. These are also the areas where most of the states'
populations reside. Two of Alaska's three biggest population centers (Anchorage: population
3-18
-------
260,000 and Juneau: population 30,000) are on the southeastern coast and these 2 cities alone
are home to just under half of the entire state's population. In Hawaii, more than 99% of the
state's population lives in the proposed EGA area. Meteorological information in Section 6
suggests that these emissions affect air quality. Based on Canadian air quality modeling,
there would be significant air quality improvements for Eastern Alaska along the Canadian
border. Therefore, it is reasonable to expect ships are contributing to ambient air
concentrations of ozone and PM2.5 in Hawaii and Alaska, even though our modeling does not
allow us to quantify these effects.
3.2 Current and Projected Air Quality
Ships are currently contributing to ambient PM2.5 and ozone concentrations and their
contribution will continue to grow into the future as more stringent controls for onshore
emission sources take effect. In this section, we present information on PM2.5 and ozone
levels in the continental United States based on air quality modeling. We also discuss the air
quality modeling methodology and impacts from ships' emissions on air quality in Alaska and
Hawaii.
Due to the imprecise science of discerning human health effects that are due solely to
SOx versus its PM derivatives (i.e. sulphate particles) or to NOx versus its derivatives, ozone
and PM, the air quality and health impacts from exposure to direct SOx and NOx from ships
are not separately quantified here.
3.2.1 Current PMZ5 Levels
As described in Section 3.1.2, PM causes adverse health effects, and the U.S.
government has set national standards to protect against those health effects. There are two
U.S. national ambient air quality standards (NAAQS) for PM2.s: an annual standard (15
l^g/m3) and a 24-hour standard (35 ^ig/m3). The most recent revisions to these standards were
in 1997 and 2006. In 2005 the U.S. EPA designated nonattainment areas for the 2006 PM2.5
NAAQS (70 FR 19844, April 14, 2005).G
In addition to the U.S. government NAAQS for PM2.5, the World Health Organization
(WHO) has also set air quality guidelines for PM2.5.119 The 2005 WHO Air Quality
Guidelines (AQG) set for the first time a guideline value for particulate matter (PM).
Although the aim is to achieve the lowest concentrations possible, since no threshold for PM
has been identified below which no damage to health is observed, the annual mean PM2.5
AQG is 10 ng/m3 and the 24-hour mean PM2.5 AQG is 25 ^ig/m3.
The IMO, the U.S. government and individual states and local areas have already put
in place many PM2.s and PM2.s precursor emission reduction programs. However, ships are
significant contributors to PM2.5 in many areas and states will need additional reductions in a
timely manner to help them meet their air quality goals.
G A nonattainment area is defined in the Clean Air Act (CAA) as an area that is violating an ambient standard or
is contributing to a nearby area that is violating the standard.
3-19
-------
3.2.2 Projected PMZ5 Air Quality
Levels of PM2.s in the ambient air are expected to continue to be a problem into the
future. Without further action, emissions from ships will contribute a larger share to the
projected levels of PM2.5 as emissions from other sources decrease. In this section we present
information on projected levels of PM2.5 in 2020, ships' contribution to these levels, and the
improvements which would occur with the proposed EGA.
3.2.2.1 Projected PM2.5 Levels without an EC A
Figure 3.2-1 presents the projected annual average PM2.5 concentrations for the
continental U.S.H based on the inventory projections described in Section 2.7.l Most of the
U.S. is projected to have annual average PM2.s levels between 5 and 12 pg/m3 with a few
areas having higher levels and some areas in the west having lower levels.
Figure 3.2-1 Annual Average PM2.g Concentrations in 2020 without an ECA
Even with the implementation of all current U.S. state and federal regulations, there
are projected to be many areas in the U.S. with levels of PM2.5 which are above health
H As discussed in Section 3.2.5.1.2 the air quality modeling domain only covers the continental United States.
1 As discussed in Section 2.7 the inventories used for the air quality modeling differ slightly from those used in
the final inventory calculations. The difference is small and was due to an error in calculating the distances and
the fact that the air quality modeling only included Tier I NOX controls in the baseline.
3-20
-------
standards.J Emission reductions from the EGA designation will be helpful for states and
counties in attaining and maintaining the PM2.s NAAQS and the WHO AQG.
3.2.2.2 Contribution of Ships to Projected PM2.5 Levels
Emissions of NOX, SOX and direct PM2.5 from ships have a significant impact on
ambient PM2.5 concentrations. The contribution from ships were determined by comparing
model results in two future year control runs, one with all sources and one without ships.
Figure 3.2-2 illustrates the projected percentage contribution of ships to annual average PM2.5
concentrations in 2020. The percentage contribution of ships to annual average PM2.s
concentrations is projected to be greater than 15% in parts of southern FL, southern LA, and
the northern and southern Pacific coastline. The impact of ship emissions on PM2.s
concentrations also extends well beyond the U.S. coastlines. As can be seen in Figure 3.2-2
the projected contribution of ships to annual average PM2.s concentrations in many inland
areas, such as Tennessee, Nevada, New York and Pennsylvania, is up to 2%.
The absolute contribution of ships to ambient PM2.5 levels is shown in Figure 3.2-3.
This shows that the contribution from ships to annual average PM2.5 concentrations is
projected to be greater than 3 pg/m3 for highly populated portions of southern California,
while both southern Louisiana and Florida are projected to show impacts greater than 1.5
pg/m3.
Perwrti Chanae u Annual Avaraga PM2 5
Kfl Ultra wfl« JKM r-
Figure 3.2-2 Percentage Contribution of Ships to Annual Average PM2.g Concentrations in 2020
J See Chapter 5, Section 5.4 for more information about existing emission reduction programs to control land-
based and other marine sources.
3-21
-------
Change in Annual Average PM2.5
2020 C3 zero oul vs 2020 basecase
Figure 3.2-3 Absolute Contribution of Ships to Annual Average PM2.s Concentrations in 2020
The modeling projections clearly show that ships affect air quality far inland on all the
U.S. coastlines. This is to be expected since ships operate along all the U.S. coastlines. It can
be concluded from looking at these results that emissions from ships need to be controlled in
order to achieve PM2.5 reductions, even in inland areas and areas without ports.
3.2.2.3 Projected PM2.5 Levels with an ECA
The impacts of the proposed ECA were determined by comparing the model results in
the 2020 control run against the baseline simulation of the same year. According to air
quality modeling performed for this analysis, the emission standards are expected to provide
significant nationwide improvements in PM2.5 levels.
Figures 3.2-4 and 3.2-5 present the projected percentage and absolute PM2.s
improvements in 2020 if an ECA were enacted 200 nm from the U.S. shoreline. Similar to
Figures 3.2-2 and 3.2-3, the PM2.5 improvements extend well inland including southern
California, the cities of Birmingham, AL and Atlanta, GA and the northeast corridor. The
entire U.S. coastline will experience large improvements in their air quality from the proposed
ECA.
3-22
-------
Pertent Changs in Annual Average PM2.i
210) Jt*>n SDi
Figure 3.2-4 Percent Improvement in Annual Average PM2.5 Concentrations in 2020 Resulting from the
Application of the Proposed ECA
Change in Annual Average PM2 5
2020 200nm ECA vs 2020 bs
Figure 3.2-5 Absolute Improvement in Annual Average PM2.5 Concentrations in 2020 Resulting from the
Application of the Proposed ECA
3-23
-------
3.2.3 Current Ozone Levels
As described in Section 3.1.2, ozone causes adverse health effects, and the U.S.
government has set national standards to protect against those health effects. The U.S. EPA
has recently amended the ozone NAAQS (73 FR 16436, March 27, 2008). The final 2008
ozone NAAQS rule addresses revisions to the previous 1997 NAAQS for ozone to provide
increased protection of public health and welfare. The 1997 8-hour ozone NAAQS was set at
0.08 ppm (effectively 0.084 ppm). In 2008 the U.S. EPA revised the level of the 8-hour
standard to 0.075 parts per million (ppm), expressed to three decimal places.
In addition to the U.S. government NAAQS for ozone, the WHO has also set an AQG
for ozone of 100 pg/m3 for an 8-hour mean.120 Comparing the WHO AQG to the U.S.
NAAQS requires converting pg/m3 to ppb and assuming a temperature of 20° Celsius and an
atmospheric pressure of 1013 mb. The conversion is approximately a factor of 2, meaning
that the AQG for ozone is approximately 50 ppb.K'121'
The IMO, the U.S. government and individual states and local areas have already put
into place many programs to reduce ozone precursors. However, ships are significant
contributors to ozone in many areas and states will need additional reductions in a timely
manner to help them meet their air quality goals.
3.2.4 Projected Ozone Air Quality
Levels of ozone in the ambient air are expected to continue to be a problem into the
future. Without further action, emissions from ships will contribute a larger share to the
projected levels of ozone as emissions from other sources decrease. In this section we present
information on projected levels of ozone in 2020, ships' contribution to these levels and the
improvements which would occur with an EGA.
3.2.4.1 Projected Ozone Levels without an ECA
Figure 3.2-6 presents the projected seasonal average of daily 8-hour maximum ozone
concentrations for the continental U.S. based on the inventory projections described in
Section 2.4L Concentrations over most of the U.S. are in the 40 to 50 ppb range with a few
scattered areas being lower, 30 to 40 ppb, or higher, up to > 70 ppb.
K The definition for standard temperature and pressure varies but both the U.S. EPA and the National Institute of
Standards and Technology use 20° Celsius and an atmospheric pressure of 1013 mb.
L As discussed in Section 3.2.5.1.2 the air quality modeling domain only covers the continental United States.
3-24
-------
Seasonal Avg S-hr Ozone (2020ce.i
Figure 3.2-6 Seasonal Average of Daily 8-hour Maximum Ozone Concentrations in 2020 without an ECA
Even with the implementation of all current U.S. state and federal regulations,
including the Acid Rain program and the NOx SIP call which target SOx and NOx emissions
that cause air quality issues far from power plants, nonroad and on-road diesel rules and the
Tier II rule for highway vehicles, there are projected to be many areas in the U.S. with levels
of ozone which are above health standards. Emission reductions from the ECA designation
would be helpful for U.S. states and counties in attaining and maintaining the ozone NAAQS
andWHOAQG.
3.2.4.2 Contribution of Ships to Projected Ozone Levels
Emissions of NOx from ships have a significant impact on ambient ozone
concentrations. The contribution from ships were determined by comparing model results in
two future year control runs, one with all sources and one without ships. Figure 3.2-7
illustrates the projected percentage contribution of ships to average daily maximum 8-hour
ozone concentrations in 2020. The percentage contribution of ships to average daily
maximum 8-hour ozone concentrations is projected to be between 5 and 15% throughout the
gulf coast, the pacific coast and the southern east coast, with southern California experiencing
contributions from ships of greater than 15%. The impacts of ship emissions on ozone
concentrations would extend well inland, diminishing with distance from a coast. As can be
M See Chapter 5, Section 5.4 for more information about existing emission reduction programs to control land-
based and other marine sources.
3-25
-------
seen in Figure 3.2-7, the projected contribution of ships to ozone concentrations in many
inland areas is up to 2%.
The absolute contribution of ships to 8-hour ozone concentrations is shown in Figure
3.2-8. This shows that the contribution from ships to 8-hour ozone concentrations is projected
to be greater than 0.2 ppb for much of the country, while most coastal areas are projected to
show impacts greater than 2.0 ppb.
The modeling projections clearly show that ships affect air quality on all the U.S.
coastlines. This is to be expected since ships operate along all the U.S. coastlines. It can be
concluded from looking at these results that emissions from ships need to be controlled in
order to achieve ozone reductions, even in inland areas and areas without ports.
Percent tfianqe in aummet-season average 3-hair max crone
Figure 3.2-7 Percentage Contribution of Ships to Summertime Maximum 8-hour Average Ozone
Concentrations in 2020
3-26
-------
Change in summer-season average 8-hour max ozone
2020 C3 tern out us 2020 basei
Figure 3.2-8 Absolute Contribution of Ships to Summertime Maximum 8-hour Average Ozone
Concentrations in 2020
3.2.4.3 Projected Ozone Levels with an ECA
The impacts of the proposed ECA were determined by comparing the model results in
the 2020 control run against the baseline simulation of the same year. According to air
quality modeling performed for this analysis, the emission standards are expected to provide
significant nationwide improvements in ozone levels.
Figures 3.2-9 and 3.2-10 present the projected percentage and absolute summertime
maximum 8-hour average ozone improvements in 2020 if an ECA were enacted 200 nm from
the U.S. shoreline. The ozone improvements are significant and extend inland including the
states of Arizona, Missouri, Kentucky, Pennsylvania and New York. The entire U.S.
coastline will experience improvements in their air quality from an ECA designation.
3-27
-------
Percent ctiange in summer-season avareaa 8-tour max o:ope
Figure 3.2-9 Percent Improvement in Summertime Maximum 8-hour Average Ozone Concentrations in
2020 Resulting from the Application of the Proposed ECA
Change in summer-season average 8-hour max ozone
2020 200nm ECA vs 2020 bas<
Figure 3.2-10 Absolute Improvement in Summertime Maximum 8-hour Average Ozone Concentrations in
2020 Resulting from the Application of the Proposed ECA
While the ECA designation would reduce ozone levels generally and provide national
ozone-related health benefits, this is not always the case at the local level. The air quality
modeling projects that in a few areas ozone levels will get higher because of the NOX
3-28
-------
disbenefit phenomenon. Due to the complex photochemistry of ozone production, NOx
emissions lead to both the formation and destruction of ozone, depending on the relative
quantities of NOx, VOC, and ozone formation catalysts such as the OH and HOz radicals. In
areas dominated by fresh emissions of NOx, ozone catalysts are removed via the production
of nitric acid which slows the ozone formation rate. Because NOx is generally depleted more
rapidly than VOC, this effect is usually short-lived and the emitted NOx can lead to ozone
formation later and further downwind. The terms "NOx disbenefits" or "ozone disbenefits"
refer to the ozone increases that result when reducing Ox emissions in localized areas.
According to the NARSTO Ozone Assessment, disbenefits are generally limited to small
regions within specific urban cores and are surrounded by larger regions in which NOx
control is beneficial.123 It is important to note the following as well: there is a level of NOx
control where enough NOx will have been reduced to result in decreases in ambient ozone
concentrations, this modeling does not include future VOC or NOx controls that local areas
are planning, and reductions in NOx are not only important to help reduce ozone but also to
help reduce PM2.5.
3.2.5 Air Quality Modeling Methodology
When considering the potential effects of any particular air quality regulation, it is
common practice to apply a photochemical air quality modeling system to estimate the change
in air quality expected to occur with the emissions reductions proposed as part of the control
program. At their root level, air quality models are quantitative approximations of the
numerous complex physical and chemical interactions in the atmosphere that determine the
formation and fate of air pollutants in the atmosphere. The U.S. government has traditionally
used air quality modeling results to support policy decisions and as inputs into regulatory
impact analyses. As part of this exercise, we have completed several air quality modeling
simulations to look at the impact of a potential EGA application on future air pollution levels
over the United States.
This section of the document describes the air quality modeling performed by the U.S.
government in support of the EGA application. A fine-scale, national air quality modeling
analysis was performed to estimate the effect in 2020 of the proposed EGA emissions
reductions on future: 8-hour ozone concentrations, annual fine particulate matter (PMz.s)
concentrations, visibility levels, and acid deposition to watersheds and ecosystems. The
following text will describe: the air quality model that was used, how it was applied, how the
model inputs were developed, how the model was evaluated, and for what scenarios it was
applied.
3.2.5.1 Modeling Methodology
For this analysis, we used a 2002-based, multi-pollutant modeling platform to assess
the impacts of reduced marine emissions from the application of an EGA. This platform
represents a structured system of connected modeling-related tools and data that provide a
consistent and transparent basis for assessing the air quality response to changes in emissions,
meteorology, and/or model formulation. The base year of data used to construct this platform
includes emissions and meteorology for 2002. The platform was developed by the U.S.
EPA's Office of Air Quality Planning and Standards in collaboration with the Office of
3-29
-------
Research and Development and is intended to support a variety of regulatory and research
model applications and analyses.
There are four key elements to the modeling platform, all of which will be described
in more detail in subsequent sections. The key elements are:
• the selected air quality model;
• the emissions, meteorological, and initial and boundary concentration data which are
input to the model;
• the emissions and meteorological models (or pre-processors) used to prepare the input
data in the form and format needed for air quality model simulations; and
• the predicted concentration and deposition values predicted by the model.
3.2.5.1.1 Air Quality Model
The CMAQ modeling system is a non-proprietary comprehensive three-dimensional,
grid-based Eulerian air quality model designed to estimate the formation and fate of oxidant
precursors, primary and secondary PM concentrations and deposition, over regional and urban
spatial scales for given input sets of meteorological conditions and emissions. 24'125'126
CMAQ is a publicly available, peer reviewedN, state-of-the-science model consisting of a
number of science attributes that are critical for simulating the oxidant precursors and non-
linear organic and inorganic chemical relationships associated with the formation of sulfate,
nitrate, and organic aerosols. CMAQ also simulates the transport and removal of directly
emitted particles which are speciated as elemental carbon, crustal material, nitrate, sulfate,
and organic aerosols. The CMAQ model version 4.6 was most recently peer-reviewed in
February of 2007 for the U.S. EPA as reported in the "Third Peer Review of the CMAQ
Model."127 The CMAQ model is a well-known and well-respected tool and has been used in
numerous national and international applications.128'129'130
The CMAQ modeling system is designed as an "open system" where new scientific
algorithms and mechanisms can be utilized and evaluated in conjunction with CMAQ
processes. Model parameterizations may also be modified to test performance characteristics
of dynamical-chemical processes within model simulations, such as tropospheric ozone,
visibility, acid deposition, and particulate matter. CMAQ offers a multi-pollutant (i.e., ozone,
particulates, acid deposition, and nitrogen loading) capability via a generalized chemistry
mechanism, general numerical solver, and comprehensive description of gaseous and aqueous
chemistry and modal aerosol dynamics. CMAQ was also designed with scaleable
atmospheric dynamics and generalized coordinates to address multi-scale capabilities (e.g.
regional or local scale) depending on a user-defined model resolution. To resolve
atmospheric dynamics at local scales, CMAQ utilizes a set of governing equations for
compressible non-hydrostatic atmospheres expressed in a generalized coordinate system. The
N Community Modeling & Analysis System (CMAS) - Reports from the CMAQ Review Process can be found
at: http://www.cmascenter.0rg/r and d/cmaq review process.cfm?temp id=99999 .
3-30
-------
generalized coordinate system allows various vertical coordinates and map projections to be
used and resolves the necessary grid and coordinates transformations.
This 2002 multi-pollutant modeling platform used the latest publicly-released CMAQ
version 4.6° with a few minor changes and new features made internally by the U.S. EPA
CMAQ model developers, all of which reflects updates to earlier versions in a number of
areas to improve the underlying science. The model enhancements in CMAQ v4.6.1 include:
1) an in-cloud sulfate chemistry module that accounts for the nonlinear sensitivity of
sulfate formation to varying pH;
2) an improved vertical asymmetric convective mixing module (ACM2) that allows
in-cloud transport from a source layer to all other-in cloud layers (combined non-local and
local closure scheme);
3) a heterogeneous reaction involving nitrate formation (gas-phase reactions involving
N205andH20);
4) the heterogeneous NzOs reaction probability is now temperature- and humidity-
dependent,
5) an updated version of the ISORROPIA aerosol thermodynamics module including
improved representation of aerosol liquid water content and correction in activity coefficients
for temperature other than 298K, and
6) an updated gas-phase chemistry mechanism, Carbon Bond 05 (CB05) and
associated Euler Backward Iterative (EBI) solver, with extensions to model explicit
concentrations of air toxic species.
3.2.5.1.2 Air Quality Model Domain and Configuration
The CMAQ modeling analyses were performed for three separate domains, as shown
in Figure 3.2-11. This modeling used a parent horizontal grid of 36 km with two nested,
finer-scale 12 km grids covering the Eastern and Western U.S. (i.e., EUS and WUS grids
respectively).P'Q The model extends vertically from the surface to 100 millibars using a
sigma-pressure coordinate system. Air quality conditions at the outer boundary of the 36 km
domain were taken from the global GEOS-Chem model and did not change over the
simulated scenarios. The 36 km grid was only used to establish the incoming air quality
concentrations along the boundaries of the 12 km grids. All of the modeling results assessing
the air quality impacts of emissions reductions from the application of EGA controls were
0 CMAQ version 4.6 was released on September 30, 2006. It is available from the Community Modeling and
Analysis System (CMAS) as well as previous peer-review reports at: http://www.cmascenter.org.
p We were unable to consider effects beyond the 48-State area due to the unavailability of gridded
meteorological data for locations like Alaska and Hawaii.
Q In the overlapping portion of the two fine grids we used the WUS results for the States of MT, WY, CO, and
NM, and the EUS results for ND, MN, SD, IA, NE, MO, KS, OK, and TX.
3-31
-------
taken from the 12 km grids. Table 3.2-1 provides some basic geographic information
regarding the CMAQ domains. Table 3.2-2 provides information on the vertical structure of
the CMAQ modeling as well as the model which provided meteorological inputs. Table 3.2-3
indicates which CMAQ configuration options were chosen for this analysis.
Figure 3.2-11. Map of the CMAQ Modeling Domains. The black outer box denotes the 36 km national
modeling domain; the red inner box is the 12 km western U.S. fine grid; and the blue inner box is the 12
km eastern U.S. fine grid.
Table 3.2-1. Geographic Elements of Domains used in the ECA Modeling.
CMAQ MODELING CONFIGURATION
Map Projection
Grid Resolution
Coordinate Center
True Latitudes
Dimensions
Vertical extent
National Grid
Western U.S. Fine Grid
Eastern U.S. Fine Grid
Lambert Conformal Projection
36km
12km
12km
97degW,40degN
33 deg N and 45 deg N
148x112x14
213x192x14
279x240x14
14 Layers: Surface to 100 millibar level (see Table 3-XX)
3-32
-------
Table 3.2-2. Vertical Layer Structure for MM5 and CMAQ (heights are layer top).
CMAQ
LAYERS
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
MM5 LAYERS
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
SIGMA P
1.000
0.995
0.990
0.985
0.980
0.970
0.960
0.950
0.940
0.930
0.920
0.910
0.900
0.880
0.860
0.840
0.820
0.800
0.770
0.740
0.700
0.650
0.600
0.550
0.500
0.450
0.400
0.350
0.300
0.250
0.200
0.150
0.100
0.050
0.000
APPROXIMAT
E HEIGHT (M)
0
38
77
115
154
232
310
389
469
550
631
712
794
961
1,130
1,303
1,478
1,657
1,930
2,212
2,600
3,108
3,644
4,212
4,816
5,461
6,153
6,903
7,720
8,621
9,625
10,764
12,085
13,670
15,674
APPROXIMATE
PRESSURE (MB)
1000
995
991
987
982
973
964
955
946
937
928
919
910
892
874
856
838
820
793
766
730
685
640
595
550
505
460
415
370
325
280
235
190
145
100
Table 3.2-3. Additional Details Regarding the CMAQ Model Configuration.
GAS-PHASE CHEMICAL
MECHANISMKRER
Gas-Phase Chemical Solver
PM Module
Inorganic PM module
CB05
Euler Backward Iterative (EBI) scheme
AER04 aerosol module which contains mechanisms
dealing with sea salt emissions. Three-mode approach:
One coarse mode, two fine modes with variable standard
deviations.
ISORROPIA
3-33
-------
Organic PM module
Advection Scheme
(vertical and horizontal)
Planetary Boundary Layer
Scheme
Dry Deposition
Aqueous Chemistry
Cloud Scheme
Vertical Coordinate
Updated SOA module based on Odum/Griffin et al.,
(1997, 1999)
Piecewise Parabolic Method (PPM)
Asymmetric Convective Mixing module (ACM 2) scheme
which permits gradual layer-by-layer downward mixing
through compensatory subsidence
M3DRY module modified RADM scheme
RADM Bulk scheme
RADM Cloud scheme
Terrain-following Sigma coordinate
The 36 km and both 12 km CMAQ modeling domains were modeled for the entire
year of 2002. We also modeled ten days at the end of December 2001 as a model "ramp up"
period. These days are used to minimize the effects of initial conditions and are not
considered as part of the output analyses. All 365 model days were used in the calculations of
the EGA impacts on annual average levels of PM2.s. For the 8-hour ozone results, we only
used the modeling results from the period between May 1 and September 30, 2002. This 153-
day period generally conforms to the ozone season across most parts of the U.S. and contains
the majority of days with observed high ozone concentrations in 2002.
3.2.5.1.3 Air Quality Model Inputs
The key inputs to the CMAQ model include emissions from anthropogenic and
biogenic sources, meteorological data describing atmospheric states and motions, and initial
and boundary conditions. A summary of these three modeling components are described
below.
3.2.5.1.3.1 Emissions Inventory Data Inputs
With the exception of the marine emissions discussed in Section 2 of this document,
the CMAQ gridded 2002 emissions input data were based on emissions from the 2002
National Emissions Inventory (NEI) version 3.0. This inventory includes emissions of criteria
pollutantsR from point, stationary area, and mobile source categories. With the exception of
California5, monthly onroad and nonroad emissions were generated from the National Mobile
Inventory Model (NMIM) using versions of MOBILE6.0 and NONROAD2005 consistent
with recent national rule analyses.T'u The 2002-based platform and its associated chemical
R Criteria pollutant emissions include sulfur dioxide, oxides of nitrogen, carbon monoxide, volatile organic
compounds, ammonia, and fine particles.
s The California Air Resources Board submitted annual emissions for California. These were allocated to
monthly resolution prior to emissions modeling using data from the National Mobile Inventory Model (NMIM).
T MOBILES version was used in the Mobile Source Air Toxics Rule: Regulatory Impact Analysis for Final Rule:
Control of Hazardous Air Pollutants from Mobile Sources, U.S. Environmental Protection Agency, Office of
3-34
-------
mechanism (CB05) employs updated speciation profiles using data included in the
SPECIATE4.0 database/ The 2002-based platform also incorporates several temporal
profile updates for both mobile and stationary sources.
The 2002-based platform includes emissions for a 2002 base year model evaluation
case, a 2002 base case and a 2020 future base case. The model evaluation case uses
prescribed burning and wildfire emissions specific to 2002, which were developed and
modeled as day-specific, location-specific emissions using an updated version of Sparse
Matrix Operator Kernel Emissions (SMOKE) system, version 2.3, which computes plume rise
and vertically allocates the fire emissions. SMOKE also provides mobile, area, and point
source emissions as gridded, temporalized, and speciated data inputs to CMAQ (Houyoux and
Vukovich, 1999).131 The 2002 evaluation case also includes continuous emissions monitoring
(CEM) data for 2002 for electric generating units (EGUs) with CEMs. The 2002 and
projection year baselines include multi-year averages for the fire sector and ECU emissions
that are temporally allocated based on a combination of multi-year average and 2002 temporal
profiles. Projections from 2002 were developed to account for the expected impact of
national regulations, consent decrees or settlements, known plant closures, and, for some
sectors, activity growth. Biogenic emissions were processed using the Biogenic Emissions
Inventory System (BEIS) version 3.13.
3.2.5.1.3.2 Meteorological Data Inputs
The CMAQ gridded meteorological input data for the entire year of 2002 were derived
from simulations of the Pennsylvania State University / National Center for Atmospheric
Research Mesoscale Model. This model, commonly referred to as MM5, is a limited-area,
nonhydrostatic, terrain-following system that solves for the full set of physical and
thermodynamic equations which govern atmospheric motions.132 Meteorological model input
fields were prepared separately for each of the domains shown in Figure 3.2-11 above. The
36 km national domain was modeled using MM5 v.3.6.0 and the 12 km Eastern U.S grid was
modeled with MM5 v3.7.2. Both of these two sets of meteorological inputs were developed
by the U.S. EPA. For the 12 km western U.S. grid, we utilized existing MM5 meteorological
model data prepared by the Western Regional Air Partnership.133 All three sets of MM5
model runs were conducted in 5.5 day segments with 12 hours of overlap for spin-up
purposes. Additionally, all three domains contained 34 vertical layers with an approximately
38 m deep surface layer and a 100 millibar top. The MM5 and CMAQ vertical structures are
shown in Table 3.2-2 and do not vary by horizontal grid resolution.
Transportation and Air Quality, Assessment and Standards Division, Ann Arbor, MI 48105, EPA420-R-07-002,
February 2007.
u NONROAD2005 version was used in the proposed rule for small spark ignition (SI) and marine SI rule: Draft
Regulatory Impact Analysis: Control of Emissions from Marine SI and Small SI Engines, Vessels, and
Equipment, U.S. Environmental Protection Agency, Office of Transportation and Air Quality, Office of
Transportation and Air Quality, Assessment and Standards Division, Ann Arbor, MI, EPA420-D-07-004, April
2007.
v See http://www.epa.gov/ttn/chief/software/speciate/index.html for more details.
3-35
-------
The meteorological outputs from MM5 were processed to create model-ready inputs
for CMAQ using the Meteorology-Chemistry Interface Processor (MCIP) version 3.1 to
derive the specific inputs to CMAQ, for example: horizontal wind components (i.e., speed and
direction), temperature, moisture, vertical diffusion rates, and rainfall rates for each grid cell
in each vertical layer. Before initiating the air quality simulations, an evaluation was
conducted to identify the biases and errors associated with the meteorological modeling
inputs. The U.S. EPA 2002 MM5 model performance evaluations used an approach which
included a combination of qualitative and quantitative analyses to assess the adequacy of the
MM5 simulated fields. More detail on the meteorological modeling evaluations can be found
1 q
-------
EUS and WUS grids and five large subregions.w The Atmospheric Model Evaluation Tool
(AMET) was used to conduct the evaluation described in this document.137
The ozone evaluation primarily focused on observed hourly ozone concentrations and
eight-hour daily maximum ozone concentrations above a threshold of 40 ppb. The ozone
model performance evaluation was limited to the ozone season modeled for the EGA: May,
June, July, August, and September. Ozone ambient measurements for 2002 were obtained
from the Air Quality System (AQS) Aerometric Information Retrieval System (AIRS). A
total of 1178 ozone measurement sites were included for evaluation. The ozone data were
measured and reported on an hourly basis.
The PM2.s evaluation focuses on PM2.s total mass and its components including sulfate
(S04), nitrate (N03), total nitrate (TN03=N03+HN03), ammonium (NH4), elemental carbon
(EC), and organic carbon (OC). The PM2.5 performance statistics were calculated for each
month and season individually and for the entire year, as a whole. Seasons were defined as:
winter (December-January-February), spring (March-April-May), summer (June-July-
August), and fall (September-October-November). PM2.5 ambient measurements for 2002
were obtained from the following networks for model evaluation: Speciation Trends Network
(STN, total of 199 sites), Interagency Monitoring of Protected Visual Environments
(IMPROVE, total of 150), and Clean Air Status and Trends Network (CASTNet, total of 83).
The pollutant species included in the evaluation for each network are listed in Table 3.2-4.
For PM2.5 species that are measured by more than one network, we calculated separate sets of
statistics for each network.
Table 3.2-4. PM2.5 Monitoring Networks and Pollutants Species Included in the CMAQ Performance
Evaluation.
AMBIENT
MONITORING
NETWORKS
IMPROVE
CASTNet
STN
PARTICULATE SPECIES
PM2.5
Mass
X
X
S04
X
X
X
N03
X
X
TN03
X
NH4
X
X
X
EC
X
X
OC
X
X
Note that TNOs = (NOs + HNOs)
There are various statistical metrics available and used by the science community for
model performance evaluation. The four evaluation statistics used to evaluate CMAQ
performance were two bias metrics, normalized mean bias and fractional bias; and two error
metrics, normalized mean error and fractional error.
w The subregions are defined by States where: Midwest is IL, IN, MI, OH, and WI; Northeast is CT, DE,
MA, MD, ME, NH, NJ, NY, PA, RI, and VT; Southeast is AL, FL, GA, KY, MS, NC, SC, TN, VA, and WV;
Central is AR, IA, KS, LA, MN, MO, NE, OK, and TX; West is AK, CA, OR, WA, AZ, NM, CO, UT, WY, SD,
ND, MT, ID, and NV.
3-37
-------
The "acceptability" of model performance was judged by comparing our CMAQ 2002
performance results to the range of performance found in recent regional ozone and PM2.s
model applications. These other modeling studies represent a wide range of modeling
analyses which cover various models, model configurations, domains, years and/or episodes,
chemical mechanisms, and aerosol modules. Overall, the statistical calculations of model bias
and error indicate that the CMAQ predicted ozone and PM2.s concentrations for 2002 are
within the range or close to that found in recent U.S. EPA applications.138 Figures 3.2-12 to
3.2-15 show the seasonal aggregate normalized mean bias for 8-hourly ozone and PM2.s over
the two 12-km grids. The CMAQ model performance results give us confidence that our
applications of CMAQ using this 2002 modeling platform provide a scientifically credible
approach for the impacts of EGA controls on ozone and PM2.5 concentrations, visibility levels,
and acid deposition amounts.
O3 HME (%Hot run KBiac_1gKiit»_EUB KM BOOJQ501 la 20080931
CIRCLE- AQS_Bhrmax;
Figure 3.2-12. Normalized Mean Bias (%) of hourly ozone (40 ppb threshold) by monitor for 12-km
Eastern U.S. domain, seasonal aggregate
3-38
-------
03 NMB |>| lor run IQUlx 12km WUS 'or JIXI2U1IM
CIRCLE=AQS 8hrmax;
Figure 3.2-13. Normalized Mean Bias (%) of hourly ozone (40 ppb threshold) by monitor for 12-km
Western U.S. domain, seasonal aggregate.
Figure 3.2-14. Normalized Mean Bias (%) of annual PM2.s by monitor for 12-km Eastern U.S. domain,
2002
3-39
-------
ClRCLE-tMPflOVE. TflEANGLE-STN
Figure 3.2-15. Normalized Mean Bias (%) of annual PM2.5 by monitor for 12-km Western U.S. domain,
2002
3.3 Impacts on Ecosystems
3.3.1 Sulfur and Nitrogen Deposition (overview)
Large ships release emissions over a wide area, and depending on prevailing winds
and other meteorological conditions, these emissions may be transported hundreds and even
thousands of kilometers across North America. Sections 3.1 and 3.2 discuss the results of U.S.
air quality modeling which documents this phenomenon. Overall these engines emit a large
amount of NOX, SOX and direct PM which impact not only ambient air concentrations but
also contribute to deposition of nitrogen and sulfur in many sensitive ecological areas
throughout the U.S.
Sulfur in marine fuel is primarily emitted as SOz, with a small fraction (about two
percent) being converted to S03.139'140'141 S03 almost immediately forms sulfate and is
emitted as primary PM by the engine and consists of carbonaceous material, sulfuric acid, and
ash (trace metals). Ships operating on high sulfur fuel therefore, emit large amounts of both
SOz and sulfate PM. The vast majority of the primary PM is less than or equal to 2.5 ^im in
diameter, and accounts for the majority of the number of particles in exhaust, but only a small
fraction of the mass of diesel PM. These particles also react in the atmosphere to form
secondary PM, which exist there as a carbon core with a coating of organic carbon
compounds, nitrate particles, or as sulfuric acid and ash, sulfuric acid aerosols, or sulfate
particles associated with organic carbon.
At the same time, ships emit large amounts of NO and NOz (NOX) emissions which
are carried into the atmosphere where they may be chemically altered and transformed into
new compounds. For example, N02 can also be further oxidized to nitric acid (HN03) and
can contribute in that form to the acidity of clouds, fog, and rain water and can also form
ambient particulate nitrate (pN03) which may be deposited either directly onto terrestrial and
aquatic ecosystems ("direct deposition") or deposited onto land surfaces where it
subsequently runs off and is transferred into downstream waters ("indirect deposition").
3-40
-------
Deposition of nitrogen and sulfur resulting from ship operations can occur either in a
wet or dry form. Wet deposition includes rain, snow, sleet, hail, clouds, or fog. Dry
deposition includes gases, dust, and minute particulate matters. Wet and dry atmospheric
deposition of PM2.s delivers a complex mixture of metals (such as mercury, zinc, lead, nickel,
arsenic, aluminum, and cadmium), organic compounds (such as polycyclic organic matter,
dioxins, and furans) and inorganic compounds (such as nitrate, sulfate). Together these
emissions from ships are deposited onto terrestrial and aquatic ecosystems across the U.S.
contributing to the problems of ecosystem acidification, ecosystem nutrient enrichment, and
ecosystem eutrophication.
Deposition of nitrogen and sulfur causes acidification, which alters biogeochemistry
and affects animal and plant life in terrestrial and aquatic ecosystems across the U.S. Major
effects include a decline in some forest tree species, such as red spruce and sugar maple; and a
loss of biodiversity of fishes, zooplankton, and macro invertebrates. The sensitivity of
terrestrial and aquatic ecosystems to acidification from nitrogen and sulfur deposition is
predominantly governed by the earth's geology.
Biological effects of acidification in terrestrial ecosystems are generally linked to
aluminum toxicity and decreased ability of plant roots to take up base cations. Decreases in
the acid neutralizing capacity and increases in inorganic aluminum concentration contribute to
declines in zooplankton, macro invertebrates, and fish species richness in aquatic ecosystems.
Across the U.S., ecosystems will continue to be acidified by current NOX and SOX emissions
from stationary sources, area sources, and mobile sources. For example, in the Adirondacks
Mountains of New York State, the current rates of nitrogen and sulfur deposition exceed the
amount that would allow recovery of the most acid sensitive lakes to a sustainable acid
neutralizing capacity (ANC) level.
In addition to the role nitrogen deposition plays in acidification, it also causes
ecosystem nutrient enrichment and eutrophication that alters biogeochemical cycles and
harms animal and plant life such as native lichens and alters biodiversity of terrestrial
ecosystems, such as forests and grasslands. Nitrogen deposition contributes to eutrophication
of estuaries and coastal waters which result in toxic algal blooms and fish kills. For example,
the Chesapeake Bay Estuary is highly eutrophic and 21 -30% of total nitrogen load comes
from deposition. Freshwater ecosystems may also be impacted by nitrogen deposition, for
example, high elevation freshwater lakes in the western U.S. experience adverse ecosystem
changes at nitrogen deposition rates as low as 2 kg N/ha/yr.142
There are a number of important quantified relationships between nitrogen deposition
levels and ecological effects. Certain lichen species are the most sensitive terrestrial taxa to
nitrogen with species losses occurring at just 3 kg N/ha/yr in the Pacific Northwest of the U.S.
and the southern portion of the State of California (See Figure 3-5 for the geographic
distribution of these lichens in the continental U.S.). The onset of declining biodiversity was
found to occur at levels of 5 kg N/ha/yr and above within grasslands in Minnesota and in
Europe. Altered species composition of Alpine ecosystems and forest encroachment into
temperate grasslands was found at 10 kg N/ha/yr and above in the U.S.
3-41
-------
The biogeochemical cycle of mercury, a well-known neurotoxin, is closely tied to the
sulfur cycle. Mercury is taken up by living organisms in the methylated form, which is easily
bioaccumulated in the food web. Sulfate-reducing bacteria in wetland and lake sediments
play a key role in mercury methylation. Changes in sulfate deposition have resulted in
changes in both the rate of mercury methylation and the corresponding mercury
concentrations in fish. In 2006, 3,080 fish advisories were issued in the U.S. due to the
presence of methyl mercury in fish. Although sulfur deposition is important to mercury
methylation, several other interrelated factors seem to also be related to mercury uptake,
including low lake water pH, dissolved organic carbon, suspended particulate matter
concentrations in the water column, temperature, and dissolved oxygen. In addition, the
proportion of upland to wetland land area within a watershed, as well as wetland type and
annual water yield, appear to be important.
3.3.1.1 Recent U.S. Deposition Data
Over the past two decades the U.S. has undertaken numerous efforts to reduce
nitrogen and sulfur deposition across the U.S. Analyses of long-term monitoring data for the
U.S. show that deposition of both nitrogen and sulfur compounds has decreased over the last
17 years although many areas continue to be negatively impacted by deposition. Deposition
of inorganic nitrogen and sulfur species routinely measured in the U.S. between 2004 and
2006 were as high as 9.6 kg N/ha/yr and 21.3 kg S/ha/yr. Figures 3.3-1 and 3.3-2 show that
annual total deposition (the sum of wet and dry deposition) decreased between 1989-1999 and
2004-2006 due to sulfur and NOx controls on power plants, motor vehicles and fuels in the
U.S. The data shows that reductions were more substantial for sulfur compounds than for
nitrogen compounds. These numbers are generated by the U.S. national monitoring network
and they likely underestimate nitrogen deposition because NH3 is not measured. In the
eastern U.S., where data are most abundant, total sulfur deposition decreased by about 36
percent between 1990 and 2005 while total nitrogen deposition decreased by 19 percent over
the same time frame.143
The U.S. is concerned that both current ship emissions and projected future ship
emissions will seriously erode environmental improvements that have been achieved in these
ecologically sensitive areas. As the air quality modeling results in section 3.3.1.7 show, both
nitrogen and sulfur deposition resulting from ship emissions impact a significant portion of
ecologically sensitive areas in the U.S.
3-42
-------
1lM9-1's|91
•13
•
*.F
bt 1fl»-» »t
0Jf* SMR* IUOP. KXtf. US. ffil MP7
^ QWT ir^WT pBtlKL
—30 .-..j,!.
cr/sulfur flipouflw! *wet9tfflvttpi
Figure 3.3-1 Total Sulfur Deposition in the Contiguous U.S., 1989-1991 and 2004 -2006
3-43
-------
il
t
i iBMi mm unrpn rp iiihn inuy •• |m in I im| JMI w»i
Sbn of drci« Imflcabi On matin rjjnltjtSc cf lobl ntrcger dipMtlnn
CHOP; tn cirv?s raicait thi tnttaynii C*HJH 1
I>ry nlrogen d^ostllco • Wat rJt'c«n
y
»
14
1.3
Sfl
•6 '
*' '~?ff^*£~^tSu
a
*•
Huirjwrs f**t »U utrc-jtr
DWT a 1-yHi
t iwcbn),
: 1 oxen In ercrs inacxi mt iwaiscMn 3l ioia mvogn
Dry N o ,'
• WB NCV «F«in«n
Ir
*fl
'CwtP*)*:
iJaniagM
UM M 19M-1991 md 73 noirttisrinj em maiM-«ia
U.S. fm
Figure 3.3-2 Total Nitrogen Deposition in the Contiguous U.S., 1989-1991 and 2004-2006
3-44
-------
3.3.1.2 Areas Potentially Sensitive to Nitrogen and Sulfur Deposition in the U.S.
Currently the secondary NAAQS for NOx and SOx are being reviewed, specifically
addressing the welfare effects of acidification and nitrogen nutrient enrichment. As part of
this review, ecosystem maps (Figures 3.3-3 through 3.3-6) 144 for the continental U.S. have
been created that depict areas that are potentially sensitive to aquatic and terrestrial
acidification, and aquatic and terrestrial nutrient enrichment. Taken together, these sensitive
ecological areas are of greatest concern with regard to the deposition of nitrogen and sulfur
compounds resulting from ship emissions. NOx and SOx emissions from ships today and in
2020 will significantly contribute to higher annual total nitrogen and sulfur deposition in all of
these potentially sensitive ecosystems. See Section 3.3.1.7 for a discussion and accompanying
maps which documents both the level and geographic impact of ship emissions in 2020 on
nitrogen and sulfur deposition in the U.S.
Terrestrial Acidification- U. S. Geography
Deposition of total nitrogen (including both oxidized and reduced forms) and sulfur
species contributing to acidification were routinely measured in the U.S. between 2004 and
2006 and those results are shown in Figures 3.3-1 and 3.3-2. Figure 3.3-3 depicts areas across
the U.S. which are potentially sensitive to terrestrial acidification including forest ecosystems in
the Adirondack Mountains located in the State of New York, the Green Mountains in the State
of Vermont, the White Mountains in the State of New Hampshire, the Allegheny Plateau in the
State of Pennsylvania, in the southeastern part of the U.S., and high-elevation ecosystems in the
southern Appalachians. In addition, areas of the Upper Midwest and parts of the State of
Florida are also at significant risk with regard to terrestrial acidification.
x The first draft risk and exposure assessment and other documents associated with this review are available at:
http://www.epa.gov/ttn/naaqs/standards/no2so2sec/cr rea.html
3-45
-------
| Area o1 Higest Potential Sersilivrty
| Top Ouartile N
I Top Ouartile S
t.OOD
1 km
Figure 3.3-3 Areas Potentially Sensitive to Terrestrial Acidification
Aquatic Acidification-U.S. Geography
A number of national and regional assessments have been conducted to estimate the
distribution and extent of surface water acidity in the U.S.145'146'147'148'149'150'151'152'153 As a result,
several regions of the U.S. have been identified as containing a large number of lakes and
streams which are seriously impacted by acidification.
Figure 3.3-4 illustrates those areas of the U.S. where aquatic ecosystems are at risk
from acidification. These sensitive ecological regions include portions of the Northeast U.S -
especially all the New England States, the Adirondacks, and the Catskill Mountains in the
State of New York; the Southeast U.S.-including the Appalachian Mountains and the northern
section of the State of Florida; all upper Midwest States and parts of the western U.S.154 -
especially the Los Angeles Basin and surrounding area and the Sierra Nevada Mountains in
the State of California. Two western mountain ranges with the greatest number of acid
sensitive lakes155 are the Cascade Mountains, stretching from northern California, through the
entire States of Oregon and Washington, and the Sierra Nevada's, found within the State of
California. The hydrologic cycles in these two mountain ranges are dominated by the annual
accumulation and melting of a dilute, mildly acidic snow pack. Finally, also in the western
U.S., many Rocky Mountain lakes in the State of Colorado are also sensitive to acidifying
deposition effects.156 However, it does not appear that chronic acidification has occurred to
any significant degree in these lakes, although episodic acidification has been reported for
some.
157
3-46
-------
High Pweruial Sensitivity
Acid Ssnsitivs Wiaere (USGS)
Starts
Figure 3.3-4 Areas Potentially Sensitive to Aquatic Acidification
Terrestrial Nutrient Enrichment- U. S. Geography
Nitrogen deposition affects terrestrial ecosystems throughout large areas of the U.S.158
Atmospheric nitrogen deposition is the main source of new nitrogen in many terrestrial
ecosystems throughout the U.S and impacts large numbers of forests, wetlands, freshwater
bogs and salt marshes.159 Figure 3.3-5 depicts those ecosystems potentially sensitive to
terrestrial nutrient enrichment resulting from nitrogen deposition - including nitrogen
deposition from ships.
Severe symptoms of nutrient enrichment or nitrogen saturation, have been observed in
forest ecosystems of the State of West Virginia's northern hardwood watersheds; 16° in high-
elevation spruce-fir ecosystems in the Appalachian Mountains;161 in spruce-fir ecosystems
throughout the northeastern U.S.;162'163 and in lower-elevation eastern U.S.
forests.164'165'166'167 In addition, mixed conifer forests in the Los Angeles Air Basin within
the State of California are also heavily impacted and exhibit the highest stream water nitrate
1RSTiRQ
concentrations documented within wild lands in North America. ' In general, it is
believed that deciduous forest stands in the eastern U.S. have not progressed toward nitrogen
saturation as rapidly or as far as coniferous stands in the eastern U.S.
In addition to these forest ecosystems, nitrogen deposition adversely impacts U.S.
grasslands or prairies which are located throughout the U.S.171 The vast majority of these
grasslands are found in the Central Plains regions of the U.S. between the Mississippi River
and the foothills of the Rocky Mountains. However, some native grasslands are scattered
throughout the Midwestern and Southeastern U.S.172 Also considered sensitive to nitrogen
3-47
-------
nutrient enrichment effects, and receiving high levels of atmospheric deposition, are some
173
arid and semi-arid ecosystems and desert ecosystems in the southwestern U.S. However,
water is generally more limiting than nitrogen in these areas. The alpine ecosystems in the
State of Colorado, chaparral watersheds of the Sierra Nevada Mountains in the State of
California, lichen and vascular plant communities in the San Bernardino Mountains in
California and the entire U.S. Pacific Northwest, and the Southern California coastal sage
scrub community are among the most sensitive terrestrial ecosystems to nitrogen deposition in
the U.S. 174'175 *
I High Potent
250 500 750 1.000km
Figure 3.3-5 Areas Potentially Sensitive to Terrestrial Nutrient Enrichment
Aquatic Nutrient Enrichment -U.S. Geography
Aquatic nutrient enrichment impacts a wide range of waters within the U.S. from
wetlands, to streams, rivers, lakes, estuaries and coastal waters. All are vital ecosystems to
the U.S. and all are impacted by ship emissions that contribute to the annual total nitrogen
deposition in the U.S.
Wetlands are found throughout the U.S. and support over 4200 native plant species, of
which 121 have been designated by the U.S. government as threatened or endangered.176
Freshwater wetlands are particularly sensitive to nutrient enrichment resulting from nitrogen
deposition since they contain a disproportionately high number of rare plant species that have
evolved under nitrogen-limited conditions.177 Freshwater wetlands receive nitrogen mainly
from precipitation, land runoff or ground water. Intertidal wetlands develop on sheltered
coasts or in estuaries where they are periodically inundated by marine water that often carries
high nitrogen loads, in addition to receiving water and nutrient inputs from precipitation and
ground/surface water. Wetlands can be divided into three general categories based on
3-48
-------
hydrology: (1) Peatlands and bogs, (2) fens, freshwater marshes, freshwater swamps and (3)
intertidal wetlands.
Fens and bogs are the most vulnerable type of wetland ecosystems with regard to
nutrient enrichment effects of nitrogen deposition.178 In the U.S. they are mostly found in the
glaciated northeast and Great Lakes regions and in the State of Alaska, but also in the
southeast U.S. along the Atlantic Coastal Plain stretching from the States of Virginia through
North Carolina to northern Florida.179 Like bogs, fens are mostly a northern hemisphere
phenomenon -- occurring in the northeastern United States, the Great Lakes region, western
Rocky Mountains, and much of Canada -- and are generally associated with low temperatures
and short growing seasons, where ample precipitation and high humidity cause excessive
moisture to accumulate.180
The third type of wetlands sensitive to nitrogen deposition are marshes, characterized
by emergent soft-stemmed vegetation adapted to saturated soil conditions. There are many
different kinds of marshes in the U.S., ranging from the prairie potholes in the interior of the
U.S. to the Everglades found in the extreme southern portion of the State of Florida. U.S.
fresh water marshes are important for recharging groundwater supplies, and moderating
stream flow by providing water to streams and as habitats for many wildlife species.181
Nitrogen deposition is the main source of nitrogen for many surface waters in the U.S.
including headwater streams, lower order streams, and high elevation lakes.182'183 Elevated
surface water nitrate concentrations due to nitrogen deposition occur in both the eastern and
western U.S. although high concentrations of nitrate in surface waters in the western U.S. are
not as widespread as in the eastern U.S.
High concentrations of lake or stream water nitrate, indicative of ecosystem nitrogen-
saturation, have been found at a variety of locations throughout the U.S. including the San
Bernardino and San Gabriel Mountains within the Los Angeles Air Basin in the State of
California 184, the Front Range Mountains in the State of Colorado,185'186 the Allegheny
Mountains in the State of West Virginia,187 the Catskill and Adirondack Mountains in the
State of New York'188'189'190 and the Great Smoky Mountains in the State of Tennessee.
Nitrogen nutrient enrichment is a major environmental problem facing all U.S. coastal
regions, but especially the Eastern, mid-Atlantic, and Gulf Coast regions, as excess nitrogen
leads to eutrophication. There is broad scientific consensus that nitrogen-driven
eutrophication of shallow estuaries in the U.S. has increased over the past several decades and
that environmental degradation of coastal ecosystems is now a widespread occurrence.191 A
recent national assessment of eutrophic conditions in U.S. estuaries found that 65% of the
assessed systems had moderate to high overall eutrophic conditions.192 Estuaries and coastal
waters tend to be nitrogen-limited and are therefore inherently sensitive to increased
atmospheric nitrogen deposition.193 Of 138 estuaries examined in the National Assessment,
44 were identified as showing symptoms of nutrient enrichment. Of the 23 estuaries
examined in the Northeast U.S. 61% were classified as moderately to severely degraded.
Other regions of the U.S. had mixtures of low, moderate, and high degree of
eutrophication.194 The contribution from atmospheric nitrogen deposition can be greater than
3-49
-------
30% of total nitrogen loads in some of the most highly eutrophic estuaries in the US,
including the Chesapeake Bay.
EPA's draft risk and exposure assessment (REA) for the NOxSOx secondary NAAQS
developed an overview map of the U.S. that identifies areas of national aquatic nutrient
enrichment sensitivity. They utilized the eutrophic estuaries from NOAA's Coastal
Assessment Framework and areas that exceed the nutrient criteria for lakes/reservoirs (U.S.
EPA, 2002). Both these were combined and compared to total nitrogen deposition. The
resulting map revealed areas of highest potential sensitivity to nitrogen deposition as shown in
Figure 3.3-6. These areas are identified in blue as nutrient sensitive estuaries contained in
NOAA's Coastal Assessment Framework (CAP), and red in areas where deposition exceeds
the nutrient criteria. Yellow areas indicate those areas that are below the nutrient criteria but
are within 5 kg N/ha/yr of exceeding it.
Total Nitrogen Nutrient Criteria for Lake/Reservoirs by Zone
Zone
TN (mg/L)
N dep {kg N ha-1yr-1)
II III i IV V VI VII VIII IX X
0.1 0.4 0.44 0.56 0,78 0.66 0.24 0.36
4.46 12.6 13.5 16.1 20.7 18.2 8.57 11.6
XI XII XIII XIV
0.46 0.52 1.27 0.32
13.9 15.3 29.7 10.6
750 1,000
ukrn
NOAA CAP Total Wet N Dep Exceedance BE -9 99 - -5
I I Nutrient Criteria Bdy H -4.99 - 0
I 1 states ^H -26.41 - -15 ^B 0.01 -1.92
Lakes I I-14.99--10
Exceedance levels determined by first converting N concentration
nutrient criteria amounts{mg/litre) to wet N deposition amounts (kg/ha/yr)
using a formula published by Bergstrom and Jansson (2006). These N
deposition amounts were then compared to wet NADP N deposition
(2002) amounts to determine areas of the US that are either above
or below the nutrient criteria levels for lakes/reservoirs.
* Blank areas do not have an EPA nutrient criteria for lake/reservoirs.
Figure 3.3-6 Areas Potentially Sensitive to Aquatic Nutrient Enrichment
The most extreme effects of nitrogen deposition on U.S. aquatic ecosystems result in
severe nitrogen-loading to these ecosystems that contribute to hypoxic zones devoid of life.
Three hypoxia zones of special concern in the U.S. are (1) the zone located in the Gulf of
Mexico straddling the States of Louisiana and Texas, (2) The Chesapeake Bay located
between the States of Maryland and Virginia, and (3) Long Island Sound located between the
States of New York and Connecticut. The largest hypoxia zone in the U.S. is in the northern
Gulf of Mexico along the continental shelf. During midsummer, this zone has regularly been
larger than 16,000km2.195 Figure 3.3-7 depicts the location of these three hypoxic zones.
3-50
-------
i_ a LJ i n i A w A
BiraolreiS oxygen:
T" <-2 3 nji'L
0 (5 30 M 60
fljfj MOJW: UiMCOH 2007b
Figure 3.3-7 a. Hypoxia Zone in 2007 for the Gulf of Mexico
CONNECTICUT
Dittohrid
H- 1 Dm^/L
•i 1 0 to <2.0 m^L
2 0 » <3
^ 3.0 to <3 5
•I 3 5 to
-------
V aiymg dissolved oxygen levels and oveiall fish catch in the Chesapeake
Bay throughMy, 2003.Source Virginia InsWuU cifManm Science
Figure 3.3-7 c Hypoxia Zone for Chesapeake Bay in 2003
3.3.1.3 Science of Nitrogen and Sulfur Deposition
Nitrogen and sulfur interactions in the environment are highly complex. Both are
essential, and sometimes limiting, nutrients needed for growth and productivity. Excess of
nitrogen or sulfur can lead to acidification, nutrient enrichment, and eutrophication.
Ships release emissions over a wide area, and depending on prevailing winds and
other meteorological conditions, these emissions may be transported hundreds and even
thousands of kilometers across North America. Section 3.2 discusses the results of U.S. air
quality modeling which documents this phenomenon. Overall, these engines emit a large
amount of NOX, SOX and direct PM which impact not only ambient air concentrations but
also contribute to deposition of nitrogen and sulfur in many sensitive ecological areas
throughout the U.S.
The sulfur in marine fuel is primarily emitted as sulfur dioxide (SOz), with a small
fraction (about two percent) being converted to sulfur trioxide (S03). 196 S03 almost
immediately forms sulfate and is also emitted as primary PM by the engine and consists of
carbonaceous material, sulfuric acid, and ash (trace metals). The vast majority of the primary
PM is less than or equal to 2.5 ^im in diameter, and accounts for the majority of the number of
3-52
-------
particles in exhaust, but only a small fraction of the mass of diesel PM. These particles also
react in the atmosphere to form secondary PM, which exist there as a carbon core with a
coating of organic carbon compounds, nitrate particles, or as sulfuric acid and ash, sulfuric
acid aerosols, or sulfate particles associated with organic carbon.
At the same time, ships emit large amounts of nitric oxide (NO) and nitrogen dioxide
(NO2) emissions which are carried into the atmosphere where they may be chemically altered
and transformed into new compounds. For example, N02 can also be further oxidized to
nitric acid (HN03) and can contribute in that form to the acidity of clouds, fog, and rain water
and can also form ambient particulate nitrate (pNOs) which may be deposited either directly
onto terrestrial and aquatic ecosystems ("direct deposition") or deposited onto land surfaces
where it subsequently runs off and is transferred into downstream waters ("indirect
deposition").
Deposition of nitrogen and Sulfur resulting from ship operations can occur either in a
wet or dry form. Wet deposition includes rain, snow, sleet, hail, clouds, or fog. Dry
deposition includes gases, dust, and minute particulate matters. Wet and dry atmospheric
deposition of PM2.5 delivers a complex mixture of metals (such as mercury, zinc, lead, nickel,
arsenic, aluminum, and cadmium), organic compounds (such as polycyclic organic matter,
dioxins, and furans) and inorganic compounds (such as nitrate, sulfate) to terrestrial and
aquatic ecosystems.
The chemical form of deposition is determined by ambient conditions (e.g.,
temperature, humidity, oxidant levels) and the pollutant source. Chemical and physical
transformations of ambient particles occur in the atmosphere and in the media (terrestrial or
aquatic) on which they deposit. These transformations influence the fate, bioavailability and
potential toxicity of these compounds. The atmospheric deposition of metals and toxic
compounds is implicated in severe ecosystem effects.197
Ships also emit primary PM. In addition, secondary PM is formed from NOx and SOx
gaseous emissions and associated chemical reactions in the atmosphere. The major
constituents of secondary PM are sulfate, nitrate, ammonium, and hydrogen ions. Secondary
aerosol formation depends on numerous factors including the concentrations of precursors;
the concentrations of other gaseous reactive species such as ozone, hydroxyl radical, peroxy
radicals, and hydrogen peroxide; atmospheric conditions, including solar radiation and
relative humidity; and the interactions of precursors and preexisting particles within cloud or
fog droplets or on or in the liquid film on solid particles. 98
The lifetimes of particles vary with particle size. Accumulation-mode particles such
as the sulfates and nitrates are kept in suspension by normal air motions and have a lower
deposition velocity than coarse-mode particles; they can be transported thousands of
kilometers and remain in the atmosphere for a number of days. They are removed from the
atmosphere primarily by cloud processes. Dry deposition rates are expressed in terms of
deposition velocity that varies with the particle size, reaching a minimum between 0.1 and 1.0
micrometer (|j,m) aerodynamic diameter.199
3-53
-------
Particulate matter is a factor in acid deposition. Particles serve as cloud condensation
nuclei and contribute directly to the acidification of rain. In addition, the gas-phase species
that lead to the dry deposition of acidity are also precursors of particles. Therefore, reductions
in NOX and S02 emissions will decrease both acid deposition and PM concentrations, but not
necessarily in a linear fashion. Sulfuric acid, ammonium nitrate, and organic particles also are
deposited on surfaces by dry deposition and can contribute to environmental effects.200
3.3.1.4 Computing Atmospheric Nitrogen and Sulfur Deposition to Specific Locations
Inputs of new nitrogen, i.e., non-recycled mostly anthropogenic in origin, are often
key factors controlling primary productivity in nitrogen-sensitive estuarine and coastal
waters.201 Increasing trends in urbanization, agricultural intensity, and industrial expansion
have led to increases in nitrogen deposited from the atmosphere on the order of a factor of
10 in the previous 100 years Direct fluxes of atmospheric nitrogen to ocean and gulf
waters along the northeast and southeast U.S. are now roughly equal to or exceed the load of
new nitrogen from riverine inputs at 11, 5.6, and 5.6 kg N/ha for the northeast Atlantic coast
of the U.S., the southeast Atlantic coast of the U.S., and the U.S. eastern Gulf of Mexico,
respectively.203 Atmospheric nitrogen is dominated by a number of sources, most importantly
transportation sources, including ships.
Nitrogen deposition takes different forms physically. Physically, deposition can be
direct, with the loads resulting from air pollutants depositing directly to the surface of a body
of water, usually a large body of water like an estuary or lake. In addition, there is an indirect
deposition component derived from deposition of nitrogen or sulfur to the rest of the
watershed, both land and water, of which some fraction is transported through runoff, rivers,
streams, and groundwater to the water body of concern.
Direct and indirect deposition of nitrogen and sulfur to watersheds depend on air
pollutant concentrations in the airshed above the watershed. The shape and extent of the
airshed is quite different from that of the watershed. In a watershed, everything that falls in
its area, by definition, flows into a single body of water. An airshed, by contrast, is a
theoretical concept that defines the source area containing the emissions contributing a given
level, often 75%, to the deposition in a particular watershed or to a given water body. Hence,
airsheds are modeled domains containing the sources estimated to contribute a given level of
deposition from each pollutant of concern. The principal NOx airsheds and corresponding
watersheds for several regions in the eastern U.S. are shown in Figure 3.3-8.204 These
airsheds extend well into U.S. coastal waters where ships operate.
3-54
-------
Figure 3.3-8 Principal Airsheds and Watersheds for Oxides of Nitrogen for Estuaries.Hudson/Raritan
Bay; Chesapeake Bay; Pamlico Sound; and Altamaha Sound (listed from north to south).
Nitrogen inputs have been studied in several U.S. Gulf Coast estuaries, as well, owing
to concerns about eutrophication there. Nitrogen from atmospheric deposition in these
locations is estimated to be 10 to 40% of the total input of nitrogen to many of these estuaries,
and could be higher for some. Estimates of total nitrogen loadings to estuaries or to other
large-scale elements in the landscape are then computed using measurements of wet and dry
deposition, where these are available, and interpolated with or without a set of air quality
model predictions such as the Extended Regional Acid Deposition Model
(Ext-RADM).205'206'207'208'209
Table 3.3-2 lists several water bodies for which atmospheric nitrogen inputs have been
computed and the ratio to total nitrogen loads is given. The contribution from the atmosphere
ranges from a low of 2-8% for the Guadalupe Estuary in the southern part of the State of
Texas to highs of -38% in the New York State Bight and the Albemarle-Pamlico Sound in the
State of North Carolina.
Table 3.3-2 Atmospheric Nitrogen Loads Relative to Total Nitrogen Loads in Selected U.S.
Great
3-55
-------
Waters*
Wat wholly
Albemarle-Pamlico fjtmmis
Chesapeake Bay
Delaware Bay
Lous bland Sound
NarrasjausettBay
New York Bight
lot alN Load
(million kjtyr)
23
1/0
M
60
^
1*4
AtiiiMnheilr N Load _ T , , , ,
.... , . Permit Load from the Atmosphere
( million kg YT)
9
36
8
12
(),6
61
38
21
1*.
20
12
38
Based on ADN N loads from the watershed onty (excluding dire ctN deposition to the bay surface):
Whgucur Bay, MA 0.021 0.0065 29
Based on ADN directly to the wfterbodj
Delaware Iiilaiid Bays
FlamlereBuy. NY
Guadnliipe Estuary, TX
Massachusetts Bays
NaiYasaiisei Bay
Newport River Coastal Wafer-;. NO
Potomac River, MD
Sai'asota Bay. FL
Taii^a Bay, FL
(excluding ADN loads from the
Li
O.J6
4.2 15.9
22- JO
9
0.2" .0.85 (
35.5
0.6
3.8
watershed):
0.2S
002"
031
16 6
0.4
I.094 O.(i8
1.9
0 16
1,1
21
-
2 8
s i~:
4
'-.55
5
2lefi'om Deposition of Air Polliitaute to the t)|-eatWaters-3j'd Report to Confess (EPA, 2000)
3.3.1.5 Summary of Ecological Effects Associated with Nitrogen and Sulfur and PM
Deposition
Deposition of reduced and oxidized nitrogen and sulfur species cause acidification,
altering biogeochemistry and affecting animal and plant life in terrestrial and aquatic
ecosystems across the U.S. Major effects include a decline in sensitive tree species, such as
red spruce and sugar maple; and a loss of biodiversity of fishes, zooplankton, and macro
invertebrates. The sensitivity of terrestrial and aquatic ecosystems to acidification from
nitrogen and sulfur deposition is predominantly governed by the earth's geology.
Biological effects of acidification in terrestrial ecosystems are generally linked to
aluminum toxicity and decreased ability of plant roots to take up base cations. Decreases in
acid neutralizing capacity and increases in inorganic aluminum concentration contribute to
declines in zooplankton, macro invertebrates, and fish species richness in aquatic ecosystems.
Across the U.S., ecosystems continue to be acidified by current emissions from both
stationary sources, area sources, and mobile sources. For example, in the Adirondack
Mountains of New York State, the current rates of nitrogen and sulfur deposition exceed the
amount that would allow recovery of the most acid sensitive lakes to a sustainable acid
neutralizing capacity (ANC) level.210
3-56
-------
In addition to the role nitrogen deposition plays in acidification, it also causes
ecosystem nutrient enrichment and eutrophication that alters biogeochemical cycles and
harms animal and plant life such as native lichens and alters biodiversity of terrestrial
ecosystems, such as forests and grasslands. Nitrogen deposition contributes to eutrophication
of estuaries and coastal waters which result in toxic algal blooms and fish kills. For example,
the Chesapeake Bay Estuary is highly eutrophic and 21 -30% of total nitrogen load comes
from deposition. Freshwater ecosystems may also be impacted by nitrogen deposition, for
example, high elevation freshwater lakes in the western U.S. experience adverse ecosystem
changes at nitrogen deposition rates as low as 2 kg N/ha/yr.211
The addition of nitrogen to most ecosystems causes changes in primary productivity
and growth of plants and algae, which can alter competitive interactions among species.
Some species grow more than others, leading to shifts in population dynamics, species
composition, and community structure. The most extreme effects of nitrogen deposition
include a shift of ecosystem types in terrestrial ecosystems, and hypoxic zones that are devoid
of life in aquatic ecosystems. 2
There are a number of important quantified relationships between nitrogen deposition
levels and ecological effects. Certain lichen species are the most sensitive terrestrial taxa to
nitrogen with species losses occurring at just 3 kg N/ha/yr in the U.S. Pacific Northwest and
in the southern portion of the State of California. The onset of declining biodiversity was
found to occur at levels of 5 kg N/ha/yr and above within grasslands in both the State of
Minnesota and in Europe. Altered species composition of Alpine ecosystems and forest
encroachment into temperate grasslands was found at 10 kg N/ha/yr and above in both the
U.S.213
A United States Forest Service study conducted in areas within the Tongass Forest in
Southeast Alaska found evidence of sulfur emissions impacting lichen communities. The
authors concluded that the main source of sulfur and nitrogen found in lichens from Mt.
Roberts is likely the burning of fossil fuels by cruise ships and other vehicles and equipment
in downtown Juneau.214
Lichen are an important food source for caribou. This is causing concern about the
potential role damage to lichens may be having on the Southern Alaska Peninsula Caribou
Herd,215 which is an important food source to local subsistence based cultures. This herd has
been decreasing in size, exhibiting both poor calf survival and low pregnancy rates, which are
signs of dietary stress. Currently there is a complete caribou hunting ban, including a ban on
subsistence hunting. If regulation of marine fuels could potentially enhance lichen biomass in
the area, it would contribute in turn to maintenance of an important subsistence resource for
local human populations.
The biogeochemical cycle of mercury, a well-known neurotoxin, is closely tied to the
sulfur cycle. Mercury is taken up by living organisms in the methylated form, which is easily
bioaccumulated in the food web. Sulfate-reducing bacteria in wetland and lake sediments
play a key role in mercury methylation. Changes in sulfate deposition have resulted in
changes in both the rate of mercury methylation and the corresponding mercury
3-57
-------
concentrations in fish. In 2006, 3,080 fish advisories were issued in the U.S. due to the
presence of methyl mercury in fish.216
Although sulfur deposition is important to mercury methylation, several other
interrelated factors seem to also be related to mercury uptake, including low lake water pH,
dissolved organic carbon, suspended particulate matter concentrations in the water column,
temperature, and dissolved oxygen. In addition, the proportion of upland to wetland land area
within a watershed, as well as wetland type and annual water yield, appear to be important.
Current international shipping emissions of PM2.s contain small amounts of metals—
nickel, vanadium, cadmium, iron, lead, copper, zinc, aluminum.217'218'219 Investigations of
trace metals near roadways and industrial facilities indicate that a substantial burden of heavy
metals can accumulate on vegetative surfaces. Copper, zinc, and nickel are shown to be
directly toxic to vegetation under field conditions. While metals typically exhibit low
solubility, limiting their bioavailability and direct toxicity, chemical transformations of metal
compounds occur in the environment, particularly in the presence of acidic or other oxidizing
species. These chemical changes influence the mobility and toxicity of metals in the
environment. Once taken up into plant tissue, a metal compound can undergo chemical
changes, accumulate and be passed along to herbivores or can re-enter the soil and further
cycle in the environment.
Although there has been no direct evidence of a physiological association between tree
injury and heavy metal exposures, heavy metals have been implicated because of similarities
between metal deposition patterns and forest decline.221 This hypothesized correlation was
further explored in high elevation forests in the northeastern U.S. These studies measured
levels of a group of intracellular compounds found in plants that bind with metals and are
produced by plants as a response to sublethal concentrations of heavy metals. These studies
indicated a systematic and significant increase in concentrations of these compounds
associated with the extent of tree injury. These data strongly imply that metal stress causes
tree injury and contributes to forest decline in Northeast U.S.222 Contamination of plant
leaves by heavy metals can lead to elevated soil levels. Trace metals absorbed into the plant
frequently bind to the leaf tissue, and then are lost when the leaf drops. As the fallen leaves
decompose, the heavy metals are transferred into the soil.223'224
Ships also emit air toxics, including polycyclic aromatic hydrocarbons (PAHs) -- a
class of polycyclic organic matter (POM) that contain compounds which are known or
suspected carcinogens. Since the majority of PAHs are adsorbed onto particles less than 1.0
[an in diameter, long range transport is possible. Particles of this size can remain airborne for
days or even months and travel distances up to 10,000km before being deposited on terrestrial
or aquatic surfaces.225 Atmospheric deposition of particles is believed to be the major source
of PAHs to the sediments of Lake Michigan in the Great Lakes, Chesapeake Bay, which is
surrounded by the States of Maryland and Virginia, Tampa Bay in the central part of the State
of Florida and in other coastal areas of the u.S.226'227'228' 29'230 PAHs tend to accumulate in
sediments and reach high enough concentrations in some coastal environments to pose an
environmental health threat that includes cancer in fish populations, toxicity to organisms
living in the sediment and risks to those (e.g., migratory birds) that consume these
3-58
-------
organisms.231'232 PAHs tend to accumulate in sediments and bioaccumulate in freshwater,
flora and fauna.
3.3.1.6 Ecological Effects Nutrient Enrichment
In general, ecosystems that are most responsive to nutrient enrichment from
atmospheric nitrogen deposition are those that receive high levels of nitrogen loading, are
nitrogen-limited, or contain species that have evolved in nutrient-poor environments. Species
that are adapted to low nitrogen supply will often be more readily outcompeted by species that
have higher nitrogen demands when the availability of nitrogen is increased'233'234'235'236 As a
consequence, some native species can be eliminated by nitrogen deposition.237'238'239' 24°
Note the terms "low" and "high" are relative to the amount of bioavailable nitrogen in the
ecosystem and the level of deposition.
Eutrophication effects resulting from excess nitrogen are more widespread than
acidification effects in western North America. Figure 3.3-9 highlights areas in the Western
U.S. where nitrogen effects have been extensively reported. The discussion of ecological
effects of nutrient enrichment which follows is organized around three types of ecosystem
categories which experience impacts from nutrient enrichment: terrestrial, transitional, and
aquatic.
U.S.
<*>-
L JL A JJ
jllf 1-2,3
"fir /(incipie
RMt
Letue
talmnal For a tijN atonal Park
ELO o-u^iil effect*
ient stages).
2,3
2, 3f 4, 5, 6, 7
(arcumslantial evictence, see text), 8
1. N onrichiTrtnl at aUlrapfikadiun u! lake-*
2 t-VL'OMC nitrate fcvolx in rv-inlf
Terrealnal
3
inciniWBCl ftjttrs or nhrogcnous trace
4 AUorod SlflOl «Wmynf1ift8 in firMpOnS* 10 N
5. PhyaktogkaJ pwluibaUoii of pbnti. cixnbn«J rrflectiaf orunea
6 '"Tipscts oni IC^WP cofnmuniscs.
7. EvWtfnue Otat lhi«at«t'«'J und eitddn^ated speci
6 OetTeaa&d dnwrnly rf myaanfurjil c£3rnmuni:ifls
9. f ivesl ftKpansxso irba grasslands (prefitarury crvrtemsfr tar)
A Ptol w«Hi lichen (jwiuwraly alTedEd by air potluinn w^h a ma\a N deposition Etrnponenl
^ htgh-flk?w3liixi lake wlh vleva^ci rtborltt tp(HirtwLhf from N ttopg&flcin
Available datj indi^tlc elevated N dcposiiicn, but ncoiogtca offeKrts have nol been studed.
Sou^e |=eir. elfll
3-59
-------
Figure 3.3-9. Map of the Western U.S. Showing the Primary Geographic Areas where Nitrogen
Deposition Effects have been Reported
Terrestrial
Ecological effects of nitrogen deposition occur in a variety of taxa and ecosystem
types including: forests, grasslands, arid and semi-arid areas, deserts, lichens, alpine, and
mycorrhizae. Atmospheric inputs of nitrogen can alleviate deficiencies and increase growth
of some plants at the expense of others. Nitrogen deposition alters the competitive
relationships among terrestrial plant species and therefore alters species composition and
diversity. '242'243 Wholesale shifts in species composition are easier to detect in short-lived
terrestrial ecosystems such as annual grasslands, in the forest understory, or mycorrhizal
associations, than for long-lived forest trees where changes are evident on a decade or longer
time scale. Note species shifts and ecosystem changes can occur even if the ecosystem does
not exhibit signs of nitrogen saturation.
There are a number of important quantified relationships between nitrogen deposition
levels and ecological effects.244 Certain lichen species are the most sensitive terrestrial taxa
to nitrogen in the U.S. with clear adverse effects occurring at just 3 kg N/ha/yr. Figure 3-5
shows the geographic distribution of lichens in the U.S. Among the most sensitive U.S.
ecosystems are Alpine ecosystems where alteration of plant covers of an individual species
(Carex rupestris) was estimated to occur at deposition levels near 4 kg N/ha/yr and modeling
indicates that deposition levels near 10 kg/N/ha/yr alter plant community assemblages.245
Within grasslands, the onset of declining biodiversity was found to occur at levels of 5 kg
N/ha/yr. Forest encroachment into temperate grasslands was found at 10 kg N/ha/yr and
above in the U.S. Table 3.3-3 provides a brief list of nitrogen deposition levels and associated
ecological effects.
Table 3.3-3 Examples of Quantified Relationship Between Nitrogen Deposition Levels and Ecological
Effects3
Kg
IV/ha/vr
-1.5
3.1
4
5
5.6 - 10
5-10
5-15
Ecological effect
Altered diatom communities in high
elevation freshwater lakes aad elevated
nitrogen in tree leaf tissue high elevation
forests in the U.S.
Decline of some lichen species in the
Western U.S. (critical load)
Altered growth and eo\rerage of alpine
plant species in U.S.
Onset of decline of species richness in
grasslands of the U.S. and U.K.
Onset of nitrate leaching in Eastern
forests of the U.S.
Multiple effects in tundra, bogs and
freshwater lakes in Europe (critical loads)
Multiple effects in arctic, alpine,
subalpine and scrub habitats in Europe
(critical loads)
3-60
-------
Note:a EPA, Integrated Science Assessment for Oxides of Nitrogen
and Sulfur-Ecological criteria
Most terrestrial ecosystems are nitrogen-limited, therefore they are sensitive to
perturbation caused by nitrogen additions.24 The factors that govern the vulnerability of
terrestrial ecosystems to nutrient enrichment from nitrogen deposition include the degree of
nitrogen limitation, rates and form of nitrogen deposition, elevation, species composition,
length of growing season, and soil nitrogen retention capacity.
Regions and ecosystems in the western U.S. where nitrogen nutrient enrichment
effects have been documented in terrestrial ecosystems are shown on Figure S.3-9.247 The
alpine ecosystems of the Colorado Front Range, chaparral watersheds of the Sierra Nevada,
lichen and vascular plant communities in the San Bernardino Mountains and the Pacific
Northwest, and the southern California coastal sage scrub community are among the most
sensitive terrestrial ecosystems in the western U.S.
In the eastern U.S., the degree of nitrogen saturation of the terrestrial ecosystem is
often assessed in terms of the degree of nitrate leaching from watershed soils into ground
water or surface water. Studies have estimated the number of surface waters at different
stages of saturation across several regions in the eastern U.S.248 Of the 85 northeastern
watersheds examined, 40% were in nitrogen-saturation Stage 0Y, 52% in Stage 1, and 8% in
Stage 2. Of the northeastern sites for which adequate data were available for assessment,
those in Stage 1 or 2 were most prevalent in the Adirondack and Catskill Mountains in the
State of New York.
Transitional
About 107.7 million acres of wetlands are widely distributed in the conterminous U.S.,
95 percent of which are freshwater wetlands and 5 percent are estuarine or marine wetlands249
(Figure 3.3-10). At one end of the spectrum, bogs or peatland are very sensitive to nitrogen
deposition because they receive nutrients exclusively from precipitation, and the species in
them are adapted to low levels of nitrogen'250'251'252 Intertidal wetlands are at the other end of
the spectrum; in these ecosystems marine/estuarine water sources generally exceed
atmospheric inputs by one or two orders of magnitude.253 Wetlands are widely distributed,
including some areas that receive moderate to high levels of nitrogen deposition.
Nitrogen deposition alters species richness, species composition and biodiversity in
U.S. wetland ecosystems.254 The effect of nitrogen deposition on these ecosystems depends
on the fraction of rainfall in its total water budget. Excess nitrogen deposition can cause shifts
in wetland community composition by altering competitive relationships among species,
Y In Stage 0, nitrogen inputs are low and there are strong nitrogen limitations on growth. Stage 1 is characterized
by high nitrogen rentention and fertilization effect of added nitrogen on tree growth. Stage 2 includes the
induction of nitrification and some nitrate leaching, though growth may still be high. In Stage 3 tree growth
declines, nitrification and nitrate loss continue to increase, but nitrogen mineralization rates begin to decline.
3-61
-------
which potentially leads to effects such as decreasing biodiversity, increasing non-native
species establishment and increasing the risk of extinction for sensitive and rare species.
U.S. wetlands contain a high number of rare plant species.255'256'257 High levels of
atmospheric nitrogen deposition increase the risk of decline and extinction of these species
that are adapted to low nitrogen conditions. In general these include the genus Isoetessp., of
which three species are federally endangered; insectivorous plants like the endangered green
pitcher Sarracenia oreophila; and the genus Sphagnum, of which there are 15 species listed as
endangered by eastern U.S. States. Roundleaf sundew (Drosera rotundifolia) is also
9^R
susceptible to elevated atmospheric nitrogen deposition. This plant is native to, and
broadly distributed across, the U.S. and is federally listed as endangered in Illinois and Iowa,
threatened in Tennessee, and vulnerable in New York.259 In the U.S., Sarraceniapurpurea
can be used as a biological indicator of local nitrogen deposition in some locations.260
Figure 3.3-10 Location of Wetlands in Continental U.S.
Freshwater Aquatic
Nitrogen deposition alters species richness, species composition and biodiversity in
freshwater aquatic ecosystems across the U.S.261 Evidence from multiple lines of research
and experimental approaches support this observation, including paleolimnological
reconstructions, bioassays, mesocosm and laboratory experiments. Increased nitrogen
deposition can cause a shift in community composition and reduce algal biodiversity.
Elevated nitrogen deposition results in changes in algal species composition, especially in
sensitive oligotrophic lakes. In the West, a hindcasting exercise determined that the change in
Rocky Mountain National Park lake algae that occurred between 1850 and 1964 was
associated with an increase in wet nitrogen deposition that was only about 1.5 kg N/ha.
Similar changes inferred from lake sediment cores of the Beartooth Mountains of Wyoming
also occurred at about 1.5 kg N/ha deposition.262
3-62
-------
Some freshwater algae are particularly sensitive to added nutrient nitrogen and
experience shifts in community composition and biodiversity with increased nitrogen
deposition. For example, two species of diatom (a taxanomic group of algae), Asterionella
formosa and Fragilaria crotonensis, now dominate the flora of at least several alpine and
montane Rocky Mountain lakes. Sharp increases have occurred in Lake Tahoe.2 3'264'
265,266,267,268 The timing Of fl]js shift has varied, with changes beginning in the 1950s in the
southern Rocky Mountains and in the 1970s or later in the central Rocky Mountains. These
species are opportunistic algae that have been observed to respond rapidly to disturbance and
slight nutrient enrichment in many parts of the world.
Estuarine Aquatic
Nitrogen deposition also alters species richness, species composition and biodiversity
in estuarine ecosystems throughout the U.S.269 Nitrogen is an essential nutrient for estuarine
and marine fertility. However, excessive nitrogen contributes to habitat degradation, algal
blooms, toxicity, hypoxia (reduced dissolved oxygen), anoxia (absence of dissolved oxygen),
reduction of sea grass habitats, fish kills, and decrease in biodiversity.270'271'272'273'274'2
Each of these potential impacts carries ecological and economic consequences. Ecosystem
services provided by estuaries include fish and shellfish harvest, waste assimilation, and
recreational activities.276
Increased nitrogen deposition can cause shifts in community composition, reduced
hypolimnetic DO, reduced biodiversity, and mortality of submerged aquatic vegetation. The
form of deposited nitrogen can significantly affect phytoplankton community composition in
estuarine and marine environments. Small diatoms are more efficient in using nitrate than
NH4+. Increasing NH4+ deposition relative to nitrate in the eastern U.S. favors small diatoms
at the expense of large diatoms. This alters the foundation of the food web. Submerged
aquatic vegetation is important to the quality of estuarine ecosystem habitats because it
provides habitat for a variety of aquatic organisms, absorbs excess nutrients, and traps
sediments. Nutrient enrichment is the major driving factor contributing to declines in
submerged aquatic vegetation coverage. The Mid-Atlantic region is the most heavily
impacted area in terms of moderate or high loss of submerged aquatic vegetation due to
eutrophication.
Estuarine and Coastal Aquatic
Estuaries and coastal waters tend to be nitrogen-limited and are therefore inherently
sensitive to increased atmospheric nitrogen loading.278 The U.S. national estuary
condition assessment completed in 2007279 found that the most impacted estuaries in the U.S.
occurred in the mid- Atlantic region and the estuaries with the lowest symptoms of
eutrophication were in the North Atlantic. Nitrogen nutrient enrichment is a major
environmental problem for coastal regions of the U.S., especially in the eastern and Gulf
Coast regions. Of 138 estuaries examined in the national estuary assessment, 44 were
identified as showing symptoms of nutrient over-enrichment. Estuaries are among the most
biologically productive ecosystems on Earth and provide critical habitat for an enormous
diversity of life forms, especially fish. Of the 23 estuaries examined in the national
3-63
-------
assessment in the Northeast, 61% were classified as moderately to severely degraded.280
Other regions had mixtures of low, moderate, and high degree of eutrophication (See Figure
3.3-11).
r
8
•.
.-..
~v--V"-V
A
.
U*. V
V"
S 40
1 M
•
•3 ZD;
£ 13'
18
s
\
V
2&
17
M
n
y»TniH*lKi!ffc'»»J^
It" f^f
.*=o.d
1S
^
_^_M
r
' i Mod*rat* high:
' ' Modera
occu' l&i
li
nd4H war A "*J (-•'vi » c
1 OVBI vrndim arff
taf «w a am»I u> fnedlum area
Low: few 3ii HJJUIIB D:CUI at more Ihwi mtntml levels.
UMknoraMt1 AU.'f A^lf
-------
studies.284'285'286'287'288 Other factors contributing to the sensitivity of soils and surface waters
to acidifying deposition, include: topography, soil chemistry, land use, and hydrologic flow
path.
Terrestrial
Acidifying deposition has altered major biogeochemical processes in the U.S. by
increasing the nitrogen and sulfur content of soils, accelerating nitrate and sulfate leaching
from soil to drainage waters, depleting base cations (especially calcium and magnesium) from
soils, and increasing the mobility of aluminum. Inorganic aluminum is toxic to some tree
roots. Plants affected by high levels of aluminum from the soil often have reduced root
growth, which restricts the ability of the plant to take up water and nutrients, especially
calcium.289 These direct effects can, in turn, influence the response of these plants to climatic
stresses such as droughts and cold temperatures. They can also influence the sensitivity of
plants to other stresses, including insect pests and disease290 leading to increased mortality of
canopy trees. In the U.S. terrestrial effects of acidification are best described for forested
ecosystems (especially red spruce and sugar maple ecosystems) with additional information
on other plant communities, including shrubs and lichen.291 There are several indicators of
stress to terrestrial vegetation including percent dieback of canopy trees, dead tree basal area
(as a percent), crown vigor index and fine twig dieback.292
Health, Vigor, and Reproduction of Tree Species in Forests
Both coniferous and deciduous forests throughout the eastern U.S. are experiencing
gradual losses of base cation nutrients from the soil due to accelerated leaching for acidifying
deposition. This change in nutrient availability may reduce the quality of forest nutrition over
the long term. Evidence suggests that red spruce and sugar maple in some areas in the eastern
U.S. have experienced declining health as a consequence of this deposition. For red spruce,
(picea rubens) dieback or decline has been observed across high elevation landscapes of the
northeastern U.S., and to a lesser extent, the southeastern U.S. Acidifying deposition has
been implicated as a causal factor.293 Since the 1980s, red spruce growth has increased at
both the higher- and lower-elevation sites corresponding to a decrease in SO2 emissions in the
U.S. (to about 20 million tons/year by 2000), while NOX emissions held fairly steady (at
about 25 million tons/year). Research indicates that annual emissions of sulfur plus NOx
explained about 43% of the variability in red spruce tree ring growth between 1940 and 1998,
while climatic variability accounted for about 8% of the growth variation for that period.294
The observed dieback in red spruce has been linked, in part, to reduced cold tolerance of the
spruce needles, caused by acidifying deposition. Results of controlled exposure studies
showed that acidic mist or cloud water reduced the cold tolerance of current-year needles by 3
to 10° F.295 More recently studies have found a link between availability of soil calcium and
winter injury.296 Figure 3.3-12 shows the distribution of red spruce (brown) and sugar maple
(green) in the eastern U.S.
3-65
-------
Figure 3.3-12 Distribution of Red Spruce (pink) and Sugar Maple (green) in the Eastern U.S.
297
In hardwood forests, species nutrient needs, soil conditions, and additional stressors
work together to determine sensitivity to acidifying deposition. Stand age and successional
stage also can affect the susceptibility of hardwood forests to acidification effects. In
northeastern hardwood forests, older stands exhibit greater potential for calcium depletion in
response to acidifying deposition than younger stands. Thus, with the successional change
from pin cherry (Prunus pensylvanica), striped maple (Acerpensylvanicum), white ash
(Fraxinus americana), yellow birch and white birch (Betula papyrifera) in younger stands to
beech and red maple in older stands, there is an increase in sensitivity to acidification.298
Sugar maple (Acer saccharum) is the deciduous tree species of the northeastern U.S.
and central Appalachian Mountain region (See Figure 3-14) that is most commonly associated
with adverse acidification-related effects of nitrogen and sulfur deposition.299 In general,
evidence indicates that acidifying deposition in combination with other stressors is a likely
contributor to the decline of sugar maple trees that occur at higher elevation, on geologies
dominated by sandstone or other base-poor substrate, and that have base-poor soils having
high percentages of rock fragments.300
Loss of calcium ions in the base cations has also been implicated in increased
susceptibility of flowering dogwood (Cornus ttorida) to its most destructive disease, dogwood
anthracnose- a mostly fatal disease. Figure 3.3-13 shows the native range of flowering
dogwood in the U.S. (dark gray) as well as the range of the anthracnose disease as of 2002 in
the eastern U.S. (red). Flowering dogwood is a dominant understory species of hardwood
forests in the eastern U.S.301
3-66
-------
Figure 3.3-13 Native Range of Flowering Dogwood (dk gray) and the Documented Range of Dogwood
Anthracnose (red) Source: Holzmueler et al (2006)
Limited data exists on the possible effects of nitrogen and sulfur deposition on the
acid-based characteristics of forests in the U.S. other than spruce-fire and northern hardwood
forests ecosystems as described above.302
Health and Biodiversity of Other Plant Communities
Shrubs
Available data suggest that it is likely that a variety of shrub and herbaceous species
are sensitive to base cation depletion and/or aluminum toxicity. However, conclusive
evidence is generally lacking.303
Lichens
Lichens and bryophytes are among the first components of the terrestrial ecosystem to
be affected by acidifying deposition.304 Vulnerability of lichens to increased nitrogen input is
generally greater than that of vascular plants.305 Even in the Pacific Northwest, which
receives uniformly low levels of nitrogen deposition, changes from acid-sensitive and
nitrogen-sensitive to pollution tolerant nitrophillic lichen taxa are occurring in some areas.306
Lichens remaining in areas affected by acidifying deposition were found to contain almost
exclusively the families Candelariaccae, Physciaceae, and Teloschistaceae.307
Effects of sulfur dioxide exposure to lichens includes: reduced photosynthesis and
respiration, damage to the algal component of the lichen, leakage of electrolytes, inhibition of
nitrogen fixation, reduced K absorption, and structural changes.308 Additional research has
concluded that the sulfunnitrogen exposure ratio is as important as pH in causing toxic effects
on lichens. Thus, it is not clear to what extent acidity may be the principal stressor under high
levels of air pollution exposure. The toxicity of sulfur dioxide to several lichen species is
3-67
-------
greater under acidic conditions than under neutral conditions.309 The effects of excess
nitrogen deposition to lichen communities are discussed in Section 3.3.1.5.
Artie and Alpine Tundra
The possible effects of acidifying deposition on arctic and alpine plant communities
are also of concern to the U.S.310 Especially important in this regard is the role of nitrogen
deposition in regulating ecosystem nitrogen supply and plant species composition. Soil
acidification and base cation depletion in response to acidifying deposition have not been
documented in arctic or alpine terrestrial ecosystems in the U.S. Such ecosystems are rare
and spatially limited in the eastern U.S., where acidifying deposition levels have been high.
These ecosystems are more widely distributed in the western U.S. and throughout much of
Alaska, but acidifying deposition levels are generally low in these areas. Key concerns are for
listed threatened or endangered species and species diversity.
Aquatic Ecosystems
Aquatic effects of acidification have been well studied in the U.S. and elsewhere at
various trophic levels. These studies indicate that aquatic biota have been affected by
acidification at virtually all levels of the food web in acid sensitive aquatic ecosystems.
Effects have been most clearly documented for fish, aquatic insects, other invertebrates, and
algae.
Biological effects are primarily attributable to a combination of low pH and high
inorganic aluminum concentrations. Such conditions occur more frequently during rainfall
and snowmelt that cause high flows of water and less commonly during low-flow conditions,
except where chronic acidity conditions are severe. Biological effects of episodes include
reduced fish condition factor, changes in species composition and declines in aquatic species
richness across multiple taxa, ecosystems and regions. These conditions may also result in
direct mortality.311 Biological effects in aquatic ecosystems can be divided into two major
categories: effects on health, vigor, and reproductive success; and effects on biodiversity.
3.3.1.8 Nitrogen and Sulfur Deposition Maps for the U.S - Contribution of
International Shipping in 2020 with and without an ECA
Air quality modeling conducted by the U.S. government shows that without any
further emission controls, in 2020, shipping activities will contribute to the serious problems
of acidification and nutrient enrichment in the U.S by adding significant amounts to nitrogen
and sulfur deposition across the U.S. Specifically, in 2020, annual total sulfur deposition
attributable to international shipping will range from 10% to more than 25% of total sulfur
deposition along the entire Atlantic, Gulf of Mexico, and Pacific coastal areas of the U.S. and
this level of impact will extend inland for hundreds of kilometers (See Figure 3.3-14). Of
equal significance, international shipping will contribute to total annual sulfur deposition not
only along all U.S. coastal areas but throughout the entire U.S. land mass, impacting sensitive
terrestrial and aquatic ecosystems in the vast interior and heartland regions of the U.S.
3-68
-------
Contributions to sulfur deposition will range from 1% to 5% in ecosystems located
throughout the interior sections of the U.S.
Percent Change in Annual Total Sulfur Deposition
2020ce zero minus 2020ce
Figure 3.3-14 Percent Contribution in 2020 of Ships to Annual Total Sulfur Deposition in the U.S.
With respect to nitrogen deposition, in 2020, annual total nitrogen deposition from
international shipping will range from about 9% to more than 25% along the entire U.S.
Atlantic, Pacific and Gulf of Mexico coastal areas. Nitrogen deposition from international
shipping will also extend inland for hundreds of kilometers. In addition, throughout the
remaining land areas of the U.S., international shipping will also contribute to annual total
nitrogen deposition-in the range of 1% to 5% by 2020 (See Figure 3.3-15).
3-69
-------
Percent Change in Annual Total Nitrogen Deposition
2020ce zero minus 2020ce
Figure 3.3-15 Percent Contribution in 2020 of Ships to Annual Total Nitrogen Deposition in the U.S.
If the proposed EGA were adopted, reductions in nitrogen deposition would result by
2020, benefiting many sensitive ecological areas throughout the U.S. Areas benefiting are
described in detail in section 3.3.1.1 and include sensitive forests, wetlands such as freshwater
bogs and marshes, lakes and streams throughout the entire U.S. Figure 3.3-16 illustrates the
nitrogen deposition reductions that would occur along U.S. coastlines in 2020 as well as
reductions occurring within the interior of the U.S. Reductions would range from 3% to 7%
along the entire Atlantic and Gulf Coasts with a few regions, such as southern Louisiana and
Florida, experiencing nitrogen reductions up to 9%. Along the Pacific Coast, modeling shows
that nitrogen deposition reductions would be higher, ranging from 3% to 15% on land and as
high as 20% in some coastal waters.
3-70
-------
Percent Change in Annual Total Nitrogen Deposition
2020ce 200nm minus 2020ce
Figure 3.3-16 Percent Change in Annual Total Nitrogen over the U.S. Modeling Domain for the ECA
Modeling Scenario.
With respect to sulfur deposition, adopting the proposed ECA would result in reducing
sulfur deposition levels in 2020; in some regions by more than 25%. Figure 3.3-17 illustrates
the sulfur deposition reductions occurring throughout the U.S. In some individual U.S.
watersheds, consisting of offshore islands or close to coastal areas, sulfur deposition levels
would be reduced by up to 80%. More generally, the Northeast Atlantic Coastal region would
experience sulfur deposition reductions from C3 vessels ranging from 7% to 25% while the
Southeast Atlantic Coastal region would experience reductions ranging from 7% to more than
25%. Sulfur deposition would be reduced in the Gulf Coast region from 3% to more than
25%. Along the West Coast of the U.S. sulfur deposition reductions exceeding 25% would
occur in the entire Los Angeles Basin in the State of California. The Pacific Northwest would
also see significant sulfur deposition reductions ranging from 4% to more than 25%. As
importantly, sulfur reductions due to the proposed ECA would also impact the entire U.S.
land mass with even interior sections of the U.S. experiencing reductions of 1%. Together,
these reductions would assist the U.S. in its efforts to reduce acidification impacts associated
with nitrogen and sulfur deposition in both terrestrial and aquatic ecosystems in coastal areas
of the U.S. as well as within the interior of the U.S.
3-71
-------
Percent Change in Annual Total Sulfur Deposition
2020ce 200nm minus 2020ce
Figure 3.3-17 Percent Change in Annual Total Sulfur over the U.S. Modeling Domain for the ECA
Modeling Scenario.
Appendix 3B presents both the range as well as the average total nitrogen and total
sulfur deposition changes in 2020 for CMAQ modeling scenarios over 18 specific U.S.
subregions. In the case of the proposed ECA, sulfur deposition levels were reduced by on
average from 0 to 19 percent over these large drainage regions. In individual HUCs
consisting of offshore islands or close to coastal areas, sulfur deposition levels in 2020 were
improved by as much as 78% in the proposed ECA while nitrogen deposition levels were
improved by as much as 13% in some coastal areas.
3.3.1.8.1 Methodology
The CMAQ model provides estimates of the amount of nitrogen and sulfur deposition
in each of the simulated scenarios. The modeling indicated that the shipping sector
contributes to acid deposition over the U.S. modeling domain and that these impacts will
grow by 2020, if no control measures are adopted by then. Figures 3-16 and 3-17 show the
percent change in total nitrogen and total sulfur deposition in 2020 expected to result from the
application of the proposed ECA. These plots are based on absolute outputs from the CMAQ
modeling.
3-72
-------
Additionally, we conducted additional analyses using a separate methodology in
which the CMAQ outputs were used to estimate the impacts on deposition levels in a manner
similar to how the model is used for ozone and fine particulate matter. In this methodology,
CMAQ outputs of annual wet deposition from the 2002 base year model run are used in
conjunction with annual wet deposition predictions from the control or future case scenarios
to calculate relative reduction factors (RRFs) for wet deposition. Separate wet deposition
RRFs are calculated for reduced nitrogen, oxidized nitrogen, and sulfur. These RRFs are
multiplied by the corresponding measured annual wet deposition of reduced nitrogen,
oxidized nitrogen, and sulfur from the National Atmospheric Deposition Program (NADP)
network. The result will be a projection of the NADP wet deposition for the control or future
case scenarios. The projected wet deposition for each of the three species is added to the
CMAQ-predicted dry deposition for each of these species to produce total reduced nitrogen,
total oxidized nitrogen, and total sulfur deposition for the control/future case scenario. The
reduced and oxidized nitrogen depositions are summed to calculate total nitrogen deposition.
This analysis was completed for each individual 8-digit hydrological unit code (HUC)
within the U.S. modeling domain. Each 8-digit HUC represents a local drainage basin. There
were 2,108 8-digit HUCs considered as part of this analysis. This assessment corroborated
the absolute deposition modeling results. Appendix 3B shows the average total nitrogen and
total sulfur deposition changes for three CMAQ modeling scenarios over 18 specific
subregions. In the case of an EGA adoption, sulfur deposition levels were reduced by 0 to 19
percent over these large drainage regions. In individual HUCs consisting of offshore islands
or close to coastal areas, sulfur deposition levels were improved by as much as 78% in the
EGA case. Nitrogen deposition levels were improved by as much as 13% in some coastal
areas.
3.3.1.9 Case Study: Critical Load Modeling in the Adirondack Mountains of New York
State and the Blue Ridge Mountains in the State of Virginia
The Adirondack Mountains of New York and the Blue Ridge Mountains of Virginia
have long been a locus for awareness of the environmental issues related to acidifying
deposition. Soils and water bodies, such as lakes and streams, usually buffer the acidity from
natural rain with "bases," the opposite of acids from the environment. The poor buffering
capability of the soils in both these regions make the lakes and streams particularly
susceptible to acidification from anthropogenic nitrogen and sulfur atmospheric deposition
resulting from nitrogen and sulfur oxides emissions. Consequently, acidic deposition has
affected hundreds of lakes and thousands of miles of headwater streams in both of these
regions. The diversity of life in these acidic waters has been reduced as a result of acidic
deposition.
The critical load approach provides a quantitative estimate of the exposure to one or
more pollutants below which significant harmful effects on specific sensitive elements of the
environment do not occur according to present knowledge. The critical load for a lake or
stream provides a means to gauge the extent to which a water body has recovered from past
acid deposition, or is potentially at risk due to current deposition levels. Acid neutralizing
capacity (ANC) is an excellent indicator of the health of aquatic organisms such as fish,
insects, and invertebrates.
3-73
-------
•. '• 1* '
' . , -J • »
•
•
|
C 40 E
Figure 3.3-18 Locations of lakes and streams used in this assessment
In this case study, the focus is on the combined load of nitrogen and sulfur and
deposition below which the ANC level would still support healthy aquatic ecosystems.
Critical loads were calculated for 169 lakes in the Adirondack region and 60 streams in
Virginia (Figure 3.3-18). The Steady-State Water Chemistry (SSWC) model was used to
calculate the critical load, relying on water chemistry data from the USEPA Temporal
Intergraded Monitoring of Ecosystems (TIME) and Long-term Monitoring (LTM) programs
and model assumptions well supported by the scientific literature. Research studies have
shown that surface water with ANC values greater than 50 micro-equivalents per Liter
(p,eq/L) tend to protect most fish (i.e., brook trout, others) and other aquatic organisms (Table
3.3-4). In this case, the critical load represents the combined deposition load of nitrogen and
sulfur to which a lake or stream could be subjected and still have an ANC of 50
3-74
-------
Table 3.3-4 Aquatic Status Categories
CATEGORY LABEL ANC LEVELS* EXPECTED ECOLOGICAL EFFECTS
Acute
Concern
<0 micro
equivalent
per Liter
(jieq/L)
Complete loss offish populations is expected. Planktonic
communities have extremely low diversity and are dominated by
acidophilic forms. The numbers of individuals in plankton species
that are present are greatly reduced.
Severe
Concern
0-20
jieq/L
Highly sensitive to episodic acidification. During episodes of high
acid deposition, brook trout populations may experience lethal
effects. Diversity and distribution of zooplankton communities
decline sharply.
Elevated
Concern
20-50
jieq/L
Fish species richness is greatly reduced (more than half of expected
species are missing). On average, brook trout populations experience
sub-lethal effects, including loss of health and reproduction (fitness).
Diversity and distribution of zooplankton communities also decline.
Moderate
Concern
50-100
jieq/L
Fish species richness begins to decline (sensitive species are lost
from lakes). Brook trout populations are sensitive and variable, with
possible sub-lethal effects. Diversity and distribution of zooplankton
communities begin to decline as species that are sensitive to acid
deposition are affected.
Low
Concern
>100 jieq/L
Fish species richness may be unaffected. Reproducing brook trout
populations are expected where habitat is suitable. Zooplankton
communities are unaffected and exhibit expected diversity and
range.
When the critical load is "exceeded," it means that the amount of combined nitrogen
and sulfur atmospheric deposition is greater than the critical load for a particular lake or
stream, preventing the water body from reaching or maintaining an ANC concentration of 50
Lieq/L. Critical loads of combined total nitrogen and sulfur are expressed in terms of ionic
charge balance as milliequivalent per square meter per year (meq/m2/yr). Exceedances were
calculated from deposition for years 2002 and 2020 with and without emissions from
shipping. In year 2002, there was no difference in the percent of lakes or streams in both
regions that exceeded the critical load for the case with and without ship emissions (Table
3.3-5). For the year 2020, when ship emissions are present, 33% of lakes in the Adirondack
Mountains and 52% of streams in the Virginia Blue Ridge Mountains received greater acid
deposition than could be neutralized. When ship emissions were removed from the modeling
domain for the year 2020, 31 and 50 percent of lakes and streams, respectively, received
greater acid deposition than could be neutralized- a 2% improvement.
Regional Assessment
A regional estimate of the benefits of the reduction in international shipping emissions
in 2020 can be derived from scaling up the results from 169 lakes to a larger population of
lakes in the Adirondack Mountains. One hundred fifteen lakes of the 169 lakes modeled for
critical loads are part of a subset of 1,842 lakes in the Adirondacks, which include all lakes
from 0.5 to 2000 ha in size and at least 0.5 meters in depth. Using weighting factors derived
from the EMAP probability survey and the critical load calculations from the 115 lakes,
3-75
-------
exceedance estimates were derived for the entire 1,842 lakes in the Adirondacks. Based on
this approach, 66 fewer lakes in the Adirondack Mountains are predicted to receive nitrogen
and sulfur deposition loads below the critical load and would be protected as a result of
removing international shipping emissions in 2020.
Currently, no probability survey has been completed for the study area in Virginia.
However, the 60 trout streams modeled are characteristic of first and second order streams on
non-limestone bedrock in the Blue Ridge Mountains of Virginia. Because of the strong
relationship between bedrock geology and ANC in this region, it is possible to consider the
results in the context of similar trout streams in the Southern Appalachians that have the same
bedrock geology and size. In addition, the 60 streams are a subset of 344 streams sampled by
the Virginia Trout Stream Sensitivity Study, which can be applied to a population of 304 out
of the original 344 streams. Using the 304 streams to which the analysis applies directly as
the total, 6 additional streams in this group would be protected as a result of removing
international shipping emissions in 2020. However, it is likely that many more of the -12,000
trout streams in Virginia would benefit from reduced international shipping emissions given
the extent of similar bedrock geology outside the study area.
Table 3.3-5 Percent of Modeled Lakes that Exceed the Critical Load for Years 2002 and 2020 with and
without International Shipping Emissions. "Zero" Indicates without International Shipping Emissions
2002
2002 ZERO
2020
2020 ZERO
Adirondack Mountains
Exceeded Critical Load
(%. Lakes)
Non-Exceeded Critical Load (%. Lakes)
45
55
45
55
33
73
31
71
Virginia Blue Ridge Mountains
Exceeded Critical Load
(%. Lakes)
Non-Exceeded Critical Load (%. Lakes)
82
18
82
18
52
48
50
50
3-76
-------
2002 with International
Shipping Emliilon1.
^•r
_ • I • 1 ..» •
2002 wntiout Inrarnatio real
Shipping Enrn-.ions
n
*
' **•*; •*. .
^.Jr-%- >
;
Critical Ltud Eico Hdences
I > ANC ol SO ueq;'U
Figure 3.3-19 a. 2002
2020 with lute-national
Shipping Emissions
2020 v
?!*?,. :.*v
. • .v s • X
Cri!iLu! Laud Exccodc^iccu
t>
diMX nsl *
dove •£c«
idl tmii
bad
| |
Figure 3.3-19 b. 2020; Critical Load Exceedance for ANC Concentration of 50 jieq/L. Green dots
represent lakes in the Adirondack Mountains where current nitrogen and sulfur deposition is below their
critical load and maintains an ANC concentration of 50 jieq/L. Red dots are lakes where current nitrogen
and sulfur deposition exceeds their limit and the biota are likely impacted
3-77
-------
2002 with International
Shipping Emissions
JOM without i f - .- i. i
Shippir.g Emissions
Critical Load
( > ANC of SO ueq/L)
| | V
Figure 3.3-20 a. 2002
2020 with lnUrnatiofl.il
Shipping I -r. .-!•:•.
2024 v> itrwut international
Snipping Emissions
Critical Load Excellences
( > ANC of 50 ueq/L)
Figure 3.3-20 b . 2020; Critical Load Exceedances for ANC Concentration of 50 jieq/L. Green dots
represent streams in the Virginia Blue Ridge Mountains where current nitrogen and sulfur deposition is
below their critical load and maintains an ANC concentration of 50 jieq/L. Red dots are streams where
current nitrogen and sulfur deposition exceeds their limit and the biota are likely impacted.
3.3.2 Ozone Impacts on Plants and Ecosystems (overview)
There are a number of environmental or public welfare effects associated with the
presence of ozone in the ambient air.312 In this section we discuss the impact of ozone on
plants, including trees, agronomic crops and urban ornamentals.
The Air Quality Criteria Document for Ozone and related Photochemical Oxidants
notes that "ozone affects vegetation throughout the United States, impairing crops, native
vegetation, and ecosystems more than any other air pollutant".313 Like carbon dioxide (CO 2)
and other gaseous substances, ozone enters plant tissues primarily through apertures (stomata)
in leaves in a process called "uptake".314 Once sufficient levels of ozone, a highly reactive
substance, (or its reaction products) reaches the interior of plant cells, it can inhibit or damage
3-78
-------
essential cellular components and functions, including enzyme activities, lipids, and cellular
membranes, disrupting the plant's osmotic (i.e., water) balance and energy utilization
patterns.315'316 If enough tissue becomes damaged from these effects, a plant's capacity to fix
carbon to form carbohydrates, which are the primary form of energy used by plants is
reduced,317 while plant respiration increases. With fewer resources available, the plant
reallocates existing resources away from root growth and storage, above ground growth or
yield, and reproductive processes, toward leaf repair and maintenance, leading to reduced
growth and/or reproduction. Studies have shown that plants stressed in these ways may
exhibit a general loss of vigor, which can lead to secondary impacts that modify plants'
responses to other environmental factors. Specifically, plants may become more sensitive to
other air pollutants, more susceptible to disease, insect attack, harsh weather (e.g., drought,
frost) and other environmental stresses. Furthermore, there is evidence that ozone can
interfere with the formation of mycorrhiza, essential symbiotic fungi associated with the roots
of most terrestrial plants, by reducing the amount of carbon available for transfer from the
host to the symbiont.318'319
This ozone damage may or may not be accompanied by visible injury on leaves, and
likewise, visible foliar injury may or may not be a symptom of the other types of plant
damage described above. When visible injury is present, it is commonly manifested as
chlorotic or necrotic spots, and/or increased leaf senescence (accelerated leaf aging). Because
ozone damage can consist of visible injury to leaves, it can also reduce the aesthetic value of
ornamental vegetation and trees in urban landscapes, and negatively affects scenic vistas in
protected natural areas.
Ozone can produce both acute and chronic injury in sensitive species depending on the
concentration level and the duration of the exposure. Ozone effects also tend to accumulate
over the growing season of the plant, so that even lower concentrations experienced for a
longer duration have the potential to create chronic stress on sensitive vegetation. Not all
plants, however, are equally sensitive to ozone. Much of the variation in sensitivity between
individual plants or whole species is related to the plant's ability to regulate the extent of gas
exchange via leaf stomata (e.g., avoidance of ozone uptake through closure of
*39n r\r)\ "399
stomata) ' ' Other resistance mechanisms may involve the intercellular production of
detoxifying substances. Several biochemical substances capable of detoxifying ozone have
been reported to occur in plants, including the antioxidants ascorbate and glutathione. After
injuries have occurred, plants may be capable of repairing the damage to a limited extent.323
Because of the differing sensitivities among plants to ozone, ozone pollution can also
exert a selective pressure that leads to changes in plant community composition. Given the
range of plant sensitivities and the fact that numerous other environmental factors modify
plant uptake and response to ozone, it is not possible to identify threshold values above which
ozone is consistently toxic for all plants. The next few paragraphs present additional
information on ozone damage to trees, ecosystems, agronomic crops and urban ornamentals.
Ozone also has been conclusively shown to cause discernible injury to forest
trees.324'325 In terms of forest productivity and ecosystem diversity, ozone may be the
pollutant with the greatest potential for regional-scale forest impacts. Studies have
3-79
-------
demonstrated repeatedly that ozone concentrations commonly observed in polluted areas can
have substantial impacts on plant function.326'327
Because plants are at the base of the food web in many ecosystems, changes to the
plant community can affect associated organisms and ecosystems (including the suitability of
habitats that support threatened or endangered species and below ground organisms living in
the root zone). Ozone impacts at the community and ecosystem level vary widely depending
upon numerous factors, including concentration and temporal variation of tropospheric ozone,
species composition, soil properties and climatic factors In most instances, responses to
chronic or recurrent exposure in forested ecosystems are subtle and not observable for many
years. These injuries can cause stand-level forest decline in sensitive ecosystems.329'330'331 It
is not yet possible to predict ecosystem responses to ozone with much certainty; however,
considerable knowledge of potential ecosystem responses has been acquired through long-
term observations in highly damaged forests in the United States.
Laboratory and field experiments have also shown reductions in yields for agronomic
crops exposed to ozone, including vegetables (e.g., lettuce) and field crops (e.g., cotton and
wheat). The most extensive field experiments, conducted under the National Crop Loss
Assessment Network (NCLAN) examined 15 species and numerous cultivars. The NCLAN
results show that "several economically important crop species are sensitive to ozone levels
typical of those found in the United States."332 In addition, economic studies have shown
reduced economic benefits as a result of predicted reductions in crop yields associated with
observed ozone levels.333'334'335
Urban ornamentals represent an additional vegetation category likely to experience
some degree of negative effects associated with exposure to ambient ozone levels. It is
estimated that more than $20 billion (1990 dollars) are spent annually on landscaping using
ornamentals, both by private property owners/tenants and by governmental units responsible
for public areas.336 This is therefore a potentially costly environmental effect. However, in
the absence of adequate exposure-response functions and economic damage functions for the
potential range of effects relevant to these types of vegetation, no direct quantitative analysis
has been conducted.
Air pollution can have noteworthy cumulative impacts on forested ecosystems by
affecting regeneration, productivity, and species composition.337 In the U.S., ozone in the
lower atmosphere is one of the pollutants of primary concern. Ozone injury to forest plants
can be diagnosed by examination of plant leaves. Foliar injury is usually the first visible sign
of injury to plants from ozone exposure and indicates impaired physiological processes in the
leaves/38
This indicator is based on data from the U.S. Department of Agriculture (USDA)
Forest Service Forest Inventory and Analysis (FIA) program. As part of its Phase 3 program,
formerly known as Forest Health Monitoring, FIA examines ozone injury to ozone-sensitive
plant species at ground monitoring sites in forest land across the country. For this indicator,
forest land does not include woodlots and urban trees. Sites are selected using a systematic
sampling grid, based on a global sampling design.339'340 At each site that has at least 30
individual plants of at least three ozone-sensitive species and enough open space to ensure
3-80
-------
that sensitive plants are not protected from ozone exposure by the forest canopy, FIA looks
for damage on the foliage of ozone-sensitive forest plant species. Because ozone injury is
cumulative over the course of the growing season, examinations are conducted in July and
August, when ozone injury is typically highest.
Monitoring of ozone injury to plants by the USDA Forest Service has expanded over
the last 10 years from monitoring sites in ten states in 1994 to nearly 1,000 monitoring sites in
41 states in 2002. The data underlying this indicator are based on averages of all observations
collected in 2002, the latest year for which data are publicly available at the time the study
was conducted, and are broken down by EPA Region. Ozone damage to forest plants is
classified using a subjective five-category biosite index based on expert opinion, but designed
to be equivalent from site to site. Ranges of biosite values translate to no injury, low or
moderate foliar injury (visible foliar injury to highly sensitive or moderately sensitive plants,
respectively), and high or severe foliar injury, which would be expected to result in tree-level
or ecosystem-level responses, respectively.341'342
3.3.2.1 Recent Ozone Impact Data for the U.S.
There is considerable regional variation in ozone-related visible foliar injury to
sensitive plants in the U.S. The U.S. EPA has developed an environmental indicator based on
data from the U.S. Department of Agriculture (USDA) Forest Service Forest Inventory and
Analysis (FIA) program which examines ozone injury to ozone-sensitive plant species at
ground monitoring sites in forest land across the country (This indicator does not include
woodlots and urban trees). Sites are selected using a systematic sampling grid, based on a
global sampling design.343'344 Because ozone injury is cumulative over the course of the
growing season, examinations are conducted in July and August, when ozone injury is
typically highest. The data underlying the indictor in Figure 3.3-21 are based on averages of
all observations collected in 2002, the latest year for which data are publicly available at the
time the study was conducted, and are broken down by U.S. EPA Regions. Ozone damage to
forest plants is classified using a subjective five-category biosite index based on expert
opinion, but designed to be equivalent from site to site. Ranges of biosite values translate to
no injury, low or moderate foliar injury (visible foliar injury to highly sensitive or moderately
sensitive plants, respectively, and high or severe foliar injury, which would be expected to
result in tree-level or ecosystem-level responses, respectively.345
The highest percentages of observed high and severe foliar injury, those which are
most likely to be associated with tree or ecosystem-level responses, are primarily found in the
Mid-Atlantic and Southeast regions. In EPA Region 3 (which comprises the States of
Pennsylvania, West Virginia, Virginia, Delaware, Maryland and Washington D.C.), 12
percent of ozone-sensitive plants showed signs of high or severe foliar damage, and in
Regions 2 (States of New York, New Jersey), and 4 (States of North Carolina, South Carolina,
Kentucky, Tennessee, Georgia, Florida, Alabama, and Mississippi) the values were 10 percent
and 7 percent, respectively. The sum of high and severe ozone injury ranged from 2 percent
to 4 percent in EPA Region 1 (the six New England States), Region 7 (States of Missouri,
Iowa, Nebraska and Kansas), and Region 9 (States of California, Nevada, Hawaii and
Arizona). The percentage of sites showing some ozone damage was about 45 percent in each
of these EPA Regions.
3-81
-------
Degree ol injury:
Nana
Low
Moderate
High
Ssvsra
Perterrl ol monitoring sites rn each taSeg&n/:
Realm 1
[54 sites)
Region 2
[42 sites)
Region 3
[111 sites)
Ragicn 4
(227 sites}
Regicn 5
! 180 sites}
Regicn 6
[59 sites)
Region 7
(63 sites'!
Regicn 8
(72 sites'!
Regicn 9
{80 sites)
Regicn 10
1-57 sites)
68.5
15.7
-•3.
61.9
21.4
71
7,1
24
55.9
IB 'D
144
72
4.5
75
1D.
G
1BJ
U
94,9
5.1
85.7
9.5
TO2
.:•! 5
100.0
12,5
9.8
tOO.D
a: 945 moniitmg sites,
tocatad in 41
:TDtals may
raurtdiiHj.
Data spares: I/f^i forasf Semce,
ans
EPA
Figure 3.3-21 Ozone Injury to Forest Plants in U.S. by EPA Regions, 2002ab
3.3.2.1.1 Indicator Limitations
Field and laboratory studies were reviewed to identify the forest plant species in each
region that are highly sensitive to ozone air pollution. Other forest plant species, or even
genetic variants of the same species, may not be harmed at ozone levels that cause effects on
the selected ozone-sensitive species.
3-82
-------
Because species distributions vary regionally, different ozone-sensitive plant species
were examined in different parts of the country. These target species could vary with
respect to ozone sensitivity, which might account for some of the apparent differences in
ozone injury among regions of the U.S.
Ozone damage to foliage is considerably reduced under conditions of low soil
moisture, but most of the variability in the index (70 percent) was explained by ozone
concentration.346 Ozone may have other adverse impacts on plants (e.g., reduced
productivity) that do not show signs of visible foliar injury.3
Though FIA has extensive spatial coverage based on a robust sample design, not all
forested areas in the U.S. are monitored for ozone injury. Even though the biosite data have
been collected over multiple years, most biosites were not monitored over the entire period,
so these data cannot provide more than a baseline for future trends.
3.3.2.1.2 Ozone Impacts on Forest Health
Air pollution can impact the environment and affect ecological systems, leading to
changes in the biological community (both in the diversity of species and the health and vigor
of individual species). As an example, many studies have shown that ground-level ozone
reduces the health of plants including many commercial and ecologically important forest tree
species throughout the United States.348
When ozone is present in the air, it can enter the leaves of plants, where it can cause
significant cellular damage. Since photosynthesis occurs in cells within leaves, the ability of
the plant to produce energy by photosynthesis can be compromised if enough damage occurs
to these cells. If enough tissue becomes damaged it can reduce carbon fixation and increase
plant respiration, leading to reduced growth and/or reproduction in young and mature trees.
Ozone stress also increases the susceptibility of plants to disease, insects, fungus, and other
environmental stressors (e.g., harsh weather). Because ozone damage can consist of visible
injury to leaves, it also reduces the aesthetic value of ornamental vegetation and trees in urban
landscapes, and negatively affects scenic vistas in protected natural areas.
Assessing the impact of ground-level ozone on forests in the eastern United States
involves understanding the risks to sensitive tree species from ambient ozone concentrations
and accounting for the prevalence of those species within the forest. As a way to quantify the
risks to particular plants from ground-level ozone, scientists have developed ozone-
exposure/tree-response functions by exposing tree seedlings to different ozone levels and
measuring reductions in growth as "biomass loss." Typically, seedlings are used because they
are easy to manipulate and measure their growth loss from ozone pollution. The mechanisms
of susceptibility to ozone within the leaves of seedlings and mature trees are identical, and the
decreases predicted using the seedlings should be related to the decrease in overall plant
fitness for mature trees, but the magnitude of the effect may be higher or lower depending on
the tree species.349
Some of the common tree species in the United States that are sensitive to ozone are
black cherry (Prunus serotina), tulip-poplar (Liriodendron tulipifera), eastern white pine
3-83
-------
(Pinus strobus). Ozone-exposure/tree-response functions have been developed for each of
these tree species, as well as for aspen (Populus tremuliodes), and ponderosa pine (Pinus
ponderosa). Other common tree species, such as oak (Quercusspp.) and hickory (Carya
spp.), are not nearly as sensitive to ozone. Consequently, with knowledge of the distribution
of sensitive species and the level of ozone at particular locations, it is possible to estimate a
"biomass loss" for each species across their range.
3.3.2.2 W126 Modeling and Projected Impact of Ship Emissions on U.S. Forests
Biomass
To estimate the biomass loss for the tree species listed above across the eastern United
States, the biomass loss for each of the five tree species was calculated using the three-month
12-hour W126 exposure metric at each location and its individual ozone-exposure/tree-
response functions. The W126 exposure metric was calculated using monitored data from the
AQS air quality monitoring sites. This analysis was done for 2020 with and without
international snipping emissions to determine the benefit of lowering shipping emissions on
these sensitive tree species in the Eastern half of the U.S.
The biomass loss in the eastern U.S. attributable to international shipping appears to
range from 0-6.5 % depending on the particular species. The most sensitive species in the
U.S. to ozone-related biomass loss is black cherry; the area of its range with more than 10%
biomass loss in 2020 decreased by 8.5% when emissions from ships were removed.
Likewise, Table 3-6 indicates that yellow-poplar, eastern white pine, aspen, and ponderosa
pine saw areas with more then 2% biomass loss reduced by 2.1% to 3.8% in 2020. The 2%
level of biomass loss is important, because a scientific consensus workshop on ozone effects
reported that a 2% annual biomass loss causes long term ecological harm due to the potential
for compounding effects over multiple years as short-term negative effects on seedlings affect
long-term forest health. 350'351 Figure 3.3-22 shows ship emissions' adverse impact on U.S.
forest biomass loss in 2020.
Table 3.3-6 The Percent Improvement in Area of the Tree Species Range Between the "Base Case" and
"Zero Out" Marine Emissions with Biomass Loss of Greater than 2,4,6, and 10% due to Ozone for Year
2020. Units are % Improvement of Area of Species Range.
Tree Species Pe.rcent. of Bionuiss Loss
2«t« 4°-o 6*0 10*o
r
L.t f
Aspen
ojwtt is tivinitlo h 'lex
Bin ok Cherry
Pnnins s&rofiiKt
Poncleiosa Puie
Pinia po'i'tck rota
tulip Poplai
E. Wliire Pine
24
ii;i
3«
2.1
2.8
1
11.
•i
0
1
4
c ...
.0
.8
1
0
i
1
n.
0
.s
.9
^
£.
4
u;
8.
H-'
IV
Ii.;
3
5
?t
a
n
Piiiiis a {robsis
ii.t. - mt tlinnet in Ilio ai tn
ii-a— oul
-------
2020 Base minus Zero C3 Emissions
| rtipon i Papa/in !.-Kriu1u/ttai)
I fitick LVnirryfPriinL's _THTil'.i»i
I P-rindfrnr-1^! Pir» iPjf
IE. WDte "if iftttts stobas)
Figure 3.3-22 U.S. Geographic Areas where the Proposed ECA would Reduce Biomass Loss by More than
2%
3.3.2.2.1 Methodology
Outputs from the CMAQ modeling were used to calculate a longer-term ozone
exposure metric known as "W126".352 Previous EPA analyses have concluded that the
cumulative, seasonal W126 index is the most appropriate index for relating vegetation
response to ambient ozone exposures. The metric is a sigmoidally weighted 3-month sum of
all hourly ozone concentrations observed during the daily 12-hr period between 8 am to 8 pm.
The three months are the maximum consecutive three months during the ozone season,
defined in the ECA modeling as May through September.
As in the ozone and PM2.s modeling, the CMAQ model was used in a relative sense to
estimate how ambient W126 levels would change as a result of future growth and/or ECA
emissions reductions. The resultant W126 outputs were fed into a separate model which
calculated biomass loss from certain tree species as a result of prolonged exposure to ozone.
The results of that analysis are discussed below. The CMAQ modeling estimated that ship
emissions contributed to high levels of W126 in some coastal areas. This contribution was
estimated to range from as much as 30-40 percent in parts of California and Florida. The
average contribution from all ship emissions was 8 percent nationally.
3.3.3 Visibility Overview
Emissions from international shipping activity contribute to poor visibility in the U.S.
through their primary PM2.5 and NOx emissions (which contribute to the formation of
secondary PM2.s). These airborne particles degrade visibility by scattering and absorbing
3-85
-------
light. Good visibility increases the quality of life where individuals live and work, and where
they engage in recreational activities.
Modeling undertaken for the EGA proposal shows that international shipping activities
negatively impact visibility by contributing to urban haze in U.S. cities which are located near
major deep sea ports and also as regional haze in national parks and wilderness areas
throughout the U.S. The U.S. government places special emphasis on protecting visibility in
national parks and wilderness areas. Section 169 of the Clean Air Act requires the U.S.
government to address existing visibility impairment and future visibility impairment in the
156 national parks exceeding 6,000 acres, and wilderness areas exceeding 5,000 acres, which
are categorized as mandatory class I federal areas.
Based on modeling for the EGA proposal, international shipping activities in 2002
contributed to visibility degradation at all of the 133 class I federal areas which have complete
Interagency Monitoring of Protected Visual Environments (IMPROVE) ambient data for
2002 or are represented by IMPROVE monitors with complete data.2 Absent further
emission controls, by 2020, international shipping activities will have an even larger impact
on visibility impairment in these class I federal areas. For example, in 2002, approximately
4% of visibility impairment in southern California's Agua Tibia Wilderness was due to
shipping activity. U.S. modeling, conducted as part of the EGA proposal, indicates that by
2020 approximately 12.5% of visibility impairment in Agua Tibia will be due to shipping.
Likewise, in 2002, 2.7% of visibility impairment in southern Florida's Everglades National
Park was due to international shipping, and this will double to 6% by 2020. Even in inland
class I federal areas shipping activity is contributing to visibility degradation. In 2020, about
2.5% of visibility degradation in the Grand Canyon National Park, located in the State of
Arizona, will be from international shipping, while almost 6% of visibility degradation in the
State of Washington's North Cascades National Park will be from shipping emissions.
3.3.3.1 Visibility Monitoring
In conjunction with the U.S. National Park Service, the U.S. Forest Service, other
federal land managers, and State organizations in the U.S., the U.S. EPA has supported
visibility monitoring in national parks and wilderness areas since 1988. The monitoring
network was originally established at 20 sites, but it has now been expanded to 110 sites that
represent all but one of the 156 mandatory federal Class I areas across the country. This long-
term visibility monitoring network is known as IMPROVE (Interagency Monitoring of
PROtected Visual Environments).
IMPROVE provides direct measurement of fine particles that contribute to visibility
impairment. The IMPROVE network employs aerosol measurements at all sites, and optical
z There are 156 federally-mandated class I areas which, under the Regional Haze Rule, are required to achieve
natural background visibility levels by 2064. These mandatory class I federal areas are mostly national parks,
national monuments, and wilderness areas. There are currently 116 IMPROVE monitoring sites (representing all
156 mandatory class I federal areas) collecting ambient PM2.5 data at mandatory class I federal areas, but not all
of these sites have complete data for 2002.
3-86
-------
and scene measurements at some of the sites. Aerosol measurements are taken for PMio and
PM2.5mass, and for key constituents of PM2.5, such as sulfate, nitrate, organic and elemental
carbon, soil dust, and several other elements. Measurements for specific aerosol constituents
are used to calculate "reconstructed" aerosol light extinction by multiplying the mass for each
constituent by its empirically-derived scattering and/or absorption efficiency, with adjustment
for the relative humidity. Knowledge of the main constituents of a site's light extinction
"budget" is critical for source apportionment and control strategy development. Optical
measurements are used to directly measure light extinction or its components. Such
measurements are taken principally with either a transmissometer, which measures total light
extinction, or a nephelometer, which measures particle scattering (the largest human-caused
component of total extinction). Scene characteristics are typically recorded 3 times daily with
35 millimeter photography and are used to determine the quality of visibility conditions (such
as effects on color and contrast) associated with specific levels of light extinction as
measured under both direct and aerosol-related methods. Directly measured light extinction
is used under the IMPROVE protocol to cross check that the aerosol-derived light extinction
levels are reasonable in establishing current visibility conditions. Aerosol-derived light
extinction is used to document spatial and temporal trends and to determine how proposed
changes in atmospheric constituents would affect future visibility conditions.
Annual average visibility conditions (reflecting light extinction due to both
anthropogenic and non-anthropogenic sources) vary regionally across the U.S. The rural East
generally has higher levels of impairment than remote sites in the West, with the exception of
urban-influenced sites such as San Gorgonio Wilderness (CA) and Point Reyes National
Seashore (CA), which have annual average levels comparable to certain sites in the Northeast.
Regional differences are illustrated by Figures 4-39a and 4-39b in the CD, which show that,
for class I areas, visibility levels on the 20% haziest days in the West are about equal to levels
on the 20% best days in the East (CD, p. 4-179).
Higher visibility impairment levels in the East are due to generally higher
concentrations of anthropogenic fine particles, particularly sulfates, and higher average
relative humidity levels. In fact, sulfates account for 60-86% of the haziness in eastern sites
(CD, p. 4-236). Aerosol light extinction due to sulfate on the 20% haziest days is
significantly larger in eastern class I areas as compared to western areas (CD, p. 4-182;
Figures 4-40a and 4-40b). With the exception of remote sites in the northwestern U.S.,
visibility is typically worse in the summer months. This is particularly true in the
Appalachian region, where average light extinction in the summer exceeds the annual average
by 40% (Sisler et al., 1996).
3.3.3.2 Addressing Visibility in the U.S.
The U.S. EPA has two programmatic approaches to address visibility. First, to
address the welfare effects of PM on visibility, EPA set secondary PM2.5 standards which
would act in conjunction with the establishment of a regional haze program. In setting this
secondary standard EPA concluded that PM2.s causes adverse effects on visibility in various
locations, depending on PM concentrations and factors such as chemical composition and
average relative humidity. Second, section 169 of the Clean Air Act provides additional
authority to address existing visibility impairment and prevent future visibility impairment in
3-87
-------
the 156 national parks, forests and wilderness areas categorized as mandatory class I federal
areas (62 FR 38680-81, July 18, 1997).AA Figure 3-18 below identifies where each of these
parks are located in the U.S. In July 1999 the regional haze rule (64 FR 35714) was put in
place to protect the visibility in mandatory class I federal areas. Visibility can be said to be
impaired in both PM2.5 nonattainment areas and mandatory class I federal areas. OGVs,
powered by Category 3 engines, contribute to visibility concerns in these areas through their
primary PM2.5 emissions and their NOx and SOx emissions which contribute to the formation
of secondary PM2.5.
Produced by NFS Air Resources Division
* Rainbow Lake, Wl and Bradwell Bay, FL are Class 1 Areas
where visibility is not an important air quality related value
'HWWDI Vgfc*nc*s
Figure 3.3-23 Mandatory Class I Areas in the U.S.
3.3.3.2.1 Current Visibility Impairment
Recently designated PM2.5 nonattainment areas indicate that, as of December 2008,
over 88 million people live in nonattainment areas for the 1997 PM2.5 NAAQS. Thus, at least
AA These areas are defined in section 162 of the Act as those national parks exceeding 6,000 acres, wilderness
areas and memorial parks exceeding 5,000 acres, and all international parks which were in existence on August
7, 1977.
BB As mentioned above, the EPA has recently proposed to amend the PM NAAQS (71 FR 2620, Jan. 17, 2006).
The proposal would set the secondary NAAQS equal to the primary standards for both PM2.5 and PM10-2.5. EPA
also is taking comment on whether to set a separate PM2.5 standard, designed to address visibility (principally in
urban areas), on potential levels for that standard within a range of 20 to 30 |ag/m3, and on averaging times for
the standard within a range of four to eight daylight hours.
3-*
-------
these populations would likely be experiencing visibility impairment, as well as many
thousands of individuals who travel to these areas. In addition, while visibility trends have
improved in mandatory class I federal areas the most recent data show that these areas
continue to suffer from visibility impairment. In eastern parks, average visual range has
decreased from 90 miles to 15-25 miles. In the West, visual range has decreased from 140
miles to 35-90 miles. In summary, visibility impairment is experienced throughout the U.S.,
in multi-state regions, urban areas, and remote mandatory class I federal areas.353'354 The
mandatory federal class I areas are listed in Figure 3.3-23 and in Table 3.3-7.
3.3.3.2.2 Projected Visibility Impairment in U.S. - Impact of Ship Emissions
Based on modeling for the EGA proposal, international shipping activities in 2002
contributed to visibility degradation at all of the 133 class I federal areas which have complete
Interagency Monitoring of Protected Visual Environments (IMPROVE) ambient data for
2002 or are represented by IMPROVE monitors with complete data.cc. Absent further
emission controls, by 2020, international shipping activities will have an even larger impact
on visibility deciview levels00 in these class I federal areas. The results suggest that
controlling emissions from C3 vessels would result in improved visibility deciview levels in
all 133 monitored class I federal areas- although areas would continue to have annual
average deciview levels above background in 2020.
The results indicate that reductions in regional haze would occur in all 133 of the areas
analyzed as a result of an EGA adoption. The model projects that for all monitored
mandatory class I federal areas combined, average visibility on the 20% worst days at these
scenic locales would improve by 0.21 deciviews, or 1.2%. The greatest improvements in
visibility are in coastal areas. For instance, the Agua Tibia Wilderness area (near Los
Angeles) would see 9.4% improvement as a result of the proposed EGA. National parks and
national wilderness areas in other parts of the country would also see improvements as a result
of EGA controls. For example, the Cape Romain National Wildlife Refuge (South Carolina)
would see a 4.6% improvement in visibility; and Acadia National Park (Maine) would see a
4.4% improvement with the proposed EGA. Likewise, in 2002, 2.7% of visibility impairment
in southern Florida's Everglades National Park was due to international shipping, and this will
double to 6% by 2020. Even in inland class I federal areas international shipping activity is
contributing to visibility degradation. In 2020, about 2.5% of visibility degradation in the
Grand Canyon National Park located in the state of Arizona will be from international
shipping, while almost 6% of visibility degradation in the State of Washington's North
cc There are 156 federally-mandated class I areas which, under the Regional Haze Rule, are required to achieve
natural background visibility levels by 2064. These mandatory class I federal areas are mostly national parks,
national monuments, and wilderness areas. There are currently 116 IMPROVE monitoring sites (representing all
156 mandatory class I federal areas) collecting ambient PM2.5 data at mandatory class I federal areas, but not all
of these sites have complete data for 2002.
DD The level of visibility impairment in an area is based on the light-extinction coefficient and a unit less
visibility index, called a "deciview", which is used in the valuation of visibility. The deciview metric provides a
scale for perceived visual changes over the entire range of conditions, from clear to hazy. Under many scenic
conditions, the average person can generally perceive a change of one deciview. The higher the deciview value,
the worse the visibility. Thus, an improvement in visibility is a decrease in deciview value.
3-89
-------
Cascades National Park will be from international shipping emissions. Table 3.3-7 which
follows contains the full visibility results from the 2020 EGA scenario over the 133 analyzed
areas.
3.3.3.3 Visibility Modeling
Many scenic areas in the U.S. have reduced visibility because of regional haze. The
U.S. EPA is in the midst of a major effort to improve air quality in national parks and
wilderness areas, especially for those meteorological situations in which visibility is most
degraded. The CMAQ modeling discussed in Section 3.2 was also used to project the impacts
of potential EGA-based emissions reductions on visibility conditions over specific national
parks and wilderness areas across the U.S. over the 20% worst visibility days at that location.
Table 3.3-7 Visibility Levels in Deciviews for Individual U.S. Class 1 Areas on the 20% Worst Days for
Several Scenarios
CLASS 1 AREA
(20% WORST DAYS)
Sipsey Wilderness
Caney Creek Wilderness
Upper Buffalo
Wilderness
Chiricahua NM
Chiricahua Wilderness
Galiuro Wilderness
Grand Canyon NP
Mazatzal Wilderness
Petrified Forest NP
Pine Mountain
Wilderness
Saguaro NM
Sierra Ancha Wilderness
Sycamore Canyon
Wilderness
Agua Tibia Wilderness
Caribou Wilderness
Cucamonga Wilderness
Desolation Wilderness
Dome Land Wilderness
Emigrant Wilderness
Hoover Wilderness
Joshua Tree NM
Lassen Volcanic NP
Lava Beds NM
Mokelumne Wilderness
Pinnacles NM
Point Reyes NS
Redwood NP
San Gabriel Wilderness
San Gorgonio
Wilderness
San Jacinto Wilderness
STATE
AL
AR
AR
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
BASELINE
VISIBILITY
29.03
26.36
26.27
13.43
13.43
13.43
11.66
13.35
13.21
13.35
14.83
13.67
15.25
23.50
14.15
19.94
12.63
19.43
17.63
12.87
19.62
14.15
15.05
12.63
18.46
22.81
18.45
19.94
22.17
22.17
2020
BASE
23.67
22.20
22.25
13.15
13.17
13.18
11.24
12.88
12.88
12.74
14.39
13.33
15.00
22.99
13.73
18.34
12.29
18.59
17.35
12.79
17.95
13.71
14.47
12.40
17.86
22.38
18.26
17.92
20.66
20.25
ECA
23.42
22.01
22.15
13.07
13.09
13.09
11.04
12.73
12.76
12.59
14.31
13.21
14.90
20.82
13.51
17.57
12.11
18.23
17.14
12.68
17.30
13.46
14.32
12.21
17.11
21.71
17.81
17.12
20.45
19.86
ZERO C3
EMISSIONS
23.32
21.88
22.11
13.00
13.02
13.00
10.96
12.61
12.70
12.48
14.22
13.10
14.84
20.11
13.43
17.27
12.07
18.14
17.08
12.65
17.21
13.37
14.24
12.16
16.89
21.54
17.48
16.84
20.35
19.55
NATURAL
BACKGROUND
10.99
11.58
11.57
7.21
7.21
7.21
7.14
6.68
6.49
6.68
6.46
6.59
6.69
7.64
7.31
7.06
6.12
7.46
7.64
7.91
7.19
7.31
7.86
6.12
7.99
15.77
13.91
7.06
7.30
7.30
3-90
-------
CLASS 1 AREA
(20% WORST DAYS)
South Warner
Wilderness
Thousand Lakes
Wilderness
Ventana Wilderness
Yosemite NP
Black Canyon of the
Gunnison NM
Eagles Nest Wilderness
Flat Tops Wilderness
Great Sand Dunes NM
La Garita Wilderness
Maroon Bells-Snowmass
Wilderness
Mesa Verde NP
Mount Zirkel Wilderness
Rawah Wilderness
Rocky Mountain NP
Weminuche Wilderness
West Elk Wilderness
Chassahowitzka
Everglades NP
St. Marks
Cohutta Wilderness
Okefenokee
Wolf Island
Craters of the Moon NM
Sawtooth Wilderness
Mammoth Cave NP
Acadia NP
Moosehorn
Roosevelt Campobello
International Park
Isle Royale NP
Seney
Voyageurs NP
Hercules-Glades
Wilderness
Anaconda-Pintler
Wilderness
Bob Marshall
Wilderness
Cabinet Mountains
Wilderness
Gates of the Mountains
Wilderness
Medicine Lake
Mission Mountains
Wilderness
Scapegoat Wilderness
STATE
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
FL
FL
FL
GA
GA
GA
ID
ID
KY
ME
ME
ME
MI
MI
MN
MO
MT
MT
MT
MT
MT
MT
MT
BASELINE
VISIBILITY
15.05
14.15
18.46
17.63
10.33
9.61
9.61
12.78
10.33
9.61
13.03
10.52
10.52
13.83
10.33
9.61
26.09
22.30
26.03
30.30
27.13
27.13
14.00
13.78
31.37
22.89
21.72
21.72
20.74
24.16
19.27
26.75
13.41
14.48
14.09
11.29
17.72
14.48
14.48
2020
BASE
14.70
13.68
18.36
17.32
9.77
9.05
9.25
12.41
9.91
9.23
12.42
10.02
10.00
13.09
9.88
9.20
22.37
21.75
22.37
23.29
23.86
23.76
13.00
13.66
25.43
20.55
19.02
19.25
18.99
21.54
17.55
22.84
13.14
14.13
13.55
10.90
16.20
14.02
14.15
ECA
14.57
13.42
17.72
17.13
9.69
9.00
9.20
12.36
9.84
9.19
12.33
9.99
9.97
13.06
9.80
9.15
21.97
21.14
21.96
23.13
23.30
22.97
12.97
13.63
25.33
19.79
18.55
18.58
18.84
21.49
17.52
22.74
13.10
14.11
13.50
10.87
16.18
13.99
14.12
ZERO C3
EMISSIONS
14.51
13.33
17.57
17.08
9.66
8.98
9.18
12.34
9.81
9.16
12.28
9.98
9.95
13.05
9.77
9.12
21.75
20.40
21.65
23.07
23.07
22.75
12.94
13.61
25.30
19.62
18.38
18.23
18.81
21.47
17.51
22.72
13.07
14.09
13.47
10.85
16.17
13.97
14.11
NATURAL
BACKGROUND
7.86
7.31
7.99
7.64
6.24
6.54
6.54
6.66
6.24
6.54
6.83
6.44
6.44
7.24
6.24
6.54
11.21
12.15
11.53
11.14
11.44
11.44
7.53
6.43
11.08
12.43
12.01
12.01
12.37
12.65
12.06
11.30
7.43
7.74
7.53
6.45
7.90
7.74
7.74
3-91
-------
CLASS 1 AREA
(20% WORST DAYS)
Selway-Bitterroot
Wilderness
UL Bend
Linville Gorge
Wilderness
Swanquarter
Lostwood
Theodore Roosevelt NP
Great Gulf Wilderness
Presidential Range-Dry
River Wilderness
Brigantine
Bandelier NM
Bosque del Apache
Gila Wilderness
Pecos Wilderness
Salt Creek
San Pedro Parks
Wilderness
Wheeler Peak
Wilderness
White Mountain
Wilderness
Jarbidge Wilderness
Wichita Mountains
Crater Lake NP
Diamond Peak
Wilderness
Eagle Cap Wilderness
Gearhart Mountain
Wilderness
Hells Canyon
Wilderness
Kalmiopsis Wilderness
Mount Hood Wilderness
Mount Jefferson
Wilderness
Mount Washington
Wilderness
Mountain Lakes
Wilderness
Strawberry Mountain
Wilderness
Three Sisters Wilderness
Cape Romain
Badlands NP
Wind Cave NP
Great Smoky Mountains
NP
Joyce-Kilmer-Slickrock
Wilderness
STATE
MT
MT
NC
NC
ND
ND
NH
NH
NJ
NM
NM
NM
NM
NM
NM
NM
NM
NV
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
SC
SD
SD
TN
TN
BASELINE
VISIBILITY
13.41
15.14
28.77
25.49
19.57
17.74
22.82
22.82
29.01
12.22
13.80
13.11
10.41
18.03
10.17
10.41
13.70
12.07
23.81
13.74
13.74
18.57
13.74
18.55
15.51
14.86
15.33
15.33
13.74
18.57
15.33
26.48
17.14
15.84
30.28
30.28
2020
BASE
13.08
14.65
22.63
21.79
17.45
16.44
19.53
19.53
25.27
11.45
12.93
12.59
10.00
16.70
9.52
9.91
12.87
11.88
20.45
13.33
13.26
17.73
13.41
17.16
15.24
14.30
14.90
14.88
13.28
17.71
14.93
23.51
15.63
14.78
24.01
23.56
ECA
13.02
14.63
22.43
21.11
17.43
16.42
19.34
19.33
24.46
11.39
12.89
12.52
9.93
16.66
9.44
9.85
12.82
11.81
20.31
13.20
13.11
17.69
13.30
17.12
14.85
13.93
14.62
14.62
13.14
17.66
14.69
22.35
15.59
14.75
23.81
23.35
ZERO C3
EMISSIONS
12.98
14.62
22.34
20.99
17.41
16.41
19.29
19.28
24.31
11.36
12.87
12.48
9.90
16.63
9.41
9.82
12.79
11.78
20.24
13.13
13.03
17.65
13.25
17.07
14.66
13.64
14.46
14.46
13.07
17.62
14.54
22.14
15.57
14.73
23.72
23.26
NATURAL
BACKGROUND
7.43
8.16
11.22
11.94
8.00
7.79
11.99
11.99
12.24
6.26
6.73
6.69
6.44
6.81
6.08
6.44
6.86
7.87
7.53
7.84
7.84
8.92
7.84
8.32
9.44
8.44
8.79
8.79
7.84
8.92
8.79
12.12
8.06
7.71
11.24
11.24
3-92
-------
CLASS 1 AREA
(20% WORST DAYS)
Big Bend NP
Carlsbad Caverns NP
Guadalupe Mountains
NP
Arches NP
Bryce Canyon NP
Canyonlands NP
Zion NP
James River Face
Wilderness
Shenandoah NP
Lye Brook Wilderness
Alpine Lake Wilderness
Glacier Peak Wilderness
Goat Rocks Wilderness
Mount Adams
Wilderness
Mount Rainier NP
North Cascades NP
Olympic NP
Pasayten Wilderness
Dolly Sods Wilderness
Otter Creek Wilderness
Bridger Wilderness
Fitzpatrick Wilderness
Grand Teton NP
North Absaroka
Wilderness
Red Rock Lakes
Teton Wilderness
Washakie Wilderness
Yellowstone NP
STATE
TX
TX
TX
UT
UT
UT
UT
VA
VA
VT
WA
WA
WA
WA
WA
WA
WA
WA
WV
WV
WY
WY
WY
WY
WY
WY
WY
WY
BASELINE
VISIBILITY
17.30
17.19
17.19
11.24
11.65
11.24
13.24
29.12
29.31
24.45
17.84
13.96
12.76
12.76
18.24
13.96
16.74
15.23
29.04
29.04
11.12
11.12
11.76
11.45
11.76
11.76
11.45
11.76
2020
BASE
16.25
16.05
16.03
10.94
11.41
10.96
12.91
23.31
22.77
21.02
16.85
13.85
12.23
12.16
17.47
13.85
16.18
14.89
22.46
22.45
10.83
10.87
11.37
11.17
11.45
11.43
11.19
11.40
ECA
16.11
15.98
15.95
10.86
11.28
10.90
12.80
23.16
22.61
20.77
16.56
13.53
11.95
11.88
17.02
13.46
15.87
14.82
22.31
22.30
10.78
10.81
11.32
11.14
11.40
11.38
11.16
11.35
ZERO C3
EMISSIONS
16.01
15.93
15.90
10.83
11.22
10.89
12.73
23.12
22.57
20.72
16.26
13.19
11.70
11.67
16.66
13.04
15.39
14.72
22.26
22.26
10.76
10.79
11.30
11.13
11.38
11.36
11.15
11.33
NATURAL
BACKGROUND
7.16
6.68
6.68
6.43
6.86
6.43
6.99
11.13
11.35
11.73
8.43
8.01
8.36
8.36
8.55
8.01
8.44
8.26
10.39
10.39
6.58
6.58
6.51
6.86
6.51
6.51
6.86
6.51
3-93
-------
Appendices
Appendix 3A
Once air pollutants have been emitted into the atmosphere, the processes that
determine pollutant concentrations in space and time are largely determined by meteorology.
This portion of the document describes the relevant meteorological conditions within the
proposed areas that contribute to at-sea emissions being transported to populated areas and
contributing to harmful human health and ecological impacts.
As noted elsewhere in this document, NOX, SOX, and direct participate matter are
emitted from ships. These pollutants and the pollutants that are secondarily formed from
these emissions can have atmospheric lifetimes of 5-10 days before being significantly
dispersed, deposited, or converted to other species (Clarke et al., 2001; Karamchandani et al.,
2006). As a result of these rather long residence times in the atmosphere, it is important to
consider similar meteorological scales when determining the potential impacts of ship
emissions on human health and ecosystems. Thus, while meteorological phenomena of all
sizes affect the eventual impacts of ship emissions, the longer range regional transport of
pollutants from shipping is largely dictated by synoptic scale meteorological patterns.
Prevailing wind patterns can vary by season and by location over the United States,
but it is common for air masses to have a maritime influence especially looking back at time
periods of 5-10 days. Over parts of the U.S., this is readily evident from regional reanalyses
of ambient meteorological conditions. Figures 3A-1 and 3A-2 show prevailing winds over
the course of last year (2008) based on the NCEP Regional Reanalysis dataset (Mesinger,
2006) which is derived from the Eta weather forecast model as guided by assimilation of large
volumes of measured meteorological data. The maps show the monthly mean wind barbs.
These wind barbs are comprised of two straight lines, the longest of which indicates the
monthly mean wind direction. The shorter line indicates the speed of the monthly mean wind
vector. The wind blows from the intersection of the two lines to the end of the longer line.
Caution should be exercised when viewing these figures, as there are certainly individual
hours and days in which the winds deviate from the monthly means. Additionally, while 2008
was generally a representative yearEE, other years strongly influenced by extreme phases of
ocean-atmospheric oscillations, such as the El Nino Southern Oscillation (ENSO) could have
different patterns.
The prevailing winds in the winter period result in westerly transport of air masses
across the U.S. On average, this results in on-shore flow over the western States, along the
Texas Gulf Coast and the east coast of Florida. The polar jet stream is a prominent feature
over the U.S. in the winter and as a result, the wind fields tend to be most dynamic in this
EE 2008 featured a waning La Nina phase of the ENSO as determined by the NOAA Climate Prediction Center.
(http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ensoyears.shtml). Mean temperatures
and precipitation patterns in 2008 were generally near long-term averages, with the exception of the Upper
Midwest which was cooler and wetter than normal as determined by the NOAA National Climatic Data Center.
(http://www.ncdc.noaa.gov/oa/climate/research/2008/cmb-prod-us-2008.html)
3-94
-------
period. The wind fields around strong low pressure cyclones can advect air masses large
distances (i.e., across the continent) in relatively short periods (i.e., less than a week).
Wind Barbs [Knots] - 1000 mb
-Jan 2008
Figure 3A-1: Monthly Mean Winds in January 2008 Based on the NCEP Regional Reanalysis Dataset
Wind Barbs [Knots] - 1000 mb
-Apr 2008
Figure 3A-2: Monthly Mean Winds in April 2008 Based on the NCEP Regional Reanalysis Dataset.
By the spring period, the mean wind flow still tends to be onshore over the Pacific
Northwest, but it takes on a more parallel-to-the-coast alignment across California as a strong
eastern Pacific anticyclone begins to set up. Along the Gulf Coast, southerly winds are
common during this period. Strong low-level jet streams frequently originate over the
3-95
-------
Midwestern U.S. during the spring resulting in the rapid northward advection of moist tropical
air from the Gulf of Mexico to parts of the U.S. otherwise well removed from maritime
influences. The mean wind fields are weak along the Atlantic Coast indicating near equal
onshore/offshore winds. Although along the highly populated portions of the East Coast
(New York, Philadelphia, Baltimore, Washington DC) there was a net tendency for transport
off the ocean.
The eastern Pacific ridge is strong in the summertime and the prevailing winds tend to
run along the West Coast. In immediate coastal environs it is common for diurnally-based
wind patterns such as sea, land, bay, and lake breezes to govern how much onshore/offshore
exchange takes place. The polar jet stream is typically located well north of the U.S./Canada
border during the summer. Conditions tend to be more stagnant in this period than other
times of the year. However, mean southerly winds over the Central U.S. expose large parts of
the country to impacts from pollutants emitted or formed in the Gulf of Mexico. Mean winds
around the Bermuda High that typically governs flows in the western Atlantic, generally
results in offshore winds over the Eastern U.S. except in far north-eastern States like
Connecticut, Massachusetts, and Maine where on average there is a considerable onshore
wind component.
The fall season is a transition back to winter. Onshore winds begin to be more
commonplace in Washington and Oregon. Subtropical trade winds result in low-level
steering of air masses (and the occasional hurricane) into the Southeastern U.S. The
predominant winds over the Northeastern U.S. are offshore as cold frontal passages from
Canada become more frequent as the polar jet is displaced southward.
As noted earlier, there can be daily deviations within the prevailing seasonal winds.
One tool that can be used to determine the origination of an air mass for a pollution event are
Lagrangian trajectory models like HYSPLIT (Draxler and Hess, 1997) which calculates the
path a plume of emissions would take given an input meteorological field. A set of three
sample HYSPLIT 48-hour back trajectories are shown in Figure 3A-5 for a chosen day in the
summer of 2008 with elevated levels of PM2.s over parts of the U.S. These figures are
intended to provide a visual for what the HYSPLIT output products look like, more than to
imply any causality between these particular trajectories and the resultant air quality on this
day. The CMAQ air quality modeling, discussed above in Chapter 3.2.5, was used to isolate
and estimate the impacts of shipping emissions on locations on land. These particular sample
back-trajectories show a relatively stagnant atmosphere over Los Angeles with potential
interactions with emissions from shipping sources just offshore. The back-trajectories over
Birmingham and Philadelphia indicate that there is no direct maritime influence over the past
two days for those locations. Of course, it is still possible that the longer-trajectories might
indicate some small contribution to the overall background from sources over the water.
3-96
-------
Wind Barbs [Knots] - 1000 mb
-Jul 2008
Figure 3A-3: Monthly Mean Winds in July 2008 Based on the NCEP Regional Reanalysis Dataset
Wind Barbs [Knots] - 1000 mb
-bet 2008
\ >
Figure 3A-4: Monthly Mean Winds in October 2008 Based on the NCEP Regional Reanalysis Dataset.
3-97
-------
-130
Figure 3A-5: 48-Hour Back-Trajectories from the HYSPLIT Trajectory Model. The red triangles
represent how the air parcel that resided over the starred locations on 0000 GMT July 19,2008 travelled
over the preceding two days, in three hour increments.
Figure 3A-5 shows the compilation of daily (1800 GMT) 24-hour back trajectories
over Los Angeles as derived from 12 years (1995-2006) of meteorological data provided by
the Eta Data Assimilations System. For this location, if the mean transport direction (as
determined from the starting point to the ending point of the trajectory) was from 150 to 300
degrees, then that day was flagged as potentially having a maritime influence. This analysis
was completed for several major U.S. population centers near a coast. The results are shown
in Table 3A-1. As can be seen, while the frequency of maritime influences can vary by
location, it is not uncommon for locations all across the United States to be potentially
affected by emissions that originate offshore.
3-98
-------
120-150 150-180 180-210 210-240 240-270
Direction of back trajectory (deg)
300-330 330-300
Figure 3A-6: 24-Hour Back Trajectory Directions in Los Angeles as Estimated by the HYSPLIT Model
over the Period from 1995 to 2006
Table 3A-1: Summary of HYSPLIT back trajectories at highly-populated urban USA areas over a 12-
year period showing the frequency at which the air mass likely emanated from a marine environment.
HIGHLY POPULATED
USA COASTAL CITY
San Francisco
Los Angeles
San Diego
Houston
New Orleans
Miami
New York City
Boston
TRAJECTORY DIRECTIONS
CONSIDERED TO BE INDICATIVE
OF MARINE AIR (DEG)
180-330
150-300
180-330
90-210
90-240
30-180
30-180
30-120
FREQUENCY OF MARINE
INTRUSION OVER THE
PERIOD 1995-2006 (%)
45.7
46.3
67.2
58.9
48.7
65.8
19.0
12.5
In addition to the prevailing winds, the atmospheric stability can also conspire to result
in land-based impacts from ship plumes. At certain locations and times of the year, the
marine environment is characterized by a shallow temperature inversion (250-500m AGL)
caused by the interaction between warmer subsiding air over cooler water (Winant et al.,
1988). When ship emissions are injected into this shallow boundary layer, especially
concentrated plumes can be maintained for long distances. This effect can be occasionally be
seen in satellite pictures when clouds are formed by the exhaust from ships. When a
persistent marine inversion exists, these clouds (and by extension the pollutant plumes from
the ships) can be maintained for hundreds of kilometers and several days as shown below.
3-99
-------
Figure 3A-7. MODIS Satellite Picture from May 11,2005 Showing Clouds Formed from Ship Tracks.
This public domain photo is from NASA's Earth Observatory at the website:
http://earthobservatory.nasa.gov/IOTD/view.php?id=5488.
The MM5 meteorological modeling (Grell, et al., 1994) that was used to drive the air
quality modeling simulations performed for this analysis captured this effect over the Eastern
Pacific Ocean, the Northwest Atlantic Ocean, and the Great Lakes. Monthly average mixing
heights over these regions were typically less than 300 m in the summer. This marine
inversion prevents the ship plumes from being diluted vertically until they reach the coastal
environs adjacent to the cool waters.
The last key meteorological element that is particularly relevant to any consideration
of shipping emissions on human health and ecosystems is acid deposition. Deposition
processes can occur in two modes: dry and wet. Wet deposition occurs when gases or
particles are 'washed' out of the air by rain, snow, fog, or some other form of precipitation.
The amount of precipitation over the water bodies surrounding North America can vary by
location and season depending upon the synoptic meteorological patterns. However,
orographical influences along the Pacific Northwest, and to a lesser extent over interior
regions (e.g., Rocky Mountains, Appalachian Mountains) can lead to enhanced precipitation
in those regions when the winds are from the ocean. Figure 3A-8 shows the monthly
precipitation patterns over the U.S. for January 2008. When moist westerly winds are lifted
up over the Cascade mountain range from Northern California through Washington State,
large amounts of precipitation can occur on the windward side of the mountains.
Additionally, in the summertime it is common for precipitation to be enhanced in coastal
areas due to sea-breeze thunderstorms as well as general proximity to the moisture source.
3-100
-------
04
Figure 3A-8. Monthly Precipitation Accumulations in January 2008 from the NCEP Regional Reanalysis
Dataset. Units are kg/m2.
The air quality modeling analyses and the meteorological discussion above focused on
the 48-state contiguous portion of the United States, but the same meteorological conditions
that result in potential impacts of ship emissions on air pollution over land in that region (e.g.,
prevailing winds, atmospheric stability, and precipitation patterns) can also result in potential
impacts over Alaska and Hawaii. In fact, the oceanic influence is likely greater over the
Hawaiian Islands and the coastal environs of Alaska (typically more populated than the
interior portions of that State).
Because of its great expanse, the climatology of Alaska can differ widely depending
upon latitude, altitude, and proximity to the ocean. Generally, the state's meteorology is
classified in three zones: maritime, continental, and arctic. The weather in the maritime
locations are strongly influenced by the relatively steady-state Pacific Ocean and as a results
there are relatively small variations in prevailing winds, humidity levels and temperatures by
season and location (Alaska Climate Research Center, 2009). Without the stabilizing
influence of the ocean waters, the continental and arctic regions can experience large seasonal
extremes in temperature, humidity, precipitation, and wind direction. The local meteorology
in these two zones is driven by the topography of the surrounding areas, the altitude, and the
fraction of sea ice in the Arctic Ocean.
The proximity of the maritime regions to the shipping lanes lead to the conclusion that
populations in these areas would be most likely to be adversely impacted by air pollution
originating from ships. While wind directions at measuring sites in Alaska can be strongly
influenced by topography, the winds typically have an easterly component in populated
locations like Anchorage, Juneau, Sitka, and Kenai (Western Regional Climate Center, 2009).
Figure 3A-9 shows the average prevailing wind direction at 850 mb (approximately 1500 m
above ground level) for the months of January and July, averaged over a recent 17 year
period. The steering winds at this level indicate the potential for the transport of shipping
3-101
-------
emissions in the North Pacific (shipping routes from Asia to North America). These winds
are driven by common synoptic features that govern weather in this region, specifically the
Aleutian low pressure cyclone in the winter and a northeastern Pacific anticyclone in the
summer.
Monthly Mean 350-HPa Wind (m/s): January
Climatology: 1979-1995
Monthly Wean 850-hPa Wind (m/s): July
Climatology: 1379-1995
10 12 14 IB
10 12 14
Figure 3A-9. Monthly Mean Winds at Approximately the 1,500 Meter Level in January (left) and July
(right) Averaged over the Period from 1979 to 1995. Figures from NOAA Climate Prediction Center
Not surprisingly, Hawaiian meteorology is also subject to strong maritime influences.
Kodama and Businger (1998) summarized the basic meteorology that occurs over this region.
Global circulations such as the Hadley cell establish east-northeasterly trade winds as the
predominant flow pattern in Hawaii, especially in the warm season. These trade winds can
comprise 50-90 percent of the hourly wind directions over the region. Typically, the average
height of the surface layer ranges from 1500-3000 m AGL in all seasons in Hawaii. Any
emissions input to this layer will remain in this layer unless ventilated by convection or
removed by deposition. Ultimately, as there are shipping lanes on all sides of the main
Hawaiian Islands; regardless of which way the wind blows, there is a high potential for ship
emissions to affect air pollution over land.
In conclusion, there is ample evidence that the meteorological conditions in the
proposed area of application have the potential to put human populations and environmental
areas at risk of adverse environmental impacts from ship emissions. This conclusion is
confirmed by the air quality modeling analyses performed for this assessment.
3-102
-------
APPENDIX 3B
Table 3B-1. Percent reduction in Nitrogen (N) and Sulfur (S) deposition averaged over a 2-digit HUC sub
region for two modeling scenarios. The range of reductions for individual HUCs within the sub region is
shown in parentheses.
HUC SUB REGION
New England (1)
Mid Atlantic (2)
South Atlantic - Gulf
(3)
Great Lakes (4)
Ohio (5)
Tennessee (6)
Upper Mississippi (7)
Lower Mississippi (8)
Souris-Red-Rainy (9)
Missouri (10)
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
ZERO C3
EMISSIONS
4.9%
(2.6 to 11.0%)
6.3%
(3.0 to 16.3%)
3.1%
(1.1 to 7.4%)
6.6%
(1.2 to 14%)
5.9%
(1.8 to 11.4%)
8.7%
(3.1 to 10.3%)
0.9%
(0.4 to 1.7%)
1.2%
(0.6 to 2.9%)
1.5%
(0.6 to 2.5%)
1.4%
(0.8 to 3.3%)
2.5%
(0.6 to 3.8%)
2.8%
(0.8 to 5.0%)
0.5%
(0.2 to 1.4%)
1.1%
(0.4 to 2.2%)
5.1%
(2.6 to 11.5%)
7.8%
(4.5 to 15.6%)
0.3%
(0.2 to 17.2%)
0.9%
(0.3 to 33.3%)
0.6%
(0.4 to 1.8%)
1.8%
(1.3 to 3. 7%)
ECA
1.3%
(0.7 to 3. 5%)
5.3%
(1.8 to 15.0%)
0.8%
(0.1 to 1.9%)
6.0%
(1.0 to 13.0%)
1.1%
(0.3 to 2.8%)
6.1%
(2.0 to 7.1%)
0.2%
(0.1 to 0.5%)
1.0%
(0.5 to 2.7%)
0.4%
(0.1 to 0.7%)
1.0%
(0.6 to 2. 2%)
0.6%
(0.1 to 1.0%)
1.9%
(0.6 to 3. 5%)
0.1%
(0.1 to 0.4%)
0.7%
(0.3 to 1.3%)
1.2%
(0.5 to 2.8%)
5.8%
(3. 2 to 11.3%)
0.1%
(0.1 to 4.8%)
0.6%
(0.2 to 28.5%)
0.2%
(0.1 to 0.5%)
1.1%
(0.7 to 2. 2%)
3-103
-------
HUC SUB REGION
Arkansas- White-Red
(11)
Texas-Gulf (12)
Rio Grande (13)
Upper Colorado (14)
Lower Colorado (15)
Great Basin (16)
Pacific Northwest (17)
California (18)
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
average reduction
(range) in N deposition
average reduction
(range) in S deposition
ZERO C3
EMISSIONS
1.5%
(0.6 to 6.8%)
3.6%
(1.6 to 7.6%)
3.3%
(1.7 to 7.7%)
7.0%
(2. 3 to 11.7%)
2.0%
(0.7 to 2.9%)
3.2%
(1.5 to 4.4%)
1.6%
(1.2 to 3.1%)
2.8%
(1.0 to 7.1%)
3.3%
(1.7 to 5. 5%)
5.2%
(3. 2 to 10.1%)
2.0%
(1.2 to 3.0%)
4.4%
(2.1 to 7.1%)
4.9%
(2.2 to 33.5%)
14.5%
(5.1 to 56.4%)
8.4%
(2.5 to 40.4%)
21.3%
(4.6 to 81. 6%)
ECA
0.3%
(0.1 to 1.7%)
2.2%
(0.8 to 5.4%)
0.5%
(0.0 to 1.4%)
4.9%
(1.3 to 8.4%)
0.4%
(0.2 to 0.5%)
1.7%
(0.8 to 2.4%)
0.6%
(0.5 to 1.2%)
2.2%
(0.8 to 5.6%)
0.9%
(0.4 to 1.5%)
3.3%
(1.6 to 7.4%)
0.8%
(0.5 to 1.5%)
3.7%
(1.7 to 6.1%)
1.0%
(0.1 to 6.1%)
11.1%
(4 to 37.5%)
2.3%
(0.7 to 13.4%)
19.4%
(3.8 to 78.1%)
1 U.S. EPA. (2005). Review of the National Ambient Air Quality Standard for Paniculate Matter: Policy
Assessment of Scientific and Technical Information, OAQPS Staff Paper. EPA-452/R-05-005a. Retrieved
March 19, 2009 from http://www.epa.gov/ttn/naaqs/standards/pm/data/pmstaffpaper 20051221.pdf. Section 2.2.
2 U.S. EPA. (2006). Air Quality Criteria for Ozone and RelatedPhotochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-
0190 at http://www.regulations.gov/.
3 U.S. EPA. (2006). National-Scale Air Toxics Assessment for 1999. This material is available electronically at
http://www.epa.gov/ttn/atw/natal999/.
4 Agrawal, H., Malloy, Q.G.J., Welch, W.A., Miller, J.W., Cocker, D.R. (2008). In-use gaseous and paniculate
matter emissions from a modern ocean going container vessel. Atmospheric Environment, 42, 5504-5510.
3-104
-------
5 Hu, S., Polidori, A., Arhami, M., Shafer, M.M., Schauer, J.J., Cho, A., Sioutas, C. (2008). Redox activity and
chemical speciation of size fractionated PM in the communities of the Los Angeles-Long Beach Harbor.
Atmospheric Chemistry and Physics Discussions, 8, II683-11672.
6 U.S. EPA (2002). Health Assessment Document for Diesel Engine Exhaust. EPA/600/8-90/057F Office of
Research and Development, Washington DC. Retrieved on March 17, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=29060. pp. 1-1 1-2.
7 U.S. EPA (2004). Air Quality Criteria for Paniculate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/.
8 U.S. EPA. (2005). Review of the National Ambient Air Quality Standard for Particulate Matter: Policy
Assessment of Scientific and Technical Information, OAQPS Staff Paper. EPA-452/R-05-005a. Retrieved
March 19, 2009 from http://www.epa.gov/ttn/naaqs/standards/pm/data/pmstaffpaper_20051221.pdf.
9 U.S. EPA. (2006). Provisional Assessment of Recent Studies on Health Effects of Particulate Matter Exposure.
EPA/600/R-06/063. Retrieved on March 19, 2009 from
http://www.epa.gov/air/particlepollution/pdfs/ord_report_20060720.pdf.
10 U.S. EPA (2004). Air Quality Criteria for Particulate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/. p. 8-305
11 U.S. EPA (2004). Air Quality Criteria for Particulate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/. p. 9-93.
12 U.S. EPA (2004). Air Quality Criteria for Particulate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/. Section 8.3.3.1.
13 U.S. EPA (2004). Air Quality Criteria for Particulate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/. Table 8-34.
14 U.S. EPA (2004). Air Quality Criteria for Particulate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/. Section 8.3.1.3.4.
15 U.S. EPA (2004). Air Quality Criteria for Particulate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/. Section 8.3.4.
16 U.S. EPA (2004). Air Quality Criteria for Particulate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/. p. 8-85.
17 Laden, F., Neas, L.M., Dockery D.W., et al. (2000). Association of fine particulate matter from different
sources with daily mortality in six U.S. cities. Environ Health Perspectives, 108(10), 941-947.
3-105
-------
18 Schwartz, J., Laden, F. Zanobetti, A. (2002). The concentration-response relation between PM(2.5) and daily
deaths. Environ Health Perspect, 110(10), 1025-1029.
19 Mar, T.F., Ito, K., Koenig, J.Q., Larson, T.V., Eatough, D.J., Henry, R.C., Kim, E., Laden, F., Lall, R., Neas,
L., Stolzel, M., Paatero, P., Hopke, P.K., Thurston, G.D. (2006). PM source apportionment and health effects. 3.
Investigation of inter-method variations in associations between estimated source contributions of PM2.5 and
daily mortality in Phoenix, AZ. J. Exposure Anal. Environ. Epidemiol, 16, 311-320.
20 Ito, K., Christensen, W.F., Eatough, D.J., Henry, R.C., Kim, E., Laden, F., Lall, R., Larson, T.V., Neas, L.,
Hopke, P.K., Thurston, G.D. (2006). PM source apportionment and health effects: 2. An investigation of
intermethod variability in associations between source-apportioned fine particle mass and daily mortality in
Washington, DC. J. Exposure Anal. Environ. Epidemiol., 16, 300-310.
21 Janssen N.A., Schwartz]., Zanobetti A., et al. (2002). Air conditioning and source-specific particles as
modifiers of the effect of PM10 on hospital admissions for heart and lung disease. Environ Health Perspect,
110(1), 43-49.
22 U.S. EPA (2004). Air Quality Criteria for Paniculate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/. p. 8-307.
23 U.S. EPA (2004). Air Quality Criteria for Paniculate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/. p. 8-313, 8-314.
24 U.S. EPA (2004). Air Quality Criteria for Paniculate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/. p.8-318.
25 U.S. EPA (2004). Air Quality Criteria for Paniculate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/. p. 8-306.
26 U.S. EPA. (2005). Review of the National Ambient Air Quality Standard for Paniculate Matter: Policy
Assessment of Scientific and Technical Information, OAQPS Staff Paper. EPA-452/R-05-005a. Retrieved
March 19, 2009 from http://www.epa.gov/ttn/naaqs/standards/pm/data/pmstaffpaper_20051221.pdf. p.3-18.
27 Dockery, D.W., Pope, C.A. Ill, Xu, X, et al. (1993). An association between air pollution and mortality in six
U.S. cities. NEnglJMed, 3^,1753-1759. Retrieved on March 19, 2009 from
http://content.nejm.org/cgi/content/full/329/24/1753.
28 Pope, C.A., III, Thun, M.J., Namboodiri, M.M., Dockery, D.W., Evans, J.S., Speizer, F.E., and Heath, C.W.,
Jr. (1995). Particulate air pollution as a predictor of mortality in a prospective study of U.S. adults. Am. J. Respir.
Crit CareMed, 151, 669-674.
29 Pope, C. A., Ill, Burnett, R.T., Thun, M. J., Calle, E.E., Krewski, D., Ito, K., Thurston, G.D., (2002). Lung
cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. J. Am. Med. Assoc,
257,1132-1141.
30 Krewski, D., Burnett, R.T., Goldberg, M.S., et al. (2000). Reanalysis of the Harvard Six Cities study and the
American Cancer Society study of particulate air pollution and mortality. A special report of the Institute's
3-106
-------
Particle Epidemiology Reanalysis Project. Cambridge, MA: Health Effects Institute. Retrieved on March 19,
2009 from http://es.epa.gov/ncer/science/pm/hei/Rean-ExecSumm.pdf
31 Jerrett, M., Burnett, R.T., Ma, R., et al. (2005). Spatial Analysis of Air Pollution and Mortality in Los Angeles.
Epidemiology, 16(6),727-736
32 U.S. EPA (2004). Air Quality Criteria for Paniculate Matter. Volume I EPA600/P-99/002aF and Volume II
EPA600/P-99/002bF. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at
http://www.regulations.gov/. Section 9.2.2.1.2.
33 Kiinzli, N., Jerrett, M., Mack, W.J., et al. (2004). Ambient air pollution and atherosclerosis in Los Angeles.
Environ Health Perspect, 113, 201-206
34 U.S. EPA. (2006). Air Quality Criteria for Ozone andRelatedPhotochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-
0190 at http://www.regulations.gov/.
35 U.S. EPA. (2007). Review of the National Ambient Air Quality Standards for Ozone: Policy Assessment of
Scientific and Technical Information, OAQPS Staff Paper. EPA-452/R-07-003. Washington, DC, U.S. EPA.
Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at http://www.regulations.gov/.
36 National Research Council (NRC), 2008. Estimating Mortality Risk Reduction and Economic Benefits from
Controlling Ozone Air Pollution. The National Academies Press: Washington, D.C.
37 Bates, D.V., Baker-Anderson, M., Sizto, R. (1990). Asthma attack periodicity: a study of hospital emergency
visits in Vancouver. Environ. Res., 51,51-70.
38 Thurston, G.D.; Ito, K.; Kinney, P.L.; Lippmann, M. (1992) A multi-year study of air pollution and
respiratory hospital admissions in three New York State metropolitan areas: results for 1988 and 1989 summers.
J. Exposure Anal. Environ. Epidemiol. 2:429-450.
39 Thurston, G.D., Ito, K., Hayes, C.G., Bates, D.V., Lippmann, M. (1994) Respiratory hospital admissions and
summertime haze air pollution in Toronto, Ontario: consideration of the role of acid aerosols. Environ. Res., 65,
271-290.
40 Lipfert, F.W., Hammerstrom, T. (1992). Temporal patterns in air pollution and hospital admissions. Environ.
Res., 55,374-399.
41 Burnett, R.T., Dales, R.E., Raizenne, M.E., Krewski, D., Summers, P.W., Roberts, G.R., Raad-Young, M.,
Dann,T., Brook, J. (1994). Effects of low ambient levels of ozone and sulfates on the frequency of respiratory
admissions to Ontario hospitals. Environ. Res., 65, 172-194.
42 U.S. EPA. (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-
0190 at http://www.regulations.gov/.
43 U.S. EPA. (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-
0190 at http://www.regulations.gov/.
44 Devlin, R. B., McDonnell, W. F., Mann, R., Becker, S., House, D. E., Schreinemachers, D., Koren, H. S.
(1991). Exposure of humans to ambient levels of ozone for 6.6 hours causes cellullar and biochemical changes in
3-107
-------
the lung. Am. J. Respir. CellMol. Biol, 4, 72-81.
45 Koren, H. S., Devlin, R. B., Becker, S., Perez, R., McDonnell, W. F. (1991). Time-dependent changes of
markers associated with inflammation in the lungs of humans exposed to ambient levels of ozone. Toxicol.
PathoL, 19, 406-411.
46 Koren, H. S., Devlin, R. B., Graham, D. E., Mann, R., McGee, M. P., Horstman, D. H., Kozumbo, W. J.,
Becker, S., House, D. E., McDonnell, W. F., Bromberg, P. A. (1989). Ozone-induced inflammation in the lower
airways of human subjects. Am. Rev. Respir. Dis., 39, 407-415.
47 Schelegle, E.S., Siefkin, A.D., McDonald, R.J. (1991). Time course of ozone-induced neutrophilia in normal
humans. Am. Rev. Respir. Dis., 745,1353-1358.
48 U.S. EPA. (1996). Air Quality Criteria for Ozone andRelatedPhotochemical Oxidants. EPA600-P-93-
004aF. Washington. D.C.: U.S. EPA. Retrieved on March 19, 2009 from EPA-HQ-OAR-2005-0161. p. 7-171.
49 Hodgkin, J.E., Abbey, D.E., Euler, G.L., Magie, A.R. (1984). COPD prevalence in nonsmokers in high and
low photochemical air pollution areas. Chest, 86, 830-838.
50 Euler, G.L., Abbey, D.E., Hodgkin, J.E., Magie, A.R. (1988). Chronic obstructive pulmonary disease
symptom effects of long-term cumulative exposure to ambient levels of total oxidants and nitrogen dioxide in
California Seventh-day Adventist residents. Arch. Environ. Health, 43, 279-285.
51 Abbey, D.E., Petersen, F., Mills, P.K., Beeson, W.L. (1993). Long-term ambient concentrations of total
suspended particulates, ozone, and sulfur dioxide and respiratory symptoms in a nonsmoking population. Arch.
Environ. Health, 48, 33-46.
52 U.S. EPA. (2007). Reviewof the National Ambient Air Quality Standards for Ozone: Policy Assessment of
Scientific and Technical Information, OAQPS Staff Paper. EPA-452/R-07-003. Washington, DC, U.S. EPA.
Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-0190 at http://www.regulations.gov/.
53 U.S. EPA. (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-
0190 at http://www.regulations.gov/.
54 U.S. EPA. (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-
0190 at http://www.regulations.gov/.
55 Avol, E.L., Trim, S. C., Little, D.E., Spier, C.E., Smith, M. N., Peng, R.-C., Linn, W.S., Hackney, J.D., Gross,
K.B., D'Arcy, J.B., Gibbons, D., Higgins, I.T.T. (1990 June). Ozone exposure and lung function in children
attending a southern California summer camp. Paper no. 90-150.3. Paper presented at the 83rd annual meeting
and exhibition of the Air & Waste Management Association, Pittsburgh, PA.
56 Higgins, I. T.T., D'Arcy, J. B., Gibbons, D. I., Avol, E. L., Gross, K.B. (1990). Effect of exposures to ambient
ozone on ventilatory lung function in children. Am. Rev. Respir. Dis., 141, 1136-1146.
57 Raizenne, M.E., Burnett, R.T., Stern, B., Franklin, C.A., Spengler, J.D. (1989) Acute lung function responses
to ambient acid aerosol exposures in children. Environ. Health Perspect, 75,179-185.
58 Raizenne, M.: Stern, B.: Burnett, R.; Spengler, J. (1987 June) Acute respiratory function and transported air
pollutants: observational studies. Paper no. 87-32.6. Paper presented at the 80th annual meeting of the Air
3-108
-------
Pollution Control Association, New York, NY.
59 Spektor, D. M, Lippmann, M. (1991). Health effects of ambient ozone on healthy children at a summer camp.
In: Berglund, R. L.; Lawson, D. R.; McKee, D. J., eds. Tropospheric ozone and the environment: papers from an
international conference; March 1990; Los Angeles, CA. Pittsburgh, PA: Air & Waste Management Association:
pp. 83-89. (A&WMA transaction series no. TR-19).
60 Spektor, D. M., Thurston, G.D., Mao, J., He, D., Hayes, C., Lippmann, M. (1991). Effects of single- and
multiday ozone exposures on respiratory function in active normal children. Environ. Res, 55,107-122.
61 Spektor, D. M., Lippman, M., Lioy, P. J., Thurston, G. D., Citak, K., James, D. J., Bock, N., Speizer, F. E.,
Hayes, C. (1988). Effects of ambient ozone on respiratory function in active, normal children. Am. Rev. Respir.
Dis., 757,313-320.
62 U.S. EPA. (2006). Air Quality Criteria for Ozone andRelatedPhotochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-
0190 at http://www.regulations.gov/.
63 Hazucha, M. J., Folinsbee, L. J., Seal, E., Jr. (1992). Effects of steady-state and variable ozone concentration
profiles on pulmonary function. Am. Rev. Respir. Dis., 146, 1487-1493.
64 Horstman, D.H., Ball, B.A., Folinsbee, L.J., Brown, J., Gerrity, T. (1995) Comparison of pulmonary responses
of asthmatic and nonasthmatic subjects performing light exercise while exposed to a low level of ozone.
Toxicol. Ind. Health., 11(4), 369-85.
65 Horstman, D.H.,; Folinsbee, L.J., Ives, P.J., Abdul-Salaam, S., McDonnell, W.F. (1990). Ozone concentration
and pulmonary response relationships for 6.6-hour exposures with five hours of moderate exercise to 0.08, 0.10,
and 0.12 ppm. Am. Rev. Respir. Dis., 142, 1158-1163.
66 U.S. EPA (2008). Integrated Science Assessment (ISA) for Sulfur Oxides - Health Criteria (Final Report).
EPA/600/R-08/047F. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=198843.
67 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645.
68 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 3.1.7 and 5.3.2.1.
69 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 5.4.
70 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 5.4.
71 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 5.4.
3-109
-------
72 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 5.3.2.1.
73 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 5.3.2.1 and Figure 3.1-2.
74 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 3.1.4.2.
75 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 5.3.2.1.
76 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 5.4.
77 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 3.3.1, Figure 3.3-2, Section 5.3.2.3.
78 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 5.3.2.2.
79 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 3.4.1, Figures 3.4-1 and 3.4-2.
80 Gauderman W.J., Avol E., Gilliland F., et al. (2004). The effect of air pollution on lung development from 10
to 18 years of age. NEnglJMed., 351, 1057-1067.
81 Rojas-Martinez R., Perez-Padilla R., Olaiz-Fernandez G., Mendoza-Alvarado L., Moreno-Macias H., Fortoul
T., McDonnell W., Loomis D., Romieu I. (2007) Lung function growth in children with long-term exposure to
air pollutants in Mexico City. AmJRespir Crit CareMed., 176(4), 377-84.
82 Oftedal, B. Brunekreef, B., Nystad, W., Madsen, C., Walker, S., Nafstad, P. (2008). Residential Outdoor Air
Pollution and Lung Function in Schoolchildren. Epidemiology, 19(1), 129-137.
83 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Sections 3.4.5 and 5.3.2.4.
84 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645. Section 3.4.5.
3-110
-------
85IPCS (1996) Environmental Health Criteria 171: Diesel Fuel and Exhaust Emissions, World Health
Organization, Geneva, 1996.
86 U.S. EPA (2002). Health Assessment Document for Diesel Engine Exhaust EPA/600/8-90/057F Office of
Research and Development, Washington DC. Retrieved on March 17, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=29060.
87 U.S. EPA. (1999). Guidelines for Carcinogen Risk Assessment. Review Draft. NCEA-F-0644, July.
Washington, DC: U.S. EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=54932.
88 U.S. EPA. (\m&).Guidelines for carcinogen risk assessment. EPA/630/R-00/004. Washington, DC: U.S. EPA.
Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=54933.
89 National Institute for Occupational Safety and Health (NIOSH). (1988). Carcinogenic effects of exposure to
diesel exhaust. NIOSH Current Intelligence Bulletin 50. DHHS (NIOSH) Publication No. 88-116. Atlanta, GA:
Centers for Disease Control. Retrieved March 19, 2009 from http://www.cdc.gov/niosh/88116_50.html.
90 International Agency for Research on Cancer - IARC. (1997). Silica, some silicates, coal dust and para-aramid
fibrils. In Monographs on the evaluation of carcinogenic risks to humans. Vol. 68. Lyon, France: IARC, pp.
362-375.
91 National Institute for Occupational Safety and Health (NIOSH). (1988). Carcinogenic effects of exposure to
diesel exhaust. NIOSH Current Intelligence Bulletin 50. DHHS (NIOSH) Publication No. 88-116. Atlanta, GA:
Centers for Disease Control. Retrieved March 19, 2009 from http://www.cdc.gov/niosh/88116_50.html.
92 World Health Organization International Program on Chemical Safety (1996). Diesel fuel and exhaust
emissions. In Environmental Health Criteria Vol. 171. Geneva: World Health Organization. Retrieved March 19,
2009 from http://www.inchem.org/documents/ehc/ehc/ehcl71.htm. pp.172-176.
93 California Environmental Protection Agency (Cal EPA, OEHHA). (1998). Health risk assessment for diesel
exhaust. Public and Scientific Review Draft. Sacramento, CA: Cal EPA. Retrieved March 19, 2009
ftp://ftp.arb.ca.gov/carbis/regact/diesltac/partb.pdf
94 National Toxicology Program (NTP). (2000). 9th Report on Carcinogens. Research Triangle Park, NC: Public
Health Service, U.S. Department of Health and Human Services. Available from: http://ntp-server.niehs.nih.gov.
95 Health Effects Institute (HEI). (1995). Diesel exhaust: a critical analysis of emissions, exposure, and health
effects. A Special Report of the Institute's Diesel Working Group. Cambridge, MA: Health Effects Institute.
96 Health Effects Institute (HEI). (1999). Diesel emissions and lung cancer: epidemiology and quantitative
risk assessment. A special report of the Institute's Diesel Epidemiology Expert Panel. Cambridge, MA:
Health Effects Institute (HEI).
97 Health Effects Institute (HEI). (2002). Research directions to improve estimates of human exposure and risk
assessment. A special report of the Institute's Diesel Epidemiology Working Group. Cambridge, MA: Health
Effects Institute.
3-111
-------
98 U.S. EPA (2002). Health Assessment Document for Diesel Engine Exhaust. EPA/600/8-90/057F Office of
Research and Development, Washington DC. Retrieved on March 17, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=29060. pp. 9-11.
99 Bhatia, R., Lopipero, P., Smith, A. (1998). Diesel exposure and lung cancer. Epidemiology, 9(1), 84-91.
100 Lipsett, M. Campleman, S. (1999). Occupational exposure to diesel exhaust and lung cancer: a meta-analysis.
Am J Public Health, 80(7), 1009-1017.
101 U.S. EPA (2002), National-Scale Air Toxics Assessment for 1996. This material is available electronically at
http://www.epa.gov/ttn/atw/nata/.
102 U.S. EPA. (2006). National-Scale Air Toxics Assessment for 1999. This material is available electronically
at http://www.epa.gov/ttn/atw/natal999/.
103 Ishinishi, N. Kuwabara, N. Takaki, Y., et al. (1988). Long-term inhalation experiments on diesel exhaust. In:
Diesel exhaust and health risks. Results of the HERP studies. Ibaraki, Japan: Research Committee for HERP
Studies: pp.11-84.
104 Heinrich, U., Fuhst, R., Rittinghausen, S., et al. (1995). Chronic inhalation exposure of Wistar rats and two
different strains of mice to diesel engine exhaust, carbon black, and titanium dioxide. Inhal Toxicol, 7, 553-556.
105 Mauderly, J.L., Jones, R.K., Griffith, W.C., et al. (1987). Diesel exhaust is a pulmonary carcinogen in rats
exposed chronically by inhalation. Fundam. Appl. Toxicol., 9, 208-221.
106 Nikula, K.J., Snipes, M.B., Barr, E.B., et al. (1995). Comparative pulmonary toxicities and carcinogenicities
of chronically inhaled diesel exhaust and carbon black in F344 rats. Fundam. Appl. Toxicol, 25,80-94.
107 U.S. EPA (2002). Health Assessment Document for Diesel Engine Exhaust. EPA/600/8-90/057F Office of
Research and Development, Washington DC. Retrieved on March 17, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=29060. p. 9-9.
108 Reger, R., Hancock, J., Hankinson, J., et al. (1982). Coal miners exposed to diesel exhaust emissions. Ann
OccupHyg, 26,799-815.
109 Attfield, MD. (1978). The effect of exposure to silica and diesel exhaust in underground metal and nonmetal
miners. In Kelley, W.D., (ed.), Industrial hygiene for mining and tunneling: proceedings of a topical symposium;
Cincinnati, OH: The American Conference of Governmental Industrial Hygienists, Inc.
110 Wade, J.F., III, Newman, L.S. (1993) Diesel asthma: reactive airways disease following overexposure to
locomotive exhaust. J OccupMed, 35, 149-154.
111 U.S. EPA. (2007). Chapter 3: Air Quality and Resulting Health and Welfare Effects of Air Pollution from
Mobile Sources. In Mobile Source Air Toxics Rule (Control of Hazardous Air Pollutants from Mobile Sources:
72 FR 8428, February 26, 2007) Regulatory Impact Analysis. Retrieved March 19, 2009 from
http://www.epa.gov/otaq/regs/toxics/420r07002.pdf.
112 State of California Air Resources Board. (2009 March). Rail Yard Health Risk Assessments and Mitigation
Measures. Retrieved March 19, 2009 from http://www.arb.ca.gov/railyard/hra/hra.htm.
3-112
-------
113 Di, P., Servin, A., Rosenkranz, K., Schwehr, B., Iran, H., (2006). Diesel Paniculate Matter Exposure
Assessment Study for the Ports of Los Angeles and Long Beach. Sacramento, CA: California EPA, California Air
Resources Board (CARB). Retrieved March 19, 2009 from
http://www.arb.ca.gov/regact/marine2005/portstudy0406.pdf.
114 Di, P., Servin, A., Rosenkranz, K., Schwehr, B., Tran, H., (2006). Diesel Particulate Matter Exposure
Assessment Study for the Ports of Los Angeles and Long Beach. Sacramento, CA: California EPA, California Air
Resources Board (CARB). Retrieved March 19, 2009 from
http://www.arb.ca.gov/regact/marine2005/portstudy0406.pdf.
115ICF International. September 28, 2007. Estimation of diesel particulate matter concentration isopleths for
marine harbor areas and rail yards. Memorandum to EPA under Work Assignment Number 0-3, Contract
Number EP-C-06-094. This memo is available in Docket EPA-HQ-OAR-2007-0121.
116 ICF International. September 28, 2007. Estimation of diesel particulate matter population exposure near
selected harbor areas and rail yards. Memorandum to EPA under Work Assignment Number 0-3, Contract
Number EP-C-06-094. This memo is available in Docket EPA-HQ-OAR-2007-0121.
117 ICF International. December 1, 2008. Estimation of diesel particulate matter concentration isopleths near
selected harbor areas with revised emissions (revised). Memorandum to EPA under Work Assignment Number
1-9, Contract Number EP-C-06-094. This memo is available in Docket EPA-HQ-OAR-2007-0121.
118 ICF International. December 10, 2008. Estimation of diesel particulate matter population exposure near
selected harbor areas with revised harbor emissions (revised). Memorandum to EPA under Work Assignment
Number 2-9, Contract Number EP-C-06-094. This memo is available in Docket EPA-HQ-OAR-2007-0121.
119 World Health Organization (2005). WHO Air quality guidelines for particulate matter, ozone, nitrogen
dioxide and sulfur dioxide - Global update 2005 - Summary of risk assessment. Retrieved April 14, 2009 from
http://www.who.int/phe/health_topics/outdoorair_aqg/en/index.html
120 World Health Organization (2005). WHO Air quality guidelines for particulate matter, ozone, nitrogen
dioxide and sulfur dioxide - Global update 2005 - Summary of risk assessment. Retrieved April 14, 2009 from
http://www.who.int/phe/health_topics/outdoorair_aqg/en/index.html
121 Air Quality Expert Group (2007) Air Quality and Climate Change: A UK Perspective. Retrieved April 14,
2009 from http://www.defra.gov.uk/environment/airquality/publications/airqual-climatechange/
122 Nationmaster.com Encycolpedia. Retrieved April 14, 2009 from
http://www.nationmaster.com/encyclopedia/Standard-conditions-for-temperature-and-pressure
123 NARSTO Synthesis Team (2000). An Assessment of Tropospheric Ozone Pollution: A North American
Perspective.
124 Byun, D. W., & Ching, J. K. S. (1999). Science algorithms of the EPA models-3 community multiscale air
quality (CMAQ) modeling system. Washington, DC: U.S. Environmental Protection Agency, Office of Research
and Development.
125 Byun, D.W., and Schere, K.L., 2006. Review of the Governing Equations, Computational Algorithms, and
Other Components of the Models-3 Community Multiscale Air Quality (CMAQ) Modeling System, J. Applied
Mechanics Reviews, 59(2), 51-77.
3-113
-------
126 Dennis, R.L., Byun, D.W., Novak, J.H., Galluppi, K.J., Coats, C.J., and Vouk, M.A., 1996. The next
generation of integrated air quality modeling: EPA's Models-3, Atmospheric Environment, 30, 1925-1938.
127 Aiyyer, A, Cohan, D., Russell, A., Stockwell, W., Tanrikulu, S., Vizuete, W., Wilczak, J., 2007. Final Report:
Third Peer Review of the CMAQ Model. Submitted to the Community Modeling and Analysis System Center,
Carolina Environmental Program, The University of North Carolina at Chapel Hill, 23pp.
128 Hogrefe, C., Biswas, J., Lynn, B., Civerolo, K., Ku, J. Y., Rosenthal, J., et al. (2004). Simulating regional-
scale ozone climatology over the eastern United States: model evaluation results. Atmospheric Environment. 38
(17), 2627-2638.
129 Lin, M., Oki, T., Holloway, T., Streets, D. G., Bengtsson, M., & Kanae, S. (2008). Long-range transport of
acidifying substances in East Asia-Part LModel evaluation and sensitivity studies. Atmospheric Environment 42
(24), 5939-5955.
130 United States Environmental Protection Agency. (2008). Technical support document for the final
locomotive/marine rule: Air quality modeling analyses. Research Triangle Park, N.C.: U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Air Quality Assessment Division.
131 Houyoux, M., and Vukovich, , J.M., 1999. Updates to the Sparse Matrix Operator Kernel Emission
(SMOKE) modeling system and integration with Models-3, presented at the Emission Inventory Regional
Strategies for the Future, October 26-28, 1999, Raleigh, NC, Air and Waste Management association.
132 Grell, G., J. Dudhia, and D. Stauffer, 1994: A Description of the Fifth-Generation Penn State/NCAR
Mesoscale Model (MM5), NCARTN-398+STR., 138 pp, National Center for Atmospheric Research, Boulder
CO.
133 Kemball-Cook, S., Y. Jia, C. Emery, R. Morris, Z. Wang and G. Tonnesen. 2004. 2002 Annual MM5
Simulation to Support WRAP CMAQ Visibility Modeling for the Section 308 SIP/TIP - MM5 Sensitivity
Simulations to Identify a More Optimal MM5 Configuration for Simulating Meteorology in the Western United
States. Western Regional Air Partnership, Regional Modeling Center. December 10.
(http://pah.cert.ucr.edu/aqm/308/reports/mm5/MM5SensitivityRevRep_Dec_10_2004.pdf)
134 Brewer]., P. Dolwick, and R. Gilliam. Regional and Local Scale Evaluation of MM5 Meteorological Fields
for Various Air Quality Modeling Applications, Presented at the 87th Annual American Meteorological Society
Annual Meeting, San Antonio, TX, January 15-18, 2007.
135 Kemball-Cook, S., Y. Jia, C. Emery, R. Morris, Z. Wang and G. Tonnesen. 2004. 2002 Annual MM5
Simulation to Support WRAP CMAQ Visibility Modeling for the Section 308 SIP/TIP - MM5 Sensitivity
Simulations to Identify a More Optimal MM5 Configuration for Simulating Meteorology in the Western United
States. Western Regional Air Partnership, Regional Modeling Center. December 10.
(http://pah.cert.ucr.edu/aqm/308/reports/mm5/MM5SensitivityRevRep_Dec_10_2004.pdf)
136 Yantosca, B., 2004. GEOS-CHEMv7-01-02 User's Guide, Atmospheric Chemistry Modeling Group, Harvard
University, Cambridge, MA, October 15, 2004.
137 Gilliam, R. C., W. Appel, and S. Phillips. The Atmospheric Model Evaluation Tool (AMET): Meteorology
Module. Presented at 4th Annual CMAS Models-3 Users Conference, Chapel Hill, NC, September 26 - 28,
2005.
3-114
-------
138 United States Environmental Protection Agency. (2008). Technical support document for the final
locomotive/marine rule: Air quality modeling analyses. Research Triangle Park, N.C.: U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Air Quality Assessment Division.
139 U.S. EPA (2004). Air quality criteria document for particulate matter. Volumes I and II (Report no.
EPA/600/P-99/002aF; 2 Volumes). Research Triangle Park, NC; National Center for Environmental
Assessment-RTF; Office of Research and Development; U.S. Environmental Protection Agency.
140 LyyranenJ; Jokiniemi, J; Kauppinen,E; JoutsensaariJ (1999). Aerosol characterisation in medium-speed
diesel engines operating with heavy fuel oils. J. Aerosol ScL, No.6, 771-784.
141 Capaldo, K; Corbett, J; Kasibhatla, P; Fischbeck, P; Pandis, S (1999). Effects of ship emissions on sulfur
cycling and radiative climate forcing over the ocean. Nature, 400, 743-746.
142 U.S. EPA (2008). Nitrogen Dioxide/SulfurDioxide SecondaryNAAQS Review: Integrated Science
Assessment (ISA). (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
143 U.S. EPA (2008). U.S. EPA's 2008 Report on the Environment (Final Report). U.S. Environmental Protection
Agency, Washington, D.C., EPA/600/R-07/045F (NTIS PB2008-112484.).
144 U.S. EPA (2008, August). Risk and Exposure Assessment for the Review of the Secondary National Ambient
Air Quality Standards for Oxides of Nitrogen and Oxides of Sulfur (Draft). First Draft Chapter 1-6. U.S.
Environmental Protection Agency,, EPA-452?P-08-005a..
145 U.S. EPA (2006) Wadeable Streams Assessment: A collaborative survey of the nation's streams.
Washington, DC; Office of Water; Office of Research and Development; U.S. Environmental Protection
Agency, EPA-841-B-06-002.
146 Lawrence GB; Roy KM; Baldigo BP; Simonin HA; Capone SB; Sutherland JW; Nierzwicki-Bauer SA;
Boylen CW, (2008). Chronic and episodic acidification of Adirondack streams from Acid rain in 2003-2005.
Journal Environ Quality, 37, 2264-2274.
147 Baker JP; Bernard DP; Christensen SW; Sale MJ, (1990). Biological effects of changes in surface water acid-
base chemistry. (State of science / technology report f 13) .Washington DC; National Acid Precipitation
Assessment Program.
148 Kaufmann PR; Herlihy AT; Elwood JW; Mitch ME; Overton WS; Sale MJ; Messer JJ; Cougan KA; Peck
DV; Reckhow KH; Kinney AJ; Christie SJ; Brown DD; Hagley CA; Jager HI. (1988). Chemical characteristics
of streams in the Mid-Atlantic and Southeastern United States. Volume I: Population descriptions and physico-
chemical relationships. Washington, DC; U.S. Environmental Protection Agency, EPA/600/3-88/02 la.
149Kaufmann PR; Herlihy AT; Mitch ME; Messer JJ; Overton WS, (1991). Stream chemistry in the eastern
United States: Synoptic survey design, acid-base status, and regional patterns. Water Resoure Res, 27, 611-627.
150 Landers DH; Filers JM; Brakke DF; Overton WS; Kellar PE; Silverstein WE; Schonbrod RD; Crowe RE;
Linthurst RA; Omernik JM; Teague SA; Meier EP,(1987). Western lake survey phase I: Characteristics of lakes
in the western United States. Volume I: Population descriptions and physico-chemical relationships.
Washington, D.C. U.S. Environmental Protection Agency, EPA/600/3-86/054a.
3-115
-------
151 Linthurst RA; Landers DH; Eilers JM; Brakke DF; Overton WS; Meier EP; Crowe RE.(1986). Characteristics
of Lakes in the Eastern United States. Volume I. Population descriptions and physico-chemical relationships.
Washington, DC; U.S. Environmental Protection Agency, EPA-600/4-86-007A.
152 Linthurst RA; Landers DH; Eilers JM; Kellar PE; Brakke DF; Overton WS; Crowe R; Meier EP; Kanciruk P;
Jeffries DS, (1986). Regional chemical characteristics of lakes in North America Part II: Eastern United States.
Water Air Soil Pollution, 31, 577-591.
153 Stoddard J; Kahl JS; Deviney FA; DeWalle DR; Driscoll CT; Herlihy AT; Kellogg JH; Murdoch PS; Webb
JR; Webster KE, (2003). Response of surface water chemistry to the Clean Air Act Amendments of 1990. Office
of Research and Development, RTF; U.S. Environmental Protection Agency. EPA 620/R-03/001.
154 Charles DF, (1991). Christie, S. (Ed.). Acidic deposition and aquatic ecosystems: Regional case studies. New
York: Springer-Verlag.
155 Landers DH; Eilers JM; Brakke DF; Overton WS; Kellar PE; Silverstein WE; Schonbrod RD; Crowe RE;
Linthurst RA; Omernik JM; league SA; Meier EP,. (1987). Western lake survey phase I: Characteristics of lakes
in the Western United States. Volume I: Population descriptions and physico-chemical relationships.
Washington, D.C.; U.S. Environmental Protection Agency. EPA/600/3-86/054a.
156 Nelson PO. (1991). Cascade Mountains: Lake chemistry and sensitivity of acid deposition. In: Charles DF,
Christie S (Eds.), Acidic deposition and aquatic ecosystems: regional case studies (pp. 531-563). New York:
Springer-Verlag.
157 Williams MW; Tonnessen KA, (2000). Critical loads for inorganic nitrogen deposition in the Colorado Front
Range, USA. EcolAppl, 10, 1648-1665.
158 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA).(Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
159 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
160 Peterjohn WT; Adams MB; Gilliam FS (1996). Symptoms of nitrogen saturation in two central Appalachian
hardwood forest ecosystems. Biogeochemistry, 35, 507-522.
161 Cook RB; Elwood JW; Turner RR; Bogle MA; Mulholland PJ; Palumbo AV (1994). Acid-base chemistry of
high-elevation streams in the Great Smoky Mountains. Water Air Soil Pollution, 72, 331-356.
162 Aber JD; Nadelhoffer KJ; Steudler P; Mellilo JM (1989). Nitrogen saturation in northern forest ecosystems.
Excess nitrogen from fossil fuel combustion may stress the biosphere. Bioscience, 39, 378-386.
163 Aber JD; McDowell W; Nadelhoffer K; Magill A; Berntson G; Kamakea M; McNulty S; Currie W; Rustad L;
Fernandez I (1998). Nitrogen saturation in temperate forest ecosystems: Hypotheses revisited. Bioscience, 48,
921-934.
164 Edwards PM; Helvey JD (1991). Long-term ionic increases from a central Appalachian forested watershed. J
Environ Quality, 20, 250-255.
165 Peterjohn WT; Adams MB; Gilliam FS (1996). Symptoms of nitrogen saturation in two central Appalachian
hardwood forest ecosystems. Biogeochemistry, 35, 507-522.
3-116
-------
166 Adams MB; Angradi TR; Kochenderfer JN (1997). Stream water and soil solution responses to 5 years of
nitrogen and sulfur additions at the Fernow Experimental Forest, West Virginia. For EcolManage, 95, 79-91.
167 Adams MB; Burger JA; Jenkins AB; Zelazny L (2000). Impact of harvesting and atmospheric pollution on
nutrient depletion of eastern U.S. hardwood forests. For Ecol Manage, 138, 301-319.
168 Bytnerowicz A; Fenn ME (1996). Nitrogen deposition in California forests: A review. Environ Pollut, 92,
127-146.
169 Fenn ME; Poth MA (1998). Indicators of nitrogen status in California forests. (General technical report PSW-
GTR-166). Washington, DC; U.S. Forest Service; U.S. Department of Agriculture (USDA).
170 Aber JD; McDowell W; Nadelhoffer K; Magill A; Berntson G; Kamakea M; McNulty S; Currie W; Rustad L;
Fernandez I (1998). Nitrogen saturation in temperate forest ecosystems: Hypotheses revisited. Bioscience, 48,
921-934.
171 Clark CM; Tilman D (2008). Loss of plant species after chronic low-level nitrogen deposition to prairie
grasslands. Nature, 451, 712-715.
172 Conner R; Seidl, A; VanTassell.L; Wilkins, N (2001). United States grasslands and related resources an
economic and biological trends assessment. Texas A & M University.
http://landinfo.tamu.edu/presentations/grasslands.html
173 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA).(Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
174 Baron JS; Ojima DS; Holland EA; Parton WJ (1994). Analysis of nitrogen saturation potential in Rocky
Mountain tundra and forest: Implications for aquatic systems. Biogeochemistry, 27, 61-82.
175 Williams MW; Tonnessen KA (2000). Critical loads for inorganic nitrogen deposition in the Colorado Front
Range, USA. EcolAppl, 10, 1648-1665.
176 U.S. EPA (2009). Office of Water website on Wetlands. http//www.epa.gov/owow/wetlands
177 Moore DRJ; Keddy PA; Gaudet CL; Wisheu 1C (1989). Conservation of wetlands: Do infertile wetlands
deserve a higher priority? Biological Conservation, 47, 203-217.
178 Krupa SV (2003). Effects of atmospheric ammonia (NH3).on terrestrial vegetation: A review. Environ
Pollution, 124, 179-221.
179 U.S. EPA (2009). Office of Water website on Wetlands. http//www.epa.gov/owow/wetlands
180 U.S. EPA (2009). Office of Water website on Wetlands. http//www.epa.gov/owow/wetlands
181 U.S. EPA (2009). Office of Water website on Wetlands. http//www.epa.gov/owow/wetlands
182 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA).(Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
183 Cook RB; Elwood JW; Turner RR; Bogle MA; Mulholland PJ; Palumbo AV (1994). Acid-base chemistry of
high-elevation streams in the Great Smoky Mountains. Water Air Soil Pollut, 72, 331-356.
3-117
-------
184 Fenn ME; Poth MA; Johnson DW (1996). Evidence for nitrogen saturation in the San Bernardino Mountains
in southern California. Forest Ecology Manage, 82, 211-230.
185 Baron JS; Ojima DS; Holland EA; Parton WJ (1994). Analysis of nitrogen saturation potential in Rocky
Mountain tundra and forest: Implications for aquatic systems. Biogeochemistry, 27, 61-82.
186 Williams MW; Baron JS; Caine N; Sommerfeld R; Sanford JR (1996). Nitrogen saturation in the Rocky
Mountains. Environ Sci Technol, 30, 640-646.
187 Gilliam FS; Adams MB; Yurish BM (1996). Ecosystem nutrient responses to chronic nutrient inputs at
Fernow Experimental Forest, West Virginia. Can J For Res, 26, 196-205.
188 Murdoch PS; Stoddard JL (1992). The role of nitrate in the acidification of streams in the Catskill Mountains
of New York. Water Resour Res,28, 2707-2720.
189 Stoddard JL; Murdoch PS (1991). Catskill Mountains: An overview of the impact of acidifying pollutants on
aquatic resources. In: Charles DF (Ed.), Acidic deposition and aquatic ecosystems: Regional case studies, (pp
237-271). New York: Springer-Verlag, Inc.
190 Wigington PJ; Baker JP; DeWalle DR; Kretser WA; Murdoch PS; Simonin HA; Van Sickle J; McDowell
MK; Peck DV; Barchet WR (1996). Episodic acidification of small streams in the northeastern United States:
Episodic response project. EcolAppl, 6, 374-388.; Wigington PJ Jr; DeWalle DR; Murdoch PS; Kretser WA;
Simonin HA; Van Sickle J; Baker JP. (1996). Episodic acidification of small streams in the northeastern United
States: Ionic controls of episodes. EcolAppl, 6, 389-407.
191 Paerl HW; Bales JD; Ausley LW; Buzzelli CP; Crowder LB; Eby LA; Go M; Peierls BL; Richardson TL;
Ramus JS (2001b) Ecosystem impacts of three sequential hurricanes (Dennis, Floyd, and Irene) on the United
States' largest lagoonal estuary, Pamlico Sound, NC. ProcNatlAcadSci USA, 98, 5655-5611; Paerl HW. (2002).
Connecting atmospheric deposition to coastal eutrophication. Environ Sci Technol, 36, 323A-326A. Paerl HW;
Dennis RL; Whitall DR (2002). Atmospheric deposition of nitrogen: Implications for nutrient over-enrichment
of coastal waters. Estuaries, 25, 677-693.
192 Bricker S; Longstaff B; Dennison W; Jones A; Boicourt K; Wicks C; Woerner J (2007). Effects of nutrient
enrichment in the nation's estuaries: A decade of change.
http://ccmaserver.nos.noaa.gov/publications/eutroupdate/. (NOAA Coastal Ocean Program Decision Analysis
Series No. 26). Silver Spring, MD: National Centers for Coastal Ocean Science, NOAA.
193 Howarth RW; Marino R (2006). Nitrogen as the limiting nutrient for eutrophication in coastal marine
ecosystems: evolving views over three decades. Limnol Oceanogr, 51, 364-376.
194 Elser JJ; Bracken MES; Cleland EE; Gruner DS; Harpole WS; Hillebrand II H; Ngai JT; Seabloom EW;
Shurin JB; Smith JE (2007). Global analysis of nitrogen and phosphorus limitation of primary producers in
freshwater, marine, and terrestrial ecosystems. EcolLett, 10, 1135-1142.
195 Bricker S; Longstaff B; Dennison W; Jones A; Boicourt K; Wicks C; Woerner J (2007). Effects of nutrient
enrichment in the nation's estuaries: A decade of change.
http://ccmaserver.nos.noaa.gov/publications/eutroupdate/. (NOAA Coastal Ocean Program Decision Analysis
Series No. 26). Silver Spring, MD: National Centers for Coastal Ocean Science, National Oceanic and
Atmospheric Administration (NOAA).
196U.S. EPA (2004, October). Air Quality Criteria for Paniculate Matter (Final Report) U.S. Environmental
Protection Agency, Washington, DC, EPA 600/P-99/002aF-bF.
3-118
-------
197 Gao, Y., E.D. Nelson, M.P. Field, etal. (2002). Characterization of atmospheric trace elements on PM2.5
particulate matter over the New York-New Jersey harbor estuary. Atmos. Environ. 36: 1077-1086.
198 U.S. EPA. (2004, October). Air Quality Criteria Document for Particulate Matter. (Final Report), U.S.
Environmental Protection Agency, Washington, DC, EPA 600/P-99/002aF-bF, 2004.
199 U.S. EPA (2004, October). Air Quality Criteria Document for Particulate Matter. (Final Report), U.S.
Environmental Protection Agency, Washington, DC, EPA 600/P-99/002aF-bF.
200 U.S. EPA (2004, October). Air Quality Criteria Document for Particulate Matter. (Final Report), U.S.
Environmental Protection Agency, Washington, DC, EPA 600/P-99/002aF-bF.
201 Paerl HW; Pinckney JL; Steppe TF (2000). Cyanobacterial-bacterial mat consortia: examining the functional
unit of microbial survival and growth in extreme environments. Environmental Microbiology, 2, 11-26.
202 Swackhamer DL; Paerl HW; Eisenreich SJ; Hurley J; Hornbuckle KC; McLachlan M; Mount D; Muir D;
Schindler D (2004). Impacts of atmospheric pollutants on aquatic ecosystems. Issues in Ecology, 12, 1-24.
203 Paerl HW (2002). Connecting atmospheric deposition to coastal eutrophication. Environ Sci Technol, 36,
323A-326A.
204 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
205 Mathur R; Dennis RL (2000). A regional modeling analysis of reduced nitrogen cycling in the eastern United
States. In: Preprints of the symposium on atmospheric chemistry: Issues in the 21st century - held in Long
Beach, CA. in January 2000 (p. 85-88). Boston, MA. American Meteorological Society. Dennis, 2001.
206 Mathur R: Dennis RL (2003). Seasonal and annual modeling of reduced nitrogen compounds over the eastern
United States: Emissions, ambient levels, and deposition amounts. J Geophys Res-Atmos, 108.
207 Dennis, R (1997). Using the regional acid deposition model to determine the nitrogen deposition airshed of
the Chesapeake Bay watershed. In: Baker JE (Ed.), Atmospheric deposition of contaminants to the Great Lakes
and coastal waters, (pp 393-413). Pensacola, FL: Society of Environmental Toxicology and Chemistry Press.
208 Dennis RL; Binkowski FS; Clark TL; McHenry JN; McHenry SJ; Raynolds SK (1990). Selected applications
of the regional acid deposition model and engineering model, appendix 5F (part 2).of NAP AP SOS/T report 5.
In: Irving PM, ed. National Acid Precipitation Assessment Program: State of Science and Technology, Volume
1. Washington, DC: National Acid Precipitation Assessment Program.
209 Dennis RL; Mathur R (2001). Airshed domains for modeling atmospheric deposition of oxidized and reduced
nitrogen to the Neuse/Pamlico system of North Carolina. HydrolSci Technol, 17, 107-118.
210 U.S. EPA (2008). Nitrogen Dioxide/SulfurDioxide SecondaryNAAQS Review: Integrated Science
Assessment (ISA).(Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
211 U.S. EPA (2008). Nitrogen Dioxide/SulfurDioxide SecondaryNAAQS Review: Integrated Science
Assessment (ISA). (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
212U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA).(Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
3-119
-------
213 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA).(Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
214 Dillman, Karen L, L.H. Geiser and G. Brenner. (2007). Air Quality Biomonitoring with Lichens: The
Tongass National Forest. JSFS Tongass National Forest. Unpublished report.
215 ADF&G (2008). Predator Management for the Southern Alaska Peninsula Caribou Herd. Alaska Dept. of
Fish and Game, Division of Wildlife Conservation
216 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA).(Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
217 Agrawal, H., et al. (2008). In-use gaseous and particulate matter emissions from a modern ocean going
container vessel. Atmospheric Environment,dol:10.1Q16/j.atmosenv.2008.02.053.
218 Isakson J., Persson T.A., E. Selin Lindgren E. (2001) Identification and assessment of ship emissions and
their effects in the harbour of Gteborg, Sweeden. Atmospheric Environment, 35(21), 3659-3666.
219 Miller, W., etal. (2008 June 10). Measuring Emissions from Ocean Going Vessels. Presentation presented at
the Fuel, Engines, and Control Devices Workshop, San Pedro, California.
220 U.S. EPA (2004, October). Air Quality Criteria Document for Particulate Matter (Final Report). U.S.
Environmental Protection Agency, Washington, DC, EPA 600/P-99/002aF-bF.
221 Gawel, J. E.; Ahner, B. A.; Friedland, A. J.; Morel, F. M. M. (1996) Role for heavy metals in forest decline
indicated by phytochelatin measurements. Nature (London), 381, 64-65.
222U.S. EPA (2004, October). Air Quality Criteria Document for Particulate Matter (Final Report). U.S.
Environmental Protection Agency, Washington, DC, EPA 600/P-99/002aF-bF, 2004.
223Cotrufo M.F., De Santo A.V., Alfani A., Bartoli G., De Cristofaro A. (1995) Effects of urban heavy metal
pollution on organic matter decomposition in Quercus ilex L. Woods. Environmental Pollution, 89(1), 81-87.
224 Niklinska M., Laskowski R., Maryanski M. (1998). Effect of heavy metals and storage time on two types of
forest litter: basal respiration rate and exchangeable metals. Ecotoxicological Environmental Safety, 41, 8-
IS.Niklinski etal., 1998 - from PM AQCD
225 U.S. EPA (2004, October). Air Quality Criteria for Particulate Matter (Final Report) U.S. Environmental
Protection Agency, Washington, DC, EPA 600/P-99/002aF-bF.
226SimcikM.F., Eisenreich, S.J., Golden K.A., et al. (1996) Atmospheric Loading of Polycyclic Aromatic
Hydrocarbons to Lake Michigan as Recorded in the Sediments. Environmental Science and Technology, 30,
3039-3046..
227 Simcik M.F., Eisenreich S.J., Lioy P.J. (1999) Source apportionment and source/sink relationship of PAHs in
the coastal atmosphere of Chicago and Lake Michigan. Atmospheric Environment, 33, 5071-5079.
228 Dickhut R.M., Canuel E.A., Gustafson K.E., Liu K., Arzayus K.M., Walker S.E., Edgecombe G., Gaylor
M.O., MacDonald E.H. (2000). Automotive Sources of Carcinogenic Polycyclic Aromatic Hydrocarbons
3-120
-------
Associated with Particulate Matter in the Chesapeake Bay Region. Environmental Science & Technology,
34(21), 4635-4640.
229 Poor N., Tremblay R., Kay H., et al. (2002) Atmospheric concentrations and dry deposition rates of
polycyclic aromatic hydrocarbons (PAHs) for Tampa Bay, Florida, USA. Atmospheric Environment, 38, 6005-
6015.
230Arzavus K.M., Dickhut R.M., Canuel E.A. (2001) Fate of Atmospherically Deposited Polycyclic Aromatic
Hydrocarbons (PAHs) in Chesapeake Bay. Environmental Science & Technology, 35, 2178-2183.
231 Simcik M.F., Eisenreich, S.J., Golden K.A., et al. (1996) Atmospheric Loading of Polycyclic Aromatic
Hydrocarbons to Lake Michigan as Recorded in the Sediments. Environmental Science and Technology, 30,
3039-3046.
232 Simcik M.F., Eisenreich S.J., Lioy P.J. (1999) Source apportionment and source/sink relationship of PAHs in
the coastal atmosphere of Chicago and Lake Michigan. Atmospheric Environment, 33, 5071-5079.
233 Aerts R. (1990). Nutrient use efficiency in evergreen and deciduous species from heathland. Oecologia, 84,
391-397.
234 Aerts R; Berendse F; De Caluwe H; Schmits M (1990). Competition in heathland along an experimental
gradient of nutrient availability. Oikos, 57, 310-318.
235 Krupa SV (2003). Effects of atmospheric ammonia (NH3).on terrestrial vegetation: A review. Environ Pollut,
124,179-221.
236 Tilman D; Wedin D. (1991). Dynamics of nitrogen competition between successional grasses. Ecology, 72,
1038-1049.
237 Ellenberg H. (1985). Veranderungen der floa mitteleuropas unter dem einfluss von diingung und
immissionen. SchweizZForstwesten, 136, 19-39.
238 Falkengren-Grerup U (1986). Soil acidification and vegetation changes in deciduous forest in southern
Sweden. Oecologia, 70, 339-347. Falkengren-Grerup U. (1989). Soil acidification and its impact on ground
vegetation. Ambio, 18, 179-183.
239 Roelofs JGM. (1986). The effect of airborne sulfur and nitrogen deposition on aquatic and terrestrial
heathland vegetation. Experientia, 42, 372-377.
240 Stevens CJ; Dise NB; Mountford OJ; Cowing DJ (2004). Impact of nitrogen deposition on the species
richness of grasslands. Science, 303, 1876-1878.
241 Ellenberg H (1987). Floristic changes due to eutrophication. In Asman WAH; Diederen HSMA (Eds.).
Proceedings of the ammonia and acidification symposium of the European association for the science of air
pollution (EURASAP), held in Bilthoven, The Netherlands. April 13 - 15, 1987; (pp 301-308). European
Association for the Science of Air Pollution (EURASAP).
242 Kenk G; Fischer H (1988). Evidence from nitrogen fertilisation in the forests of Germany. Environ Pollut, 54,
199-218.
243 U.S. EPA (1993). Air Quality Criteria Document for Oxides of Nitrogen (Final). U.S. Environmental
Protection Agency. EPA/600/8-9!/049aF-cF; 3 Volumes.
3-121
-------
244 U.S. EPA 2008. Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science Assessment
(ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
245 U.S. EPA 2008. Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science Assessment
(7X4;(Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
246 LeBauer DS; Treseder KK (2008). Nitrogen limitation of net primary productivity in terrestrial ecosystems is
globally distributed. Ecology, 89, 371-379.
247 Fenn ME; Baron; JS, Allen, EB, Rueth, HM, Nydick KR, Geiser, L, Bowman, WD, Sickman, JO, Meixner, T,
Johnson, DW, Neitlich, P. (2003). Ecological effects of nitrogen deposition in the western United States.
Bioscience, 53, 404-420.
248 Stoddard JL (1994). Long-term changes in watershed retention of nitrogen: its causes and aquatic
consequences. In Baker LA (Ed.), Environmental chemistry of lakes and reservoirs, (pp. 223-284). Washington,
D.C.: American Chemical Society.
249 U.S. Fish and Wildlife Service (2005). Dahl TE (Ed.). Status and Trends of Wetlands in the Conterminous
United States 1998 to 2004. Washington, DC; Fisheries and Habitat Conservation; U.S., Fish and Wildlife
Service (FWS); Department of the Interior.
http://wetlandsfws.er.usgs.gov/status_trends/national_reports/trends_2005_report.pdf.
250 Bridgham SD; Pastor J; McClaugherty CA; Richardson CJ (1995). Nutrient-use efficiency: A litterfall index,
a model, and a test along a nutrient availability gradient in North Carolina peatlands. American Naturalist, 145,
1-21.
251 Bridgham SD; Pastor J; Janssens J; Chapin C; Malterer T. (1996). Multiple limiting gradients in peatlands: A
call for a new paradigm. Wetlands, 16, 45-65.
252 Shaver GR; Melillo JM (1984). Nutrient budgets of marsh plants: Efficiency concepts and relation to
availability. Ecology, 65, 1491-1510.
253 Morris JT (1991). Effects of nitrogen loading on wetland ecosystems with particular reference to atmospheric
deposition. Annual Review of Ecological Systems, 22, 257-279.
254 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
255 Bedford BL; Godwin KS (2003). Fens of the United States: Distribution, characteristics, and scientific
connection versus legal isolation. Wetlands, 23, 608-629.
256 Moore DRJ; Keddy PA; Gaudet CL; Wisheu 1C (1989). Conservation of wetlands: Do infertile wetlands
deserve a higher priority? Biological Conservation, 47, 203-217.
257 U.S. EPA. (1993). Air Quality Criteria Document for Oxides of Nitrogen. (Final). U.S. EPA EPA/600/8-
91/049aF-cF; 3 Volumes.
258 Redbo-Torstensson P. (1994). The demographic consequences of nitrogen fertilization of a population of
sundew, Drosera rotundifolia. Acta Botanica Neerlandica, 43, 175-188.
3-122
-------
259 U.S. Department of Agriculture (2009). Natural Resource Conservation Services Website. Plant Database:
http://plants.usda.gov/
260 Ellison AM: Gotelli NJ (2002). Nitrogen availability alters the expression of carnivory in the northern pitcher
plant, Sarracenia purpurea. Proceedings Natl Academy of SciencellSA, 99, 4409-4412.
261 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
262 Baron JS (2006). Hindcasting nitrogen deposition to determine ecological critical load. EcolAppl, 16, 433-
439.
263 Baron JS: Rueth HM; Wolfe AM: Nydick KR; Allstott EJ; Minear JT; Moraska B (2000). Ecosystem
responses to nitrogen deposition in the Colorado Front Range. Ecosystems, 3, 352-368.
264 Interlandi SJ; Kilham SS (1998). Assessing the effects of nitrogen deposition on mountain waters: a study of
phytoplankton community dynamics. Water Science and Technology, 38, 139-146.
265 Saros JE; Michel TJ; Interlandi SJ: Wolfe AP (2005). Resource requirements of Asterionella formosa and
Fragilaria crotonensis'm oligotrophic alpine lakes: implications for recent phytoplankton community
reorganizations. Can JFish Aquat Sci, 62, 1681-1689.
266 Saros JE: Interlandi SJ: Wolfe AP: Engstrom DR (2003). Recent changes in the diatom community structure
of lakes in the Beartooth Mountain Range, USA. Arct Antarct Alp Res, 35, 18-23.
267 Wolfe AP; Van Gorpe AC: Baron JS. (2003). Recent ecological and biogeochemical changes in alpine lakes
of Rocky Mountain National Park (Colorado, USA): A response to anthropogenic nitrogen deposition.
Geobiology, 1, 153-168.
268 Wolfe AP: Baron JS: Cornett RJ. (2001). Anthropogenic nitrogen deposition induces rapid ecological changes
in alpine lakes of the Colorado Front Range (USA). JPaleolimnol, 25, 1-7.
269 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
270 Boynton WR: Garber JH; Summers R: Kemp WM (1995). Inputs, transformations, and transport of nitrogen
and phosphorus in Chesapeake Bay and selected tributaries. Estuaries, 18, 285-314.
271 Howarth RW: Billen G: Swaney D: Townsend A: Jaworski N; Lajtha K; Downing JA: Elmgren R; Caraco N;
Jordan T; Berendse F; Freney J: Kudeyarov V: Murdoch PS: Zhao-Liang Z (1996). Regional nitrogen budgets
and riverine N & P fluxes for the drainages to the North Atlantic Ocean: natural and human influences.
Biogeochemistry, 35, 75-139.
272Paerl H (1995). Coastal eutrophication in relation to atmospheric nitrogen deposition: current perspectives.
Ophelia, 41,237-259.
273 Paerl H (1997). Coastal eutrophication and harmful algal blooms: importance of atmospheric deposition and
groundwater as "new" nitrogen and other nutrient sources. Limnol Oceanogr, 42, 1154-1162.
274 Valiela I; Costa JE (1988). Eutrophication of Buttermilk Bay, a Cape Cod coastal embayment: concentrations
of nutrients and watershed nutrient budgets. Environ Manage, 12, 539-553.
3-123
-------
275 Valiela I; Costa JE; Foreman K (1990). Transport of groundwater-borne nutrients from watersheds and their
effects on coastal waters. Biogeochemistry, 10, 177-197.
276 Costanza R; d'Arge R; de Groot R; Farber S; Grasso M; Hannon B; Limberg K; Naeem S; O'Neill RV;
Paruelo J; Raskin RG; Sutton P; Van Den Belt M (1997). The value of the world's ecosystem services and
natural capital. Nature, 387, 253-259.
277 D'Elia CJ; Sanders JG; Boynton WR (1986). Nutrient enrichment studies in a coastal plain estuary:
phytoplankton growth in large-scale, continuous cultures. Can J Fish Aquat Sci, 43, 397-406.
278 Howarth RW; Marino R (2006). Nitrogen as the limiting nutrient for eutrophication in coastal marine
ecosystems: evolving views over three decades. Limnol Oceanogr, 51, 364-376.
279 Bricker S: Longstaff B; Dennison W; Jones A: Boicourt K: Wicks C; Woerner J (2007). Effects of nutrient
enrichment in the nation's estuaries: A decade of change. (NOAA Coastal Ocean Program Decision Analysis
Series No. 26). Silver Spring, MD: National Centers for Coastal Ocean Science, National Oceanic and
Atmospheric Administration (NOAA). http://ccmaserver.nos.noaa.gov/publications/eutroupdate/.
280 Bricker S: Longstaff B: Dennison W: Jones A: Boicourt K: Wicks C; Woerner J (2007). Effects of nutrient
enrichment in the nation's estuaries: A decade of change. (NOAA Coastal Ocean Program Decision Analysis
Series No. 26). Silver Spring, MD: National Centers for Coastal Ocean Science, National Oceanic and
Atmospheric Administration (NOAA). http://ccmaserver.nos.noaa.gov/publications/eutroupdate/.
281 Bricker SB: Clement CG; Pirhalla DE; Orlando SP; Farrow DGG (1999). National estuarine eutrophication
assessment: Effects of nutrient enrichment in the nation's estuaries. Silver Spring, MD: Special Projects Office
and the National Centers for Coastal Ocean Science, National Ocean Service, National Oceanic and Atmospheric
Administration (NOAA).
282 Bricker S: Longstaff B: Dennison W: Jones A: Boicourt K: Wicks C; Woerner J (2007). Effects of nutrient
enrichment in the nation's estuaries: A decade of change. (NOAA Coastal Ocean Program Decision Analysis
Series No. 26). Silver Spring, MD: National Centers for Coastal Ocean Science, National Oceanic and
Atmospheric Administration (NOAA). http://ccmaserver.nos.noaa.gov/publications/eutroupdate/.
283 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (7X4,)(Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
284 Bricker OP: Rice KC (1989). Acidic deposition to streams: a geology-based method predicts their sensitivity.
Environ Sci Technol, 23, 379-385.
285 Stauffer RE (1990). Granite weathering and the sensitivity of alpine lakes to acid deposition. Limnol
Oceanogr, 35(5), 1112-1134.
286 Stauffer RE: Wittchen BD (1991). Effects of silicate weathering on water chemistry in forested, upland, felsic
terrain of the USA. Geochim Cosmochim Acta, 55, 3253-3271.
287 Vertucci FA; Filers JM (1993). Issues in monitoring wilderness lake chemistry: a case study in the Sawtooth
Mountains, Idaho. Environ Monit Assess, 28, 277-294.
288 Sullivan TJ; Webb JR; Snyder KU; Herlihy AT; Cosby BJ (2007). Spatial distribution of acid-sensitive and
acid-impacted streams in relation to watershed features in the southern Appalachian mountains. Water Air Soil
Pollut, 182, 57-71.
3-124
-------
289 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
290 Joslin JD; Kelly JM; van Miegroet H (1992). Soil chemistry and nutrition of North American spruce-fir
stands: evidence for recent change. J Environ Quality, 21, 12-30.
291 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
292 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
292 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
293 DeHayes DH; Schaberg PG; Hawley GJ; Strimbeck GR (1999). Acid rain impacts on calcium nutrition and
forest health. Bioscience, 49, 789-800.
294 Webster KL; Creed IF; Nicholas NS; Miegroet H (2004). Exploring interactions between pollutant emissions
and climatic variability in growth of red spruce in the Great Smoky Mountains National Park. Water Air Soil
Pollut, 159, 225-248.
295 DeHayes DH; Schaberg PG; Hawley GJ; Strimbeck GR. (1999). Acid rain impacts on calcium nutrition and
forest health. Bioscience, 49, 789-800.
296 Hawley GJ; Schaberg PG; Eagar C; Borer CH. (2006). Calcium addition at the Hubbard Brook Experimental
Forest reduced winter injury to red spruce in a high-injury year. Can J Forest Res, 36, 2544-2549.
297 U.S. EPA (2008) Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F. Tree distribution data was
obtained from Little's Atlas. Little EL Jr. (1971). Atlas of United States trees. Vol. 1: Conifers and important
hardwoods (USDA. Misc. Publ, no. 1146).Washington, DC, U.S. Forest Service; U.S. Department of
Agriculture (USDA).
298 Hamburg SP; Yanai RD; Arthur MA; Blum JD; Siccama TG (2003). Biotic control of calcium cycling in
northern hardwood forests: Acid rain and aging forests. Ecosystems, 6, 399-406.
299 Driscoll CT; Lawrence GB; Bulger AJ; Butler TJ; Cronan CS; Eagar C; Lambert KF; Likens GE; Stoddard
JL; Weather KC (2001). Acidic deposition in the northeastern United States: Sources and inputs, ecosystem
effects, and management strategies. Bioscience, 51, 180-198.
300 Drohan PJ; Stout SL; Petersen GW (2002). Sugar maple (AcersaccharumMarsh.) decline during 1979-1989
in northern Pennsylvania. ForEcolManage, 170, 1-17.
301 Holzmueller E; Jose S; Jenkins M; Camp A; Long A (2006). Dogwood anthracnose in eastern hardwood
forests: What is known and what can be done? Journal of Forestry, 104, 21-26.
302 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
3-125
-------
303 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
304 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
305 Fremstad E; Paal J; Mols T (2005). Impacts of increased nitrogen supply on Norwegian lichen-rich alpine
communities: A 10 year experiment. JEcol, 93, 471-481.
306 Fenn ME; Baron JS; Allen EB; Rueth HM; Nydick KR; Geiser L; Bowman WD; Sickman JO; Meixner T;
Johnson DW; Neitlich P (2003). Ecological effects of nitrogen deposition in the western United States.
Bioscience, 53, 404-420.
307 Davies L; Bates JW; Bell JNB; James PW; Purvis OW (2007). Diversity and sensitivity of epiphytes to
oxides of nitrogen in London. Environ Pollut, 146, 299-310.
308 Farmer AM; Bates JW; Bell JNB (1992). Ecophysiological effects of acid rain on bryophytes and lichens. In:
Bates JW; Farmer AM (Eds.), Bryophytes and lichens in a changing environment. Oxford, UK: Claredon Press;
and Fields RF. (1988). Physiological responses of lichens to air pollutant fumigations. In: Nash TH III; Wirth V
(Eds.), Lichens, bryophytes and air quality. Volume 30; Bibl. Lichenol. (pp 175-200). Berlin/ Stuttgart,
Germany: Cramer Publisher.
309 Scott MG; Hutchinson TC; Feth MJ (1989). A comparison of the effects on Canadian boreal forest lichens of
nitric and sulfuric acids as sources of rain acidity. NewPhytol, 111, 663-671.; Scott MG; Hutchinson TC; Feth
MJ. (1989b). Contrasting responses of lichens and Vaccinium angustifolium to long-term acidification of a
boreal forest ecosystem. CanJBot, 67, 579-588.
310 U.S. EPA (2008). Nitrogen Dioxide/Sulfur Dioxide Secondary NAAQS Review: Integrated Science
Assessment (ISA) (Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F.
311 Van Sickle J; Baker JP; Simonin HA; Baldigo BP; Kretser WA; Sharpe WE (1996). Episodic acidification of
small streams in the northeastern United States: Fish mortality in field bioassays. EcolAppl, 6, 408-421.
312 U.S. EPA (1999). The Benefits and Costs of the Clean Air Act, 1990-2010. Prepared for U.S. Congress by
U.S. EPA, Office of Air and Radiation, Office of Policy Analysis and Review, Washington, DC, November;
EPA report no. EPA410-R-99-001.
313 U.S. EPA (2006). Air Quality Criteria Document for Ozone and Related Photochemical Oxidants (Final).
U.S. EPA, Washington, DC, EPA/600/R-05/004aF-cF, 2006.
314 Winner WE; Atkinson CJ (1986). Absorption of air pollution by plants, and consequences for growth. Trends
in Ecology and Evolution 1:15-18.
315 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
316 Tingey DT; Taylor GE (1982). Variation in plant response to ozone: a conceptual model of physiological
events. In: Effects of Gaseous Air Pollution in Agriculture and Horticulture (Unsworth, M.H., Omrod, D.P.,
eds.) London, UK: Butterworth Scientific, pp.113-138.
3-126
-------
317 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
318 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
319 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
320 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
321 Ollinger SV; Aber JD; Reich, PB (1997). Simulating ozone effects on forest productivity: interactions
between leaf canopy and stand level processes. Ecological Applications 7:1237'-1251.
322 Winner WE (1994). Mechanistic analysis of plant responses to air pollution. Ecological Applications,
4(4):651-661.
323 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
324 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
325 Fox S; Mickler R A (1996). Impact of Air Pollutants on Southern Pine Forests. Springer-Verlag, NY, Ecol.
Studies.Vol 118,513pp.
326 De Steiguer; Pye J ; Love C (1990). Air Pollution Damage to U.S. Forests. Journal of Forestry, Vol 88 (8)
pp. 17-22.
327 Pye JM (1988). Impact of ozone on the growth and yield of trees: A review. Journal of Environmental
Quality 17 pp.347-360.
328 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
329 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
330 McBride JR; Miller PR: Laven RD (1985). Effects of oxidant air pollutants on forest succession in the mixed
conifer forest type of southern California. In: Air Pollutants Effects On Forest Ecosystems, Symposium
Proceedings, St. P, 157-167.
331 Miller PR: Taylor OC: Wilhour RG (1982). Oxidant air pollution effects on a western coniferous forest
ecosystem. Corvallis, OR: U.S. Environmental Protection Agency, Environmental Research Laboratory.
EPA600-D-82-276.
332 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
3-127
-------
333 Kopp RJ; Vaughn WJ; HazillaM; Carson R (1985). Implications of environmental policy for U.S.
agriculture: the case of ambient ozone standards. J. Environ. Manage. 20:321-331.
334 Adams R M; Hamilton S A: McCarl B A (1986). The benefits of pollution control: the case of ozone and
U.S. agriculture. Am. J. Agric. Econ. 34:3-19.
335 Adams R M: Glyer J D; Johnson S L; McCarl BA (1989). A reassessment of the economic effects of ozone
on U.S. agriculture. JAPCA 39:960-968.
336 Abt Associates, Inc (1995). Urban ornamental plants: sensitivity to ozone and potential economic losses.
U.S. EPA, Office of Air Quality Planning and Standards, Research Triangle Park. Under contract to RADIAN
Corporation, contract no. 68-D3-0033, WA no. 6. pp. 9-10.
337 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
338 Grulke NE (2003). The physiological basis of ozone injury assessment attributes in Sierran conifers. In:
Bytnerowicz, A., M.J. Arbaugh, and R. Alonso, eds. Ozone air pollution in the Sierra Nevada: Distribution and
effects on forests. New York, NY: Elsevier Science, Ltd. pp. 55-81.
339 White D: Kimerling AJ; Overton WS (1992). Cartographic and geometric component of a global sampling
design for environmental monitoring. Cartogr. Geograph. Info. Sys. 19:5-22.
340 Smith G; Coulston J: Jepsen E; Prichard T (2003). A national ozone biomonitoring program—results from
field surveys of ozone sensitive plants in Northeastern forests (1994-2000). Environ. Monit. Assess. 87:271-291.
341 Coulston JW; Riitters KH; Smith GC (2004). A preliminary assessment of the Montreal process indicators of
air pollution for the United States. Environ. Monit. Assess. 95:57-74.
342U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
343 White D: Kimerling AJ; Overton WS (1992). Cartographic and geometric component of a global sampling
design for environmental monitoring. Cartogr. Geograph. Info. Sys. 19:5-22.
344 Smith G: Coulston J: Jepsen E: Prichard T (2003). A national ozone biomonitoring program-results from
field surveys of ozone sensitive plants in Northeastern forests (1994-2000). Environ. Monit. Assess. 87:271-291.
345 Coulston JW: Riitters KH: Smith GC (2004). A preliminary assessment of the Montreal process indicators
of air pollution for the United States. Environ. Monit. Assess. 95:57-74.
346 Smith, GJ Coulston E: Jepsen: Prichard T (2003). A national ozone biomonitoring program—results from
field surveys of ozone senstive plans in Northeastern forests (1994-2000). Environ. Monit. Assess. 87:271-291.
347 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF.
348 US EPA (2007). Review of the National Ambient Air Quality Standards for Ozone: Policy assessment of
scientific and technical information. Office of Air Quality Planning and Standards staff paper. EPA-452/R-07-
003.
3-128
-------
349 Chappelka AH: Samuelson LJ (1998). Ambient ozone effects on forest trees of the eastern United States: a
review. NewPhytologist 139, 91-108.
350 Prasad A M; Iverson LR (2003). Little's range and FIA importance value database for 135 eastern US tree
species. Northeastern Research Station, USDA Forest Service, Delaware, Ohio:
http://www.fs.fed.us/ne/delaware/4153/global/littlefia/index.html,
351 Heck WW; Cowling EB (1997). The need for a long term cumulative secondary ozone standard-an
ecological perspective. Environmental Management, January, 23-33.
352 Lefohn, A.S, Runeckles, V.C., 1987. Establishing a standard to protect vegetation - ozone exposure/dose
considerations. Atmospheric Environment 21, 561-568.
353 US EPA (2005). Air Quality Designations and Classifications for the Fine Particles (PM2.5) National Ambient
Air Quality Standards, 70 FR 943, Jan 5. 2005. This document is also available on the web at:
http://www.epa.gov/pmdesignations/
354 US EPA (1999). Regional Haze Regulations, 64 FR 35714, July 1, 1999.
3-129
-------
4 Quantified Health Impacts Analysis
Ship emissions are responsible for a large number of adverse human health and
environmental impacts, especially in densely populated coastal areas. As demonstrated in
Chapters 2 and 3, ships that would operate in the proposed EGA generate emissions of NOx (a
precursor to ozone formation and secondarily-formed PMz.s), SOx (a precursor to
secondarily-formed PMz.s) and directly-emitted PM2.5. These pollutants contribute to ambient
concentrations of PM2.5 and ozone that cause harm to human health and the environment.
This chapter presents the U.S.-related health impacts associated with emissions from ships,
both in terms of the expected contribution of overall ship emissions to adverse health impacts
on land and the reductions in adverse health impacts that can be expected to occur from the
adoption of the proposed EGA. Reductions in ambient PM2.s and ozone that will result from
the proposed EGA are expected to benefit human health in the form of avoided premature
deaths and other serious human health effects, as well as other important public health and
environmental effects.
The most conservative premature mortality estimates (Pope et al., 2002 for PM2.s and
Bell et al., 2004 for ozone)ll2 suggest that implementation of the proposed EGA would reduce
approximately 3,500 premature mortalities in 2020. The upper end of the premature mortality
estimates (Laden et al., 2006 for PM2.s and Levy et al., 2005 for ozone)3'4 suggest that
implementation of the proposed EGA would increase the estimate of avoided premature
mortalities to approximately 8,100 in 2020. Thus, even taking the most conservative
premature mortality assumptions, the health impacts of the proposed EGA are clearly
substantial.
The health impacts modeling presented in this Chapter is based on peer-reviewed
studies of air quality and health and welfare effects associated with improvements in air
quality. The health impact estimates for the proposed EGA are based on an analytical
structure and sequence consistent with health impacts analyses performed by the United
States Environmental Protection Agency (US EPA) for its recent analyses in support of the
final Ozone National Ambient Air Quality Standard (NAAQS) and the final PM NAAQS as
well as all of its recent mobile source emission control programs.5'6 For a more detailed
discussion of the principles of health impacts analysis used here, we refer the reader to those
NAAQS documents.
Benefits estimated for this analysis were generated using the Environmental Benefits
Mapping and Analysis Program (BenMAP). BenMAP is a computer program developed by
the US EPA that integrates a number of modeling elements (e.g., interpolation functions,
population projections, health impact functions, valuation functions, analysis and pooling
methods) to translate modeled air concentration estimates into health effect incidence
estimates. Interested parties may wish to consult the webpage
http://www.epa.gov/ttn/ecas/benmodels.html for more information.
The general health impacts analysis framework is as follows:
4-1
-------
• Using baseline and control emissions inventories for the emission species expected to
affect ambient air quality (NOx, SOz, and PlV^.s; see Chapter 2), we carried out
sophisticated photochemical air quality models to estimate baseline and control ambient
concentrations of PM and ozone for 2020 (see Chapter 3).
• The estimated changes in ambient concentrations are then combined with monitoring data
to estimate population-level potential exposures to changes in ambient concentrations for
use in estimating health effects (see Chapter 3). Modeled changes in ambient data are also
used to estimate changes in visibility.
• Changes in population exposure to ambient air pollution are used along with impact
functionsA to generate estimated reductions in the incidence of health effects. Because
these estimates contain uncertainty, we characterize the health impact estimates
probabilistically when appropriate information is available.
Table 4-1 presents the human health impacts we are able to quantify using this
methodology. However, the full complement of human health and welfare effects associated
with PM and ozone remains unquantified because of current limitations in methods or
available data. We have not quantified a number of known or suspected health effects linked
with ozone and PM for which appropriate health impact functions are not available or which
do not provide easily interpretable outcomes (i.e., changes in heart rate variability).
Additionally, we are unable to quantify a number of known environmental (welfare) effects,
including reduced acid and particulate deposition damage to cultural monuments and other
materials, and environmental benefits due to reductions of impacts of eutrophication in coastal
areas. These unquantified welfare effects are also listed in Table 4-1. Both the unquantified
and quantified environmental benefits of the proposed EGA are described further in Chapter
5. In sum, the health benefits quantified in this Chapter are likely underestimates of the total
benefits attributable to the implementation of the proposed EGA.
A The term "impact function" as used here refers to the combination of a) an effect estimate obtained from the
epidemiological literature, b) the baseline incidence estimate for the health effect of interest in the modeled
population, c) the size of that modeled population, and d) the change in the ambient air pollution metric of
interest. These elements are combined in the impact function to generate estimates of changes in incidence of
the health effect. The impact function is distinct from the concentration-response (C-R) function, which strictly
refers to the estimated equation from the epidemiological study relating incidence of the health effect and
ambient pollution. We refer to the specific value of the relative risk or estimated coefficients in the
epidemiological study as the "effect estimate." In referencing the functions used to generate changes in
incidence of health effects for this analysis, we use the term "impact function" rather than C-R function because
"impact function" includes all key input parameters used in the incidence calculation.
4-2
-------
Table 4-1 Human Health and Welfare Effects of Pollutants Affected by the Proposed ECA
POLLUTANT/
EFFECT
PM/Healthb
PM/Welfare
Ozone/Health6
Ozone/Welfare
Nitrogen
Deposition/
Welfare
NOx/Health
QUANTIFIED ESTIMATED
Premature mortality based on both
cohort study estimates c'
Bronchitis: chronic and acute
Hospital admissions: respiratory and
cardiovascular
Emergency room visits for asthma
Nonfatal heart attacks (myocardial
infarction)
Lower and upper respiratory illness
Minor restricted-activity days
Work loss days
Asthma exacerbations (asthmatic
population)
Respiratory symptoms (asthmatic
population)
Infant mortality
Premature mortality: short-term
exposures
Hospital admissions: respiratory
Emergency room visits for asthma
Minor restricted-activity days
School loss days
Asthma attacks
Acute respiratory symptoms
Decreased outdoor worker
productivity
Forest biomass
UNQUANTIFIED EFFECTS - CHANGES IN:
Subchronic bronchitis cases
Low birth weight
Pulmonary function
Chronic respiratory diseases other than chronic bronchitis
Nonasthma respiratory emergency room visits
Value of recreational and residential visibility
Household soiling
Cardiovascular emergency room visits
Chronic respiratory damagef
Premature aging of the lungsf
Nonasthma respiratory emergency room visits
Yields for commercial crops
Yields for commercial forests and noncommercial crops
Damage to urban ornamental plants
Recreational demand from damaged forest aesthetics
Ecosystem functions
Commercial forests due to acidic sulfate and nitrate
deposition
Commercial freshwater fishing due to acidic deposition
Recreation in terrestrial ecosystems due to acidic
deposition
Commercial fishing, agriculture, and forests due to nitrogen
deposition
Recreation in estuarine ecosystems due to nitrogen
deposition
Ecosystem functions
Passive fertilization
Lung irritation
Lowered resistance to respiratory infection
Hospital admissions for respiratory and cardiac diseases
a Primary quantified effects are those included in this analysis.
In addition to primary endpoints, there are a number of biological responses that have been associated with PM
and ozone health effects including morphological changes and altered host defense mechanisms. The public
health impact of these biological responses may be partly represented by our quantified endpoints.
4-3
-------
c Cohort estimates are designed to examine the effects of long term exposures to ambient pollution, but relative
risk estimates may also incorporate some effects due to shorter term exposures (see Kunzli, 2001 for a discussion
of this issue).
While some of the effects of short-term exposure are likely to be captured by the cohort estimates, there may be
additional premature mortality from short-term PM exposure not captured in the cohort estimates included in the
primary analysis.
e The public health impact of biological responses such as increased airway responsiveness to stimuli,
inflammation in the lung, acute inflammation and respiratory cell damage, and increased susceptibility to
respiratory infection are likely partially represented by our quantified endpoints.
f The public health impact of effects such as chronic respiratory damage and premature aging of the lungs may
be partially represented by quantified endpoints such as hospital admissions or premature mortality, but a
number of other related health impacts, such as doctor visits and decreased athletic performance, remain
unquantified.
4.1 Health Impacts Analysis Results for the Proposed EGA
Tables 4.1-1 and 4.1-2 present the annual PM2.s and ozone health impacts for two
scenarios. The first scenario assesses the annual health impact of ship emissions if current
levels of per-unit emissions are assumed to occur in 2020. The second scenario assesses the
annual reduction of ship-related health impacts if the EGA standards are in place in 2020.
Table 4.1-1. Estimated PM2.5-Related Health Impacts Associated with Ship Emissions3
Health Effect
Premature Mortality15
Adult, age 30+, ACS Cohort Study (Pope et al, 2002)
Adult, age 25+, Six-Cities Study (Laden et al., 2006)
Infant, age <1 year (Woodruff et al., 1997)
Chronic bronchitis (adult, age 26 and over)
Non-fatal myocardial infarction (adult, age 18 and
over)
Hospital admissions - respiratory (all ages)c
Hospital admissions - cardiovascular (adults, age >18)d
Emergency room visits for asthma (age 18 years and
younger)
Acute bronchitis, (children, age 8-12)
Lower respiratory symptoms (children, age 7-14)
Upper respiratory symptoms (asthmatic children, age
9-18)
Asthma exacerbation (asthmatic children, age 6-18)
2020 Annual Ship-Related
Incidence
(5th% - 95th%ile)
4,300
(1,700-7,000)
9,800
(5,400-14,000)
16
(0-42)
4,300
(810-7,800)
8,900
(4,900-13,000)
990
(490-1,500)
2,100
(1,500-2,400)
2,500
(1,500-3,500)
11,000
(0-22,000)
84,000
(40,000-130,000)
62,000
(19,000-100,000)
79,000
(8,600-220,000)
2020 Annual Reduction in
Ship-Related Incidence w/
200nm EGA
(5*% - 95*%ile)
3,400
(1,300-5,500)
7,800
(4,300-11,000)
12
(0 - 33)
3,300
(620 - 6,000)
7,200
(3,900-10,000)
780
(380-1,200)
1,600
(1,200- 1,900)
1,900
(1,100-2,700)
8,500
(0-17,000)
66,000
(32,000-99,000)
48,000
(15,000-82,000)
62,000
(6,700- 180,000)
4-4
-------
Work loss days
Minor restricted activity days (adults age 18-65)
580,000
(510,000-650,00)
3,400,000
(2,900,000-4,000,000)
460,000
(400,000 - 520,000)
2,700,000
(2,300,000-3,100,000)
Notes:
a Incidence is rounded to two significant digits. Estimates represent incidence within the 48 contiguous United States.
b PM-related adult mortality based upon the American Cancer Society (ACS) Cohort Study (Pope et al, 2002) and the Six-
Cities Study (Laden et al., 2006). Note that these are two alternative estimates of adult mortality and should not be summed.
PM-related infant mortality based upon a study by Woodruff, Grillo, and Schoendorf, (1997).
c Respiratory hospital admissions for PM include admissions for chronic obstructive pulmonary disease (COPD), pneumonia
and asthma.
d Cardiovascular hospital admissions for PM include total cardiovascular and subcategories for ischemic heart disease,
dysrhythmias, and heart failure.
Table 4.1-2. Estimated Ozone-Related Health Impacts Associated with Ship Emissions"
Health Effect
Premature Mortality, All agesb
Multi-City Analyses
Bell et al (2004) - Non-accidental
Huang et al (2005) - Cardiopulmonary
Schwartz, (2005) - Non-accidental
Meta-analyses:
Bell et al (2005) - All cause
Ito et al (2005) - Non-accidental
Levy et al (2005) - All cause
Hospital admissions- respiratory causes (adult,
65 and older)0
Hospital admissions -respiratory causes
(children, under 2)
Emergency room visit for asthma (all ages)
Minor restricted activity days (adults, age 18-
65)
School absence days
2020 Annual Ship-Related
Incidence
(5*% - 95*%ile)
370
(160-570)
620
(290-940)
560
(240-890)
1,200
(660-1,700)
1,600
(1,100-2,200)
1,700
(1,200-2,100)
2,900
(400-4,800)
2,400
(1,200-3,500)
1,300
(0-3,500)
2,300,000
(1,100,000-3,400,000)
810,000
(360,000-1,100,000)
2020 Annual Reduction in Ship-
Related Incidence w/ 200nm
EGA
(5th% - 95th%ile)
61
(23 - 98)
100
(43 - 160)
93
(34 - 150)
200
(100 - 290)
270
(170-370)
280
(200 - 360)
470
(46 - 830)
380
(180 - 590)
210
(0 - 550)
360,000
(160,000-570,000)
130,000
(51,000- 190,000)
' Incidence is rounded to two significant digits. Estimates represent incidence within the 48 contiguous United States.
b Estimates of ozone-related premature mortality are based upon incidence estimates derived from several alternative studies:
Bell et al. (2004); Huang et al. (2005); Schwartz (2005) ; Bell et al. (2005); Ito et al. (2005); Levy et al. (2005). The
estimates of ozone-related premature mortality should therefore not be summed.
c Respiratory hospital admissions for ozone include admissions for all respiratory causes and subcategories for COPD and
pneumonia.
As can be seen in Tables 4.1-1 and 4.1-2, ship emissions contribute to large numbers
of adverse health impacts within the U.S. By designating an EGA, we estimate that by 2020,
4-5
-------
emission reductions will result in major reductions in health impacts, especially those
associated with PM exposure. For example, we estimate that in 2020, ships emitting at their
current performance would be responsible for approximately 4,300 - 9,800 cases of
premature mortality in adults (range based on the health impact function used - Pope et al.,
2002 and Laden et al., 2006, respectively). Improving ship emissions to EGA standards will
avoid between 3,400 - 7,800 premature deaths in 2020, a reduction of approximately 79%.
We also estimate that ships are responsible for a large number of PlV^.s-related
morbidity impacts. For example, we estimate that in 2020, ships emitting at their current
performance would be responsible for approximately 4,300 cases of chronic bronchitis, 8,900
non-fatal heart attacks, 5,600 hospital admissions and emergency room visits, 580,000 days of
work lost, and 3,400,000 days of restricted physical activity. Improving ship emissions to
EGA standards will result in the avoidance of 3,300 cases of chronic bronchitis, 7,200 non-
fatal heart attacks, 4,400 hospital admissions and emergency room visits, 460,000 days of
work lost, and 2,700,000 days of restricted physical activity. Again, improving to EGA
standards will reduce the incidence of PlV^.s-related non-fatal health impacts associated with
ships by approximately 78%.
Similarly, ship emissions contribute to adverse health impacts associated with ozone
exposure. For example, we estimate that in 2020, ships emitting at their current performance
would be responsible for approximately 370 - 1,700 cases of premature mortality, depending
on the health impact function, 6,600 hospital admissions and emergency room visits, 810,000
days of school absence, and 2,300,000 day of restricted physical activity. Improving to EGA
standards will avoid between 61 - 280 premature deaths in 2020. Furthermore, it will result
in the avoidance of 1,100 hospital admissions and emergency room visits, 130,000 days of
school absence, and 360,000 days of restricted physical activity.
It is clear that the avoided health impacts associated with the proposed EGA are
substantial. Implementation of a North American EGA would significantly improve human
health, both in terms of reduced premature mortality and avoided morbidity effects.
4.2 Methodology
4.2.1 Human Health Impact Functions
Health impact functions measure the change in a health endpoint of interest, such as
hospital admissions, for a given change in ambient ozone or PM concentration. Health impact
functions are derived from primary epidemiology studies, meta-analyses of multiple
epidemiology studies, or expert elicitations. A standard health impact function has four
components: 1) an effect estimate from a particular study; 2) a baseline incidence rate for the
health effect (obtained from either the epidemiology study or a source of public health
statistics such as the Centers for Disease Control); 3) the size of the potentially affected
population; and 4) the estimated change in the relevant ozone or PM summary measures.
A typical health impact function might look like:
4-6
-------
where y0 is the baseline incidence (the product of the baseline incidence rate times the
potentially affected population), p is the effect estimate, and Ax is the estimated change in the
summary pollutant measure. There are other functional forms, but the basic elements remain
the same. The following subsections describe the sources for each of the first three elements:
size of the potentially affected populations; PM2.s and ozone effect estimates; and baseline
incidence rates. Section 4.2.2 describes the ozone and PM air quality inputs to the health
impact functions.
4.2.1.1 Potentially Affected Populations
The starting point for estimating the size of potentially affected populations is the
2000 U.S. Census block level dataset.7 Benefits Modeling and Analysis Program (BenMAP)
incorporates 250 age/gender/race categories to match specific populations potentially affected
by ozone and other air pollutants. The software constructs specific populations matching the
populations in each epidemiological study by accessing the appropriate age-specific
populations from the overall population database. BenMAP projects populations to 2020
using growth factors based on economic projections.8
4.2.1.2 Effect Estimate Sources
The most significant quantifiable benefits of reducing ambient concentrations of ozone
and PM are attributable to reductions in human health risks. EPA's Ozone and PM Criteria
,9,10
11,12
Documents ' and the World Health Organization's 2003 and 2004llllz reports outline
numerous human health effects known or suspected to be linked to exposure to ambient ozone
and PM. US EPA recently evaluated the ozone and PM literature for use in the benefits
analysis for the final 2008 Ozone NAAQS and final 2006 PM NAAQS analyses. We use the
same literature in this analysis.
It is important to note that we are unable to separately quantify all of the possible PM
and ozone health effects that have been reported in the literature for three reasons: (1) the
possibility of double counting (such as hospital admissions for specific respiratory diseases
versus hospital admissions for all or a sub-set of respiratory diseases); (2) uncertainties in
applying effect relationships that are based on clinical studies to the potentially affected
population; or (3) the lack of an established concentration-response (CR) relationship. Table
4-1 lists the possible human health and welfare effects of pollutants affected by the proposed
EGA. Table 4.2-1 lists the health endpoints included in this analysis.
Table 4.2-1 Ozone- and PM-Related Health Endpoints
ENDPOINT
POLLUTANT
STUDY
STUDY POPULA TION
Premature Mortality
Premature mortality
- daily time series
03
Bell et al (2004) (NMMAPS study) 13 - Non-
accidental
Huang et al (2005) u - Cardiopulmonary
Schwartz (2005) 15 - Non-accidental
Meta-analyses:
Bell etal (2005) 16 - All cause
Ito et al (2005) ll - Non-accidental
All ages
4-7
-------
ENDPOINT
Premature mortality
— cohort study, all-
cause
Premature mortality
— all-cause
POLLUTANT
PM2.5
PM2.5
STUDY
Levy et al (2005) 18 - All cause
Popeetal. (2002) 19
Laden et al. (2006) 20
Woodruff etal. (1997)21
STUDY POPULA TION
>29 years
>25 years
Infant (<1 year)
Chronic Illness
Chronic bronchitis
Nonfatal heart
attacks
PM2.5
PM2.5
Abbey et al. (1995)22
Peters etal. (200 1)23
>26 years
Adults (> 18 years)
Hospital Admissions
Respiratory
03
PM2.5
PM2.5
PM2.5
PM2.5
Pooled estimate:
Schwartz (1995) - ICD 460-519 (all resp)24
Schwartz (1994a; 1994b) - ICD 480-486
(pneumonia)25'26
Moolgavkar et al. (1997) - ICD 480-487
(pneumonia) 27
Schwartz (1994b) - ICD 491-492, 494-496
(COPD)
Moolgavkar et al. (1997) - ICD 490-496
(COPD)
Burnett etal. (200 1)28
Pooled estimate:
Moolgavkar (2003)— ICD 490-496 (COPD)29
Ito (2003)— ICD 490-496 (COPD)30
Moolgavkar (2000)— ICD 490-496 (COPD)31
Ito (2003)— ICD 480-486 (pneumonia)
Sheppard (2003)— ICD 493 (asthma)32
>64 years
<2 years
>64 years
20-64 years
>64 years
<65 years
4-8
-------
ENDPOINT
Cardiovascular
Asthma-related ER
visits
Asthma-related ER
visits (con't)
POLLUTANT
PM2.5
PM2.5
03
PM2.5
STUDY
Pooled estimate:
Moolgavkar (2003)— ICD 390-429 (all
cardiovascular)
Ito (2003)— ICD 410-414, 427-428 (ischemic
heart disease, dysrhythmia, heart failure)
Moolgavkar (2000)— ICD 390-429 (all
cardiovascular)
Pooled estimate:
Jaffeetal(2003)33
Peel etal (2005) 34
Wilson etal (2005) 35
Norrisetal. (1999)36
STUDY POPULA TION
>64 years
20-64 years
5-34 years
All ages
All ages
0-1 8 years
Other Health Endpoints
Acute bronchitis
Upper respiratory
symptoms
Lower respiratory
symptoms
Asthma
exacerbations
Work loss days
School absence
days
Minor Restricted
Activity Days
(MRADs)
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
03
03
PM2.5
Dockeryetal. (1996)37
Pope etal. (1991) 38
Schwartz and Neas (2000) 39
Pooled estimate:
Ostro et al. (200 1)40 (cough, wheeze and
shortness of breath)
Vedaletal. (1998)41 (cough)
Ostro (1987)42
Pooled estimate:
Gilliland et al. (200 1)43
Chen etal. (2000) 44
Ostro and Rothschild (1989) 45
Ostro and Rothschild (1989)
8- 12 years
Asthmatics, 9-11
years
7- 14 years
6- 18 years3
18-65 years
5-17yearsb
18-65 years
18-65 years
a The original study populations were 8 to 13 for the Ostro et al. (2001) study and 6 to 13 for the Vedal et al.
(1998) study. Based on advice from the Science Advisory Board Health Effects Subcommittee (SAB-HES),
we extended the applied population to 6 to 18, reflecting the common biological basis for the effect in
children in the broader age group. See: U.S. Science Advisory Board. 2004. Advisory Plans for Health
Effects Analysis in the Analytical Plan for EPA's Second Prospective Analysis -Benefits and Costs of the
Clean Air Act, 1990—2020. EPA-SAB-COUNCIL-ADV-04-004. See also National Research Council
(NRC). 2002. Estimating the Public Health Benefits of Proposed Air Pollution Regulations. Washington,
DC: The National Academies Press.
b Gilliland et al. (2001) studied children aged 9 and 10. Chen et al. (2000) studied children 6 to 11. Based on
recent advice from the National Research Council and the EPA SAB-HES, we have calculated reductions in
school absences for all school-aged children based on the biological similarity between children aged 5 to 17.
In selecting epidemiological studies as sources of effect estimates, we applied several
criteria to develop a set of studies that is likely to provide the best estimates of impacts in the
U.S. To account for the potential impacts of different health care systems or underlying
health status of populations, we give preference to U.S. studies over non-U.S. studies. In
addition, due to the potential for confounding by co-pollutants, we give preference to effect
estimates from models including both ozone and PM over effect estimates from single-
4-9
-------
pollutant models.46'47
4.2.1.2.1 PM25-RelatedHealth Impact Functions
PM2.s-Related Adult Premature Mortality
Both long- and short-term exposures to ambient levels of air pollution have been
associated with increased risk of premature mortality. The size of the mortality risk estimates
from epidemiological studies, the serious nature of the effect itself, and the high monetary
value ascribed to prolonging life make mortality risk reduction the most significant health
endpoint quantified in this analysis.
Although a number of uncertainties remain to be addressed by continued research
(NRC, 1998),4 a substantial body of published scientific literature documents the correlation
between elevated PM concentrations and increased mortality rates (US EPA, 2004).49 Time-
series methods have been used to relate short-term (often day-to-day) changes in PM
concentrations and changes in daily mortality rates up to several days after a period of
elevated PM concentrations. Cohort methods have been used to examine the potential
relationship between community-level PM exposures over multiple years (i.e., long-term
exposures) and community-level annual mortality rates. Researchers have found statistically
significant associations between PM and premature mortality using both types of studies. In
general, the risk estimates based on the cohort studies are larger than those derived from time-
series studies. Cohort analyses are thought to better capture the full public health impact of
exposure to air pollution over time, because they capture the effects of long-term exposures
and possibly some component of short-term exposures (Kunzli et al., 2001; NRC, 2002).50'51
This section discusses some of the issues surrounding the estimation of premature mortality.
Over a dozen studies have found significant associations between various measures of
long-term exposure to PM and elevated rates of annual mortality, beginning with Lave and
Seskin (1977).52 Most of the published studies found positive (but not always statistically
significant) associations with available PM indices such as total suspended particles (TSP).
However, exploration of alternative model specifications sometimes raised questions about
causal relationships (e.g., Lipfert, Morris, and Wyzga [1989]).53 These early "ecological
cross-sectional" studies (e.g., Lave and Seskin [1977]; Ozkaynak and Thurston [1987]54)
were criticized for a number of methodological limitations, particularly for inadequate control
at the individual level for variables that are potentially important in causing mortality, such as
wealth, smoking, and diet. Over the last 10 years, several studies using "prospective cohort"
designs have been published that appear to be consistent with the earlier body of literature.
These new "prospective cohort" studies reflect a significant improvement over the earlier
work because they include individual-level information with respect to health status and
residence. The most extensive analyses have been based on data from two prospective cohort
groups, often referred to as the Harvard "Six-Cities Study" (Dockery et al., 1993;55 Laden et
al, 2006) and the "American Cancer Society or ACS study" (Pope et al., 1995;56 Pope et al,
2002; Pope et al, 200457); these studies have found consistent relationships between fine
particle indicators and premature mortality across multiple locations in the United States. A
third major data set comes from the California-based 7th Day Adventist Study (e.g., Abbey et
al., 1999),58 which reported associations between long-term PM exposure and mortality in
4-10
-------
men. Results from this cohort, however, have been inconsistent, and the air quality results are
not geographically representative of most of the United States, and the lifestyle of the
population is not reflective of much of the U.S. population. Analysis is also available for a
cohort of adult male veterans diagnosed with hypertension has been examined (Lipfert et al.,
2000; Lipfert et al, 2003, 2006). '60'61 The characteristics of this group differ from the
cohorts in the Six-Cities, ACS, and 7th Day Adventist studies with respect to income, race,
health status, and smoking status. Unlike previous long-term analyses, this study found some
associations between mortality and ozone but found inconsistent results for PM indicators.
Because of the selective nature of the population in the veteran's cohort, we have chosen not
to include any effect estimates from the Lipfert et al. (2000) study in our benefits
assessment.B
Given their consistent results and broad geographic coverage, and importance in
informing the NAAQS development process, the Six-Cities and ACS data have been
particularly important in benefits analyses. The credibility of these two studies is further
enhanced by the fact that the initial published studies (Pope et al, 1995 and Dockery et al
1993) were subject to extensive reexamination and reanalysis by an independent team of
scientific experts commissioned by the Health Effects Institute (HEI) (Krewski et al., 2000).62
The final results of the reanalysis were then independently peer reviewed by a Special Panel
of the HEI Health Review Committee. The results of these reanalyses confirmed and
expanded those of the original investigators. While the HEI reexamination lends credibility to
the original studies, it also highlights sensitivities concerning the relative impact of various
pollutants, such as S02, the potential role of education in mediating the association between
pollution and mortality, and the influence of spatial correlation modeling.
Further confirmation and extension of the findings of the 1993 Six City Study and the
1995 ACS study were recently completed using more recent air quality and a longer follow-
up period for the ACS cohort was recently published (Pope et al, 2002, 2004; Laden et al,
2006). The follow up to the Harvard Six City Study both confirmed the effect size from the
first analysis and provided additional confirmation that reductions in PM2.5 are likely to result
in reductions in the risk of premature death. This additional evidence stems from the
observed reductions in PM2.5 in each city during the extended follow-up period. Laden et al.
(2006) found that mortality rates consistently went down at a rate proportionate to the
observed reductions in PM2.s.
B US EPA recognizes that the ACS cohort also is not representative of the demographic mix in the general
population. The ACS cohort is almost entirely white and has higher income and education levels relative to the
general population. US EPA's approach to this problem is to match populations based on the potential for
demographic characteristics to modify the effect of air pollution on mortality risk. Thus, for the various ACS-
based models, we are careful to apply the effect estimate only to ages matching those in the original studies,
because age has a potentially large modifying impact on the effect estimate, especially when younger individuals
are excluded from the study population. For the Lipfert analysis, the applied population should be limited to that
matching the sample used in the analysis. This sample was all male, veterans, and diagnosed hypertensive.
There are also a number of differences between the composition of the sample and the general population,
including a higher percentage of African Americans (35%) and a much higher percentage of smokers (81%
former smokers, 57% current smokers) than the general population (12% African American, 24% current
smokers).
4-11
-------
The extended analyses of the ACS cohort data (Pope et al, 2002, 2004) provides
additional refinements to the analysis of PM-related mortality by a) extending the follow-up
period for the ACS study subjects to 16 years, which triples the size of the mortality data set;
b) substantially increasing exposure data, including additional measurement of cohort
exposure to PM2.5 following implementation of the PM2.5 standard in 1999; c) controlling for
a variety of personal risk factors including occupational exposure and diet; and d) using
advanced statistical methods to evaluate specific issues that can adversely affect risk estimates
including the possibility of spatial autocorrelation of survival times in communities located
near each other.
For this analysis, we use the ACS study because it includes a large sample size and
longer exposure interval and covers more locations (e.g., 50 cities compared to the Six-Cities
Study) than other studies of its kind. The relative risks derived from the ACS study are based
on the average exposure to PM2.5, measured by the average of two PM2.5 measurements, over
the periods 1979-1983 and 1999-2000. In addition to relative risks for all-cause mortality,
the ACS study provides relative risks for cardiopulmonary, lung cancer, and all-other cause
mortality. Because of concerns regarding the statistical reliability of the "all-other" cause
mortality relative risk estimates, we calculate mortality impacts for this analysis using the all-
cause relative risk.
We also include a separate estimate based on the Six-cities study to complement the
estimate based on the ACS study. We use this specific estimate because it reflects the most
up-to-date science and reflects the weight that experts have placed on both the ACS and
Harvard Six-city studies (see the results of the PM mortality expert elicitation).63
Because of the differences in the study designs and populations considered in the ACS
and Harvard Six-cities studies, we do not pool the results of the studies and instead present a
range of estimates reflecting the two sources of impact estimates.
A number of additional analyses have been conducted on the ACS cohort data (Jerrett
et al., 2005;64 Krewski et al., 2005;^5 Pope et al., 2004). These studies have continued to find
a strong significant relationship between PM2.5 and mortality outcomes. Specifically, much of
the recent research has suggested a stronger relationship between cardiovascular mortality and
lung cancer mortality with PM2.5, and a less significant relationship between respiratory-
related mortality and PM2.5.
PM2.5-Related Infant Mortality
Recently published studies have strengthened the case for an association between PM
exposure and respiratory inflammation and infection leading to premature mortality in
children under 5 years of age. Specifically, the release of the WHO Global Burden of Disease
Study focusing on ambient air cites several recently published time-series studies relating
daily PM exposure to mortality in children The study by Belanger et al. (2003)66 also
corroborates findings linking PM exposure to increased respiratory inflammation and
infections in children. A study by Chay and Greenstone (2003)67 found that reductions in
TSP caused by the recession of 1981-1982 were related to reductions in infant mortality at the
county level. With regard to the cohort study conducted by Woodruff et al. (1997),68 we note
4-12
-------
several strengths of the study, including the use of a larger cohort drawn from a large number
of metropolitan areas and efforts to control for a variety of individual risk factors in infants
(e.g., maternal educational level, maternal ethnicity, parental marital status, and maternal
smoking status). Based on these findings, the US EPA estimates infant mortality using an
impact function developed from the Woodruff et al. (1997) study.
Chronic Bronchitis
Chronic bronchitis (CB) is characterized by mucus in the lungs and a persistent wet
cough for at least 3 months a year for several years in a row. CB affects an estimated 5% of
the U.S. population (American Lung Association, 1999).69 A limited number of studies have
estimated the impact of air pollution on new incidences of CB. Schwartz (1993)70 and Abbey
et al. (1995)71 provide evidence that long-term PM exposure gives rise to the development of
CB in the United States. Because the proposed EGA is expected to reduce primarily PM2.5,
this analysis uses only the Abbey et al. (1995) study, because it is the only study focusing on
the relationship between PM2.s and new incidences of CB.
Nonfatal Myocardial Infarctions (heart attacks)
Nonfatal heart attacks have been linked with short-term exposures to PM2.s in the
United States (Peters et al., 2001)72 and other countries (Poloniecki et al., 1997).73 We used a
recent study by Peters et al. (2001) as the basis for the impact function estimating the
relationship between PM2.5 and nonfatal heart attacks. A more recent study by Zanobetti and
Schwartz (2005)74 used a similar method to Peters et al. (2001), but focused on adults 65 and
older, and used PMio as the PM indicator. They found a significant relationship between
nonfatal heart attacks and PMio, although the magnitude of the effect was much lower than
Peters et al. This may reflect the use of PMio, the more limited age range, or the less precise
diagnosis of heart attack used in defining the outcome measure. Other studies, such as
Domenici et al. (2006),75 Samet et al. (2000),76 and Moolgavkar (2000),77 show a consistent
relationship between all cardiovascular hospital admissions, including those for nonfatal heart
attacks, and PM. Given the lasting impact of a heart attack on long-term health costs and
earnings, we provide a separate estimate for nonfatal heart attacks. The estimate used in the
analysis of the proposed EGA is based on the single available U.S. PM2.5 effect estimate from
Peters et al. (2001). The finding of a specific impact on heart attacks is consistent with
hospital admission and other studies showing relationships between fine particles and
cardiovascular effects both within and outside the United States. Several epidemiologic
studies (Liao et al., 1999; Gold et al., 2000; Magari et al., 2001)78'79'80 have shown that heart
rate variability (an indicator of how much the heart is able to speed up or slow down in
response to momentary stresses) is negatively related to PM levels. Heart rate variability is a
risk factor for heart attacks and other coronary heart diseases (Carthenon et al., 2002; Dekker
et al., 2000; Liao et al., 1997; Tsuji et al., 1996).81'82'83'84 As such, significant impacts of PM
on heart rate variability are consistent with an increased risk of heart attacks.
Hospital and Emergency Room Admissions
Because of the availability of detailed hospital admission and discharge records, there
is an extensive body of literature examining the relationship between hospital admissions and
4-13
-------
air pollution. Because of this, many of the hospital admission endpoints use pooled impact
functions based on the results of a number of studies. In addition, some studies have
examined the relationship between air pollution and emergency room (ER) visits. Since most
emergency room visits do not result in an admission to the hospital (the majority of people
going to the emergency room is treated and return home), we treat hospital admissions and
emergency room visits separately, taking account of the fraction of emergency room visits
that are admitted to the hospital.
The two main groups of hospital admissions estimated in this analysis are respiratory
admissions and cardiovascular admissions. There is not much evidence linking PM with
other types of hospital admissions. The only type of emergency room visits that have been
consistently linked to PM in the United States are asthma-related visits.
To estimate avoided incidences of PM2.s related cardiovascular hospital admissions in
populations aged 65 and older, we use effect estimates from studies by Moolgavkar (2003)85
and Ito (2003).86 However, only Moolgavkar (2000)87 provided a separate effect estimate for
populations 20 to 64.c Total cardiovascular hospital admissions are thus the sum of the
pooled estimates from Moolgavkar (2003) and Ito (2003) for populations over 65 and the
Moolgavkar (2000) based impacts for populations aged 20 to 64. Cardiovascular hospital
admissions include admissions for myocardial infarctions. To avoid double-counting benefits
from reductions in myocardial infarctions when applying the impact function for
cardiovascular hospital admissions, we first adjusted the baseline cardiovascular hospital
admissions to remove admissions for myocardial infarctions.
To estimate total avoided incidences of respiratory hospital admissions, we used
impact functions for several respiratory causes, including chronic obstructive pulmonary
disease (COPD), pneumonia, and asthma. As with cardiovascular admissions, additional
published studies show a statistically significant relationship between PMio and respiratory
hospital admissions. We used only those focusing on PM2.s. Both Moolgavkar (2000) and Ito
(2003) provide effect estimates for COPD in populations over 65, allowing us to pool the
impact functions for this group. Only Moolgavkar (2000) provides a separate effect estimate
for populations 20 to 64. Total COPD hospital admissions are thus the sum of the pooled
estimate for populations over 65 and the single study estimate for populations 20 to 64. Only
Ito (2003) estimated pneumonia and only for the population 65 and older. In addition,
Sheppard (2003) provided an effect estimate for asthma hospital admissions for populations
under age 65. Total avoided incidence of PM-related respiratory-related hospital admissions
is the sum of COPD, pneumonia, and asthma admissions.
c Note that the Moolgavkar (2000) study has not been updated to reflect the more stringent GAM convergence
criteria. However, given that no other estimates are available for this age group, we chose to use the existing
study. Updates have been provided for the 65 and older population, and showed little difference. Given the
very small (<5%) difference in the effect estimates for people 65 and older with cardiovascular hospital
admissions between the original and reanalyzed results, we do not expect the difference in the effect estimates
for the 20 to 64 population to differ significantly. As such, the choice to use the earlier, uncorrected analysis will
likely not introduce much bias.
4-14
-------
To estimate the effects of PM air pollution reductions on asthma-related ER visits, we
use the effect estimate from a study of children 18 and under by Norris et al. (1999).88 As
noted earlier, there is another study by Schwartz examining a broader age group (less than
65), but the Schwartz study focused on PMi0 rather than PM2.s. We selected the Norris et al.
(1999) effect estimate because it better matched the pollutant of interest. Because children
tend to have higher rates of hospitalization for asthma relative to adults under 65, we will
likely capture the majority of the impact of PM2.5 on asthma emergency room visits in
populations under 65, although there may still be significant impacts in the adult population
under 65.
Acute Health Events and Work Loss Days
As indicated in Table 4.2-1, in addition to mortality, chronic illness, and hospital
admissions, a number of acute health effects not requiring hospitalization are associated with
exposure to ambient levels of PM. The sources for the effect estimates used to quantify these
effects are described below.
Around four percent of U.S. children between the ages of 5 and 17 experience
episodes of acute bronchitis annually (American Lung Association, 2002).89 Acute bronchitis
is characterized by coughing, chest discomfort, slight fever, and extreme tiredness, lasting for
a number of days. According to the MedlinePlus medical encyclopedia,0 with the exception
of cough, most acute bronchitis symptoms abate within 7 to 10 days. Incidence of episodes of
acute bronchitis in children between the ages of 5 and 17 were estimated using an effect
estimate developed from Dockery et al. (1996).90
Incidences of lower respiratory symptoms (e.g., wheezing, deep cough) in children
aged 7 to 14 were estimated using an effect estimate from Schwartz and Neas (2000).91
Because asthmatics have greater sensitivity to stimuli (including air pollution),
children with asthma can be more susceptible to a variety of upper respiratory symptoms (e.g.,
runny or stuffy nose; wet cough; and burning, aching, or red eyes). Research on the effects of
air pollution on upper respiratory symptoms has thus focused on effects in asthmatics.
Incidences of upper respiratory symptoms in asthmatic children aged 9 to 11 are estimated
using an effect estimate developed from Pope et al. (1991).92
Health effects from air pollution can also result in missed days of work (either from
personal symptoms or from caring for a sick family member). Days of work lost due to PM2.5
were estimated using an effect estimate developed from Ostro (1987).93
Minor restricted activity days (MRADs) result when individuals reduce most usual
daily activities and replace them with less strenuous activities or rest, yet not to the point of
missing work or school. For example, a mechanic who would usually be doing physical work
D See http://www.nlm.nih.gov/medlineplus/ency/article/000124.htm, accessed January 2002.
4-15
-------
most of the day will instead spend the day at a desk doing paper and phone work because of
difficulty breathing or chest pain. The effect of PM2.s and ozone on MRAD was estimated
using an effect estimate derived from Ostro and Rothschild (1989).94
In analyzing the proposed EGA, we focused the estimation on asthma exacerbations
occurring in children and excluded adults from the calculation to avoid double counting.E
Asthma exacerbations occurring in adults are assumed to be captured in the general
population endpoints such as work loss days and MRADs. Consequently, if we had included
an adult-specific asthma exacerbation estimate, we would likely double-count incidence for
this endpoint. However, because the general population endpoints do not cover children (with
regard to asthmatic effects), an analysis focused specifically on asthma exacerbations for
children (6 to 18 years of age) could be conducted without concern for double-counting.
To characterize asthma exacerbations in children, we selected two studies (Ostro et al.,
2001; Vedal et al., 1998)95'96 that followed panels of asthmatic children. Ostro et al. (2001)
followed a group of 138 African-American children in Los Angeles for 13 weeks, recording
daily occurrences of respiratory symptoms associated with asthma exacerbations (e.g.,
shortness of breath, wheeze, and cough). This study found a statistically significant
association between PM2.5, measured as a 12-hour average, and the daily prevalence of
shortness of breath and wheeze endpoints. Although the association was not statistically
significant for cough, the results were still positive and close to significance; consequently,
we decided to include this endpoint, along with shortness of breath and wheeze, in generating
incidence estimates (see below). Vedal et al. (1998) followed a group of elementary school
children, including 74 asthmatics, located on the west coast of Vancouver Island for 18
months including measurements of daily peak expiratory flow (PEF) and the tracking of
respiratory symptoms (e.g., cough, phlegm, wheeze, chest tightness) through the use of daily
diaries. Association between PMio and respiratory symptoms for the asthmatic population
was only reported for two endpoints: cough and PEF. Because it is difficult to translate PEF
measures into clearly defined health endpoints that can be monetized, we only included the
cough-related effect estimate from this study in quantifying asthma exacerbations. We
employed the following pooling approach in combining estimates generated using effect
estimates from the two studies to produce a single asthma exacerbation incidence estimate.
First, we pooled the separate incidence estimates for shortness of breath, wheeze, and cough
generated using effect estimates from the Ostro et al. study, because each of these endpoints is
aimed at capturing the same overall endpoint (asthma exacerbations) and there could be
overlap in their predictions. The pooled estimate from the Ostro et al. study is then pooled
E Estimating asthma exacerbations associated with air pollution exposures is difficult, due to concerns about
double-counting of benefits. Concerns over double-counting stem from the fact that studies of the general
population also include asthmatics, so estimates based solely on the asthmatic population cannot be directly
added to the general population numbers without double-counting. In one specific case (upper respiratory
symptoms in children), the only study available is limited to asthmatic children, so this endpoint can be readily
included in the calculation of total benefits. However, other endpoints, such as lower respiratory symptoms and
MRADs, are estimated for the total population that includes asthmatics. Therefore, to simply add predictions of
asthma-related symptoms generated for the population of asthmatics to these total population-based estimates
could result in double-counting, especially if they evaluate similar endpoints.
4-16
-------
with the cough-related estimate generated using the Vedal study. The rationale for this
second pooling step is similar to the first; both studies are attempting to quantify the same
overall endpoint (asthma exacerbations).
Additional epidemiological studies are available for characterizing asthma-related
health endpoints (the full list of epidemiological studies considered for modeling asthma-
related incidence is presented in Table 4.2-2). However, we do not use these additional
studies in this analysis. In particular, the Yu et al. (2000)97 estimates show a much higher
baseline incidence rate than other studies, which may lead to an overstatement of the expected
impacts in the overall asthmatic population. The Whittemore and Korn (1980)98 study did not
use a well-defined endpoint, instead focusing on a respondent-defined "asthma attack." Other
studies looked at respiratory symptoms in asthmatics but did not focus on specific
exacerbations of asthma.
Treatment of Potential Thresholds in PM2.s-Related Health Impact Functions
Unless specifically noted, our premature mortality benefits estimates are based on an
assumed cutpoint in the premature mortality concentration-response function at 10 pg/m3, and
an assumed cutpoint of 10 pg/m3for the concentration-response functions for morbidity
associated with short term exposure to PM2.5. The 10 pg/m3 threshold reflects comments from
the U.S. EPA's Science Advisory Board Clean Air Science Advisory Committee (CASAC)
(U.S. EPA Science Advisory Board, 2005)."
4-17
-------
Table 4.2-2. Studies Examining Health Impacts in the Asthmatic Population Evaluated for Use in the
Health Impacts Analysis
ENDPOINT
DEFINITION
POLLUTANT
STUDY
Asthma Attack Indicators
Shortness of breath
Cough
Wheeze
Asthma
exacerbation
Cough
Prevalence of shortness of
breath; incidence of
shortness of breath
Prevalence of cough;
incidence of cough
Prevalence of wheeze;
incidence of wheeze
>= 1 mild asthma
symptom: wheeze, cough,
chest tightness, shortness of
breath
Prevalence of cough
PM2.5
PM2.5
PM2.5
PMio, PMi.o
PM10
Ostroetal. (2001)
Ostroetal. (2001)
Ostroetal. (2001)
Yu et al. (2000)
Vedaletal. (1998)
STUDY
POPULATION
African-American
asthmatics, 8-13
African-American
asthmatics, 8-13
African-American
asthmatics, 8-13
Asthmatics, 5-13
Asthmatics, 6-13
Other Symptoms/Illness Endpoints
Upper respiratory
symptoms
Moderate or worse
asthma
Acute bronchitis
Phlegm
Asthma attacks
>= 1 of the following:
runny or stuffy nose; wet
cough; burning, aching, or
red eyes
Probability of moderate (or
worse) rating of overall
asthma status
>= 1 episodes of bronchitis
in the past 12 months
"Other than with colds,
does this child usually seem
congested in the chest or
bring up phlegm?"
Respondent-defined asthma
attack
PM10
PM2.5
PM2.5
PM2.5
PM2.5
Popeetal. (1991)
Ostroetal. (1991)
McConnell et al.
(1999)
McConnell et al.
(1999)
Whittemore and
Korn (1980)
Asthmatics, 9-11
Asthmatics, all
ages
Asthmatics, 9-15
Asthmatics, 9-15
Asthmatics, all
ages
4.2.1.2.2 Ozone-Related Health Impact Functions
Ozone-Related Premature Mortality
While particulate matter is the criteria pollutant most clearly associated with
premature mortality, research suggests that short-term repeated ozone exposure likely
contributes to premature death. In a recent report on the estimation of ozone-related
premature mortality published by the National Research Council (NRC), 10° a panel of experts
and reviewers concluded that ozone-related mortality should be included in estimates of the
health benefits of reducing ozone exposure. The report also recommended that little or no
weight be given to the assumption that there is no causal association between ozone exposure
and premature mortality.
4-18
-------
We estimate the change in mortality incidence and estimated credible interval
resulting from application of the effect estimate from the following studies: the Bell et al.
(2004) NMMAPS analysis, Huang et al. (2004), Schwartz (2004), and effect estimates from
the three meta-analyses - Bell et al. (2005), Ito et al. (2005), and Levy et al. (2005). The
results from each study are presented separately to reflect differences in the study designs and
assumptions about causality. However, it is important to note that this procedure only
captures the uncertainty in the underlying epidemiological work, and does not capture other
sources of uncertainty, such as uncertainty in the estimation of changes in air pollution
exposure.
Respiratory Hospital Admissions Effect Estimates
Detailed hospital admission and discharge records provide data for an extensive body
of literature examining the relationship between hospital admissions and air pollution. This is
especially true for the portion of the population aged 65 and older, because of the availability
of detailed Medicare records. In addition, there is one study (Burnett et al., 2001)101
providing an effect estimate for respiratory hospital admissions in children under two.
Because the number of hospital admission studies we considered is so large, we used
results from a number of studies to pool some hospital admission endpoints. Pooling is the
process by which multiple study results may be combined in order to produce better estimates
of the effect estimate, or p. For a complete discussion of the pooling process, see the
BenMAP manual for technical details.0 To estimate total respiratory hospital admissions
associated with changes in ambient ozone concentrations for adults over 65, we first estimated
the change in hospital admissions for each of the different effects categories that each study
provided for each city. These cities included Minneapolis, Detroit, Tacoma and New Haven.
To estimate total respiratory hospital admissions for Detroit, we added the pneumonia and
COPD estimates, based on the effect estimates in the Schwartz study (1994).102 Similarly, we
summed the estimated hospital admissions based on the effect estimates the Moolgavkar
study reported for Minneapolis (Moolgavkar et al., 1997).103 To estimate total respiratory
hospital admissions for Minneapolis using the Schwartz study (1994),104 we simply estimated
pneumonia hospital admissions based on the effect estimate. Making this assumption that
pneumonia admissions represent the total impact of ozone on hospital admissions in this city
will give some weight to the possibility that there is no relationship between ozone and
COPD, reflecting the equivocal evidence represented by the different studies. We then used a
fixed-effects pooling procedure to combine the two total respiratory hospital admission
estimates for Minneapolis. Finally, we used random effects pooling to combine the results for
Minneapolis and Detroit with results from studies in Tacoma and New Haven from Schwartz
(1995). As noted above, this pooling approach incorporates both the precision of the
individual effect estimates and between-study variability characterizing differences across
study locations.
F A credible interval is a posterior probability interval used in Bayesian statistics, which is similar to a
confidence interval used in frequentist statistics.
G BenMAP and its supporting manual are available for download at http://www.epa.gov/air/benmap. Accessed
January 9, 2009.
4-19
-------
Asthma-Related Emergency Room Visits Effect Estimates
We used three studies as the source of the concentration-response functions we used to
estimate the effects of ozone exposure on asthma-related emergency room (ER) visits: Peel et
al. (2005);106 Wilson et al. (2005);107 and Jaffe et al. (2003).10^ We estimated the change in
ER visits using the effect estimate (s) from each study and then pooled the results using the
random effects pooling technique (see the BenMAP manual for technical details). The study
by Jaffe et al. (2003) examined the relationship between ER visits and air pollution for
populations aged five to 34 in the Ohio cities of Cleveland, Columbus and Cincinnati from
1991 through 1996. In single-pollutant Poisson regression models, ozone was linked to
asthma visits. We use the pooled estimate across all three cities as reported in the study. The
Peel et al. study (2005) estimated asthma-related ER visits for all ages in Atlanta, using air
quality data from 1993 to 2000. Using Poisson generalized estimating equations, the authors
found a marginal association between the maximum daily 8-hour average ozone level and ER
visits for asthma over a 3-day moving average (lags of 0, 1, and 2 days) in a single pollutant
model. Wilson et al. (2005) examined the relationship between ER visits for respiratory
illnesses and asthma and air pollution for all people residing in Portland, Maine from 1998-
2000 and Manchester, New Hampshire from 1996-2000. For all models used in the analysis,
the authors restricted the ozone data incorporated into the model to the months ozone levels
are usually measured, the spring-summer months (April through September). Using the
generalized additive model, Wilson et al. (2005) found a significant association between the
maximum daily 8-hour average ozone level and ER visits for asthma in Portland, but found no
significant association for Manchester. Similar to the approach used to generate effect
estimates for hospital admissions, we used random effects pooling to combine the results
across the individual study estimates for ER visits for asthma. The Peel et al. (2005) and
Wilson et al. (2005) Manchester estimates were not significant at the 95 percent level, and
thus, the confidence interval for the pooled incidence estimate based on these studies includes
negative values. This is an artifact of the statistical power of the studies, and the negative
values in the tails of the estimated effect distributions do not represent improvements in health
as ozone concentrations are increased. Instead these should be viewed as a measure of
uncertainty due to limitations in the statistical power of the study. Note that we included both
hospital admissions and ER visits as separate endpoints associated with ozone exposure,
because our estimates of hospital admission costs do not include the costs of ER visits, and
because most asthma ER visits do not result in a hospital admission.
Minor Restricted Activity Days Effects Estimate
Minor restricted activity days (MRADs) occur when individuals reduce most usual
daily activities and replace them with less-strenuous activities or rest, but do not miss work or
school. We estimated the effect of ozone exposure on MRADs using a concentration-
response function derived from Ostro and Rothschild (1989).109 These researchers estimated
the impact of ozone and PM2.5 on MRAD incidence in a national sample of the adult working
population (ages 18 to 65) living in metropolitan areas. We developed separate coefficients
for each year of the Ostro and Rothschild analysis (1976-1981), which we then combined for
use in EPA's analysis. The effect estimate used in the impact function is a weighted average
of the coefficients in Ostro and Rothschild (1989, Table 4), using the inverse of the variance
as the weight.
4-20
-------
School Absences Effect Estimate
Children may be absent from school due to respiratory or other acute diseases caused,
or aggravated by, exposure to air pollution. Several studies have found a significant
association between ozone levels and school absence rates. We use two studies (Gilliland et
al., 2001; Chen et al., 2000)110'111 to estimate changes in school absences resulting from
changes in ozone levels. The Gilliland et al. study estimated the incidence of new periods of
absence, while the Chen et al. study examined daily absence rates. We converted the
Gilliland et al. estimate to days of absence by multiplying the absence periods by the average
duration of an absence. We estimated 1.6 days as the average duration of a school absence,
the result of dividing the average daily school absence rate from Chen et al. (2000) and
Ransom and Pope (1992) by the episodic absence duration from Gilliland et al. (2001). Thus,
each Gilliland et al. period of absence is converted into 1.6 absence days.
Following recent advice from the National Research Council (2002),112 we calculated
reductions in school absences for the full population of school age children, ages five to 17.
This is consistent with recent peer-reviewed literature on estimating the impact of ozone
exposure on school absences (Hall et al. 2003).113 We estimated the change in school
absences using both Chen et al. (2000) and Gilliland et al. (2001) and then, similar to hospital
admissions and ER visits, pooled the results using the random effects pooling procedure.
4.2.1.3 Baseline PM Health Effect Incidence Rates
The epidemiological studies of the association between pollution levels and adverse
health effects generally provide a direct estimate of the relationship of air quality changes to
the relative risk of a health effect, rather than an estimate of the absolute number of avoided
cases. For example, a typical result might be that a 10 pg/m3 decrease in daily PM2.5 levels
might decrease hospital admissions by 3 percent. To then convert this relative change into a
number of cases, the baseline incidence of the health effect is necessary. The baseline
incidence rate provides an estimate of the incidence rate (number of cases of the health effect
per year, usually per 10,000 or 100,000 general population) in the assessment location
corresponding to baseline pollutant levels in that location. To derive the total baseline
incidence per year, this rate must be multiplied by the corresponding population number (e.g.,
if the baseline incidence rate is number of cases per year per 100,000 population, it must be
multiplied by the number of 100,000s in the population).
Some epidemiological studies examine the association between pollution levels and
adverse health effects in a specific subpopulation, such as asthmatics or diabetics. In these
cases, it is necessary to develop not only baseline incidence rates, but also prevalence rates for
the defining condition (e.g., asthma). For both baseline incidence and prevalence data, we use
age-specific rates where available. Impact functions are applied to individual age groups and
then summed over the relevant age range to provide an estimate of total population benefits.
In most cases, because of a lack of data or methods, we have not attempted to project
incidence rates to future years, instead assuming that the most recent data on incidence rates is
the best prediction of future incidence rates. In recent years, better data on trends in incidence
and prevalence rates for some endpoints, such as asthma, have become available. We are
4-21
-------
working to develop methods to use these data to project future incidence rates. However, for
our primary benefits analysis, we continue to use current incidence rates. The one exception
is in the case of premature mortality. In this case, we have projected mortality rates such that
future mortality rates are consistent with our projections of population growth. Compared
with previous analyses, this will result in a reduction in the mortality related impacts of air
pollution in future years.
Table 4.2-3 summarizes the baseline incidence data and sources used in the benefits
analysis. We use the most geographically disaggregated data available. For premature
mortality, county-level data are available. For hospital admissions, regional rates are
available. However, for all other endpoints, a single national incidence rate is used, due to a
lack of more spatially disaggregated data. In these cases, we used national incidence rates
whenever possible, because these data are most applicable to a national assessment of
benefits. However, for some studies, the only available incidence information comes from the
studies themselves; in these cases, incidence in the study population is assumed to represent
typical incidence at the national level.
Table 4.2-3: Baseline Incidence Rates and Population Prevalence Rates for Use in Impact Functions,
General Population
ENDPOINT
Mortality
Hospitalizations
Asthma ER Visits
Chronic Bronchitis
Nonfatal
Myocardial
Infarction (heart
attacks)
Asthma
Exacerbations
PARAMETER
Daily or annual mortality
rate
Daily hospitalization rate
Daily asthma ER visit rate
Annual prevalence rate per
person
- Aged 18-44
- Aged 45-64
- Aged 65 and older
Annual incidence rate per
person
Daily nonfatal myocardial
infarction incidence rate per
person, 18+
- Northeast
- Midwest
- South
- West
Incidence (and prevalence)
among asthmatic African-
American children
- daily wheeze
- daily cough
- daily dyspnea
Prevalence among asthmatic
children
RATES
Value
Age-, cause-, and
county-specific rate
Age-, region-, and
cause-specific rate
Age- and region-
specific visit rate
0.0367
0.0505
0.0587
0.00378
0.0000159
0.0000135
0.0000111
0.0000100
0.076 (0.173)
0.067 (0.145)
0.037 (0.074)
Source*
CDC Wonder (1996-1998)
1999 NHDS public use data files"
2000 NHAMCS public use data
files"; 1999 NHDS public use data
filesb
1999 NHIS (American Lung
Association, 2002, Table 4)
Abbey et al. (1993, Table 3)
1999 NHDS public use data files";
adjusted by 0.93 for probability of
surviving after 28 days (Rosamond
etal, 1999)
Ostroetal. (2001)
Vedaletal. (1998)
4-22
-------
Acute Bronchitis
Lower Respiratory
Symptoms
Upper Respiratory
Symptoms
Work Loss Days
Minor Restricted-
Activity Days
- daily wheeze
- daily cough
- daily dyspnea
Annual bronchitis incidence
rate, children
Daily lower respiratory
symptom incidence among
childrend
Daily upper respiratory
symptom incidence among
asthmatic children
Daily WLD incidence rate
per person (18-65)
- Aged 18-24
- Aged 25-44
- Aged 45-64
Daily MRAD incidence rate
per person
0.038
0.086
0.045
0.043
0.0012
0.3419
0.00540
0.00678
0.00492
0.02137
American Lung Association (2002,
Table 11)
Schwartz et al. (1994, Table 2)
Popeetal. (1991, Table 2)
1996 HIS (Adams, Hendershot, and
Marano, 1999, Table 41); U.S.
Bureau of the Census (2000)
Ostro and Rothschild (1989, p. 243)
The following abbreviations are used to describe the national surveys conducted by the National Center for
Health Statistics: HIS refers to the National Health Interview Survey; NHDS—National Hospital Discharge
Survey; NHAMCS—National Hospital Ambulatory Medical Care Survey.
Seeftp://ftp.cdc.gov/pub/Health_Statistics/NCHS/Datasets/NHDS/.
Seeftp://ftp.cdc.gov/pub/Health_Statistics/NCHS/Datasets/NHAMCS/.
Lower respiratory symptoms are defined as two or more of the following: cough, chest pain, phlegm, and
wheeze.
Baseline age, cause, and county-specific mortality rates were obtained from the U.S.
Centers for Disease Control and Prevention (CDC) for the years 1996 through 1998. CDC
maintains an online data repository of health statistics, CDC Wonder, accessible at
http://wonder.cdc.gov/. The mortality rates provided are derived from U.S. death records and
U.S. Census Bureau postcensal population estimates. Mortality rates were averaged across 3
years (1996 through 1998) to provide more stable estimates. When estimating rates for age
groups that differed from the CDC Wonder groupings, we assumed that rates were uniform
across all ages in the reported age group. For example, to estimate mortality rates for
individuals ages 30 and up, we scaled the 25- to 34-year-old death count and population by
one-half and then generated a population-weighted mortality rate using data for the older age
groups.
To estimate age- and county-specific mortality rates in years 2000 through 2020, we
calculated adjustment factors, based on a series of Census Bureau projected national mortality
rates, to adjust the CDC Wonder age- and county-specific mortality rates in 1996-1998 to
corresponding rates for each future year. For the analysis year 2020, these adjustment factors
ranged across age categories from 0.76 to 0.86
For the set of endpoints affecting the asthmatic population, in addition to baseline
incidence rates, prevalence rates of asthma in the population are needed to define the
applicable population. Table 4.2-3 lists the baseline incidence rates and their sources for
asthma symptom endpoints. Table 4.2-4 lists the prevalence rates used to determine the
4-23
-------
applicable population for asthma symptom endpoints. Note that these reflect current asthma
prevalence and assume no change in prevalence rates in future years.
Table 4.2-4. Asthma Prevalence Rates Used to Estimate Asthmatic Populations in Impact Functions
POPULATION GROUP
All Ages
< 18
5-17
18-44
45-64
65+
Male, 27+
African American, 5 to 17
African American, <18
ASTHMA PREVALENCE RATES
Value
0.0386
0.0527
0.0567
0.0371
0.0333
0.0221
0.021
0.0726
0.0735
Source
American Lung Association (2002, Table 7)— based on 1999 HIS
American Lung Association (2002, Table 7)— based on 1999 HIS
American Lung Association (2002, Table 7) — based on 1999 HIS
American Lung Association (2002, Table 7) — based on 1999 HIS
American Lung Association (2002, Table 7)— based on 1999 HIS
American Lung Association (2002, Table 7)— based on 1999 HIS
2000 HIS public use data files"
American Lung Association (2002, Table 9)— based on 1999 HIS
American Lung Association (2002, Table 9) — based on 1999 HIS
a Seeftp://ftp.cdc.gov/pub/Health_Statistics/NCHS/Datasets/NHIS/2000/.
4.2.1.4 Baseline Incidence Rates for Ozone-related Health Impacts
Epidemiological studies of the association between pollution levels and adverse health
effects generally provide a direct estimate of the relationship of air quality changes to the
relative risk of a health effect, rather than estimating the absolute number of avoided cases.
For example, a typical result might be that a 100 ppb decrease in daily ozone levels might, in
turn, decrease hospital admissions by 3 percent. The baseline incidence of the health effect is
necessary to convert this relative change into a number of cases. A baseline incidence rate is
the estimate of the number of cases of the health effect per year in the assessment location, as
it corresponds to baseline pollutant levels in that location. To derive the total baseline
incidence per year, this rate must be multiplied by the corresponding population number. For
example, if the baseline incidence rate is the number of cases per year per 100,000 people,
that number must be multiplied by the number of 100,000s in the population.
Table 4.2-5 summarizes the sources of baseline incidence rates and provides average
incidence rates for the endpoints included in the analysis. For both baseline incidence and
prevalence data, we used age-specific rates where available. We applied concentration-
response functions to individual age groups and then summed over the relevant age range to
provide an estimate of total population benefits. In most cases, we used a single national
incidence rate, due to a lack of more spatially disaggregated data. Whenever possible, the
national rates used are national averages, because these data are most applicable to a national
assessment of benefits. For some studies, however, the only available incidence information
comes from the studies themselves; in these cases, incidence in the study population is
assumed to represent typical incidence at the national level. Regional incidence rates are
available for hospital admissions, and county-level data are available for premature mortality.
We have projected mortality rates such that future mortality rates are consistent with our
projections of population growth.
4-24
-------
Table 4.2-5. National Average Baseline Incidence Rates"
ENDPOINT
Mortality
Respiratory
Hospital
Admissions.
Asthma ER
visits
Minor
Restricted
Activity Days
(MRADs)
School Loss
Days
SOURCE
CDC Compressed Mortality
File, accessed through CDC
Wonder (1996-1998)
1999 NHDS public use data
filesb
2000 NHAMCS public use
data files0; 1999 NHDS
public use data filesb
Ostro and Rothschild
(1989, p. 243)
National Center for
Education Statistics (1996)
and 1996 HIS (Adams etal,
1999, Table 4 7); estimate of
180 school days per year
NOTES
non-
accidental
incidence
incidence
incidence
all-cause
RATE PER 100 PEOPLE PER YEARU BY AGE
GROUP
<18
0.03
0.04
1.01
990
18-
24
0.02
0.08
1.09
780
25-
34
0.06
0.21
0.75
780
35-
44
0.15
0.68
0.44
780
45-
54
0.38
1.93
0.35
780
55-
64
1.01
4.40
0.43
780
65+
4.94
11.63
0.23
The following abbreviations are used to describe the national surveys conducted by the National Center for Health
Statistics: HIS refers to the National Health Interview Survey; NHDS - National Hospital Discharge Survey; NHAMCS -
National Hospital Ambulatory Medical Care Survey.
b See ftp://ftp.cdc.gov/pub/Health Statistics/NCHS/Datasets/NHDS/
c See ftp://ftp.cdc.gov/pub/Health Statistics/NCHS/Datasets/NHAMCS/
All of the rates reported here are population-weighted incidence rates per 100 people per year. Additional details on the
incidence and prevalence rates, as well as the sources for these rates are available upon request.
Table 4.2-5. National Average Baseline Incidence Rates (continued)
ENDPOINT
Asthma Exacerbations
SOURCE
Ostro et al. (2001)
Vedaletal. (1998)
NOTES
Incidence (and
prevalence) among
asthmatic African-
American children
Incidence (and
prevalence) among
asthmatic children
Daily wheeze
Daily cough
Daily
dyspnea
Daily wheeze
Daily cough
Daily
dyspnea
RATE PER 100
PEOPLE PER YEAR
0.08 (0.17)
0.07 (0.15)
0.04 (0.07)
0.04
0.09
0.05
4.2.2 Manipulating Air Quality Modeling Data for Health Impacts Analysis
In Chapter 3, we summarized the methods for and results of estimating air quality for
the 2020 base case and proposed EGA scenario. These air quality results are in turn
associated with human populations to estimate changes in health effects. For the purposes of
this analysis, we focus on the health effects that have been linked to ambient changes in ozone
and PM2.s related to emission reductions estimated to occur due to the proposed EGA. We
4-25
-------
estimate ambient PM2.5 and ozone concentrations using the Community Multiscale Air
Quality model (CMAQ). This section describes how we converted the CMAQ modeling
output into full-season profiles suitable for the health impacts analysis.
4.2.2.1 General Methodology
First, we extracted hourly, surface-layer PM and ozone concentrations for each grid
cell from the standard CMAQ output files. For ozone, these model predictions are used in
conjunction with the observed concentrations obtained from the Aerometric Information
Retrieval System (AIRS) to generate ozone concentrations for the entire ozone season.H>1 The
predicted changes in ozone concentrations from the future-year base case to future-year
control scenario serve as inputs to the health and welfare impact functions of the benefits
analysis (i.e., BenMAP).
To estimate ozone-related health effects for the contiguous United States, full-season
ozone data are required for every BenMAP grid-cell. Given available ozone monitoring data,
we generated full-season ozone profiles for each location in two steps: (1) we combined
monitored observations and modeled ozone predictions to interpolate hourly ozone
concentrations to a grid of 12-km by 12-km population grid cells for the contiguous 48 states,
and (2) we converted these full-season hourly ozone profiles to an ozone measure of interest,
such as the daily 8-hour maximum.J'K
For PM2.5, we also use the model predictions in conjunction with observed monitor
data. CMAQ generates predictions of hourly PM species concentrations for every grid. The
species include a primary coarse fraction (corresponding to PM in the 2.5 to 10 micron size
range), a primary fine fraction (corresponding to PM less than 2.5 microns in diameter), and
several secondary particles (e.g., sulfates, nitrates, and organics). PM2.5 is calculated as the
sum of the primary fine fraction and all of the secondarily formed particles. Future-year
estimates of PM2.5 were calculated using relative reduction factors (RRFs) applied to 2002
ambient PM2.s and PM2.s species concentrations. A gridded field of PM2.s concentrations was
created by interpolating Federal Reference Monitor ambient data and IMPROVE ambient
data. Gridded fields of PM2.5 species concentrations were created by interpolating US EPA
speciation network (ESPN) ambient data and IMPROVE data. The ambient data were
interpolated to the CMAQ 12 km grid.
The procedures for determining the RRFs are similar to those in US EPA's draft
guidance for modeling the PM2.5 standard (EPA, 1999). The guidance recommends that
model predictions be used in a relative sense to estimate changes expected to occur in each
major PM2.s species. The procedure for calculating future-year PM2.5 design values is called
H The ozone season for this analysis is defined as the 5-month period from May to September.
1 Based on AIRS, there were 961 ozone monitors with sufficient data (i.e., 50 percent or more days reporting at
least nine hourly observations per day [8 am to 8 pm] during the ozone season).
J The 12-km grid squares contain the population data used in the health benefits analysis model, BenMAP.
K This approach is a generalization of planar interpolation that is technically referred to as enhanced Voronoi
Neighbor Averaging (EVNA) spatial interpolation. See the BenMAP manual for technical details, available for
download at http://www.epa.gov/air/benmap.
4-26
-------
the "Speciated Modeled Attainment Test (SMAT)." EPA used this procedure to estimate the
ambient impacts of the proposed EGA controls.
4.2.2.2 Emissions Inventory Boundary Distance Error
As noted in Appendix 2F to Chapter 2, the air quality modeling used for this analysis
is based on inventory estimates that were modeled using incorrect boundary information. The
impact of this difference, while modest, leads to an underestimate of the benefits that are
presented in this Chapter. Please refer to Appendix 2F for more information on the emissions
excluded from the health impacts analysis of the proposed EGA.
4.3 Methods for Describing Uncertainty
For this analysis, consistent with the approach used in the analyses for the recent PM
and Ozone NAAQS, we addressed key sources of uncertainty through Monte Carlo
propagation of uncertainty in the concentration-response (CR) functions. It should be noted
that the Monte Carlo-generated distributions of health impacts reflect only some of the
uncertainties in the input parameters. Uncertainties associated with emissions, air quality
modeling, populations, and baseline health effect incidence rates are not represented in the
distributions of avoided health impacts associated with the implementation of the proposed
EGA. A complete description of uncertainty related to health impacts analyses can be found
in the regulatory impact analysis drafted in support of the final Ozone NAAQS analysis.114
4-27
-------
1 Pope, C.A., III, R.T. Burnett, M.J. Thun, E.E. Calle, D. Krewski, K. Ito, and G.D. Thurston. 2002. "Lung
Cancer, Cardiopulmonary Mortality, and Long-term Exposure to Fine Particulate Air Pollution." Journal of the
American Medical Association 287:1132-1141.
2 Bell, M.L., et al. 2004. Ozone and short-term mortality in 95 US urban communities, 1987-2000. Jama, 2004.
292(19): p. 2372-8.
3 Laden, F., J. Schwartz, F.E. Speizer, and D.W. Dockery. 2006. Reduction in Fine Particulate Air
Pollution and Mortality. American Journal of Respiratory and Critical Care Medicine. 173: 667-672.
4 Levy, J.I., S.M. Chemerynski, and J.A. Sarnat. 2005. Ozone exposure and mortality: an empiric bayes
metaregression analysis. Epidemiology, 2005. 16(4): p. 458-68.
5 U.S. Environmental Protection Agency. March 2008. Final Ozone NAAQS Regulatory Impact Analysis.
Prepared by: Office of Air and Radiation, Office of Air Quality Planning and Standards.
6 U.S. Environmental Protection Agency. October 2006. Final Regulatory Impact Analysis (RIA) for the
Proposed National Ambient Air Quality Standards for Particulate Matter. Prepared by: Office of Air and
Radiation. Available at HTTP://www.epa.gov/ttn/ecas/ria.html.
7 GeoLytics Inc. 2002. Geolytics CensusCD® 2000 Short Form Blocks. CD-ROM Release 1.0. GeoLytics,
Inc. East Brunswick, NJ. Available: http://www.geolytics.com/ [accessed 29 September 2004].
8 Woods & Poole Economics Inc. 2001. Population by Single Year of Age CD. CD-ROM. Woods & Poole
Economics, Inc. Washington, D.C.
9 U.S. Environmental Protection Agency (2006) Air quality criteria for ozone and related photochemical
oxidants (second external review draft) Research Triangle Park, NC: National Center for Environmental
Assessment: report no. EPA/600R-05/004aB-cB, Sv.Available:
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=137307[March 2006]
10 U.S. Environmental Protection Agency, 2004. Air Quality Criteria for Particulate Matter Volume II of II.
National Center for Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC EPA/600/P-99/002bF
11 World Health Organization (WHO). 2003. Health Aspects of Air Pollution with Particulate Matter, Ozone
and Nitrogen Dioxide: Report on a WHO Working Group. World Health Organization. Bonn, Germany.
EUR/03/5042688.
12 Anderson HR, Atkinson RW, Peacock JL, Marston L, Konstantinou K. 2004. Meta-analysis of time-series
studies and panel studies of Particulate Matter (PM) and Ozone (03): Report of a WHO task group. Copenhagen,
Denmark: World Health Organization.
13 Bell, M.L., et al. 2004. Ozone and short-term mortality in 95 US urban communities, 1987-2000. Jama, 2004.
292(19): p. 2372-8.
14 Huang, Y.; Dominici, F.; Bell, M. L. (2005) Bayesian hierarchical distributed lag models for summer ozone
exposure and cardio-respiratory mortality. Environmetrics 16: 547-562.
4-28
-------
15 Schwartz, J. (2005) How sensitive is the association between ozone and daily deaths to control for
temperature? Am. J. Respir. Crit. Care Med. 171: 627-631.
16 Bell, M.L., F. Dominici, and J.M. Samet. A meta-analysis of time-series studies of ozone and mortality with
comparison to the national morbidity, mortality, and air pollution study. Epidemiology, 2005. 16(4): p. 436-45.
17 Ito, K., S.F. De Leon, and M. Lippmann. Associations between ozone and daily mortality: analysis and meta-
analysis. Epidemiology, 2005. 16(4): p. 446-57.
18 Levy, J.I., S.M. Chemerynski, and J.A. Sarnat. 2005. Ozone exposure and mortality: an empiric bayes
metaregression analysis. Epidemiology, 2005. 16(4): p. 458-68.
19 Pope, C.A., III, R.T. Burnett, M.J. Thun, E.E. Calle, D. Krewski, K. Ito, and G.D. Thurston. 2002. "Lung
Cancer, Cardiopulmonary Mortality, and Long-term Exposure to Fine Particulate Air Pollution." Journal of the
American Medical Association 287:1132-1141.
20 Laden, F., J. Schwartz, F.E. Speizer, and D.W. Dockery. 2006. Reduction in Fine Particulate Air
Pollution and Mortality. American Journal of Respiratory and Critical Care Medicine. 173: 667-672.
21 Woodruff, T.J., J. Grillo, and K.C. Schoendorf. 1997. "The Relationship Between Selected Causes of
Postneonatal Infant Mortality and Particulate Air Pollution in the United States." Environmental Health
Perspectives 105(6):608-612.
99
Abbey, D.E., B.L. Hwang, R.J. Burchette, T. Vancuren, and P.K. Mills. 1995. "Estimated Long-Term
Ambient Concentrations of PM(10) and Development of Respiratory Symptoms in a Nonsmoking Population."
Archives of Environmental Health 50(2): 139-152.
23
Peters, A., D.W. Dockery, J.E. Muller, and M.A. Mittleman. 2001. "Increased Particulate Air Pollution and
the Triggering of Myocardial Infarction." Circulation 103:2810-2815.
24 Schwartz J. 1995. Short term fluctuations in air pollution and hospital admissions of the elderly for
respiratory disease. Thorax 50(5):531-538.
25 Schwartz J. 1994a. PM(10) Ozone, and Hospital Admissions For the Elderly in Minneapolis St Paul,
Minnesota. Arch Environ Health 49(5):366-374.
26 Schwartz J. 1994b. Air Pollution and Hospital Admissions For the Elderly in Detroit, Michigan. Am J Respir
Crit Care Med 150(3):648-655.
27 Moolgavkar SH, Luebeck EG, Anderson EL. 1997. Air pollution and hospital admissions for respiratory
causes in Minneapolis St. Paul and Birmingham. Epidemiology 8(4):364-370.
28 Burnett RT, Smith-Doiron M, Stieb D, Raizenne ME, Brook JR, Dales RE, et al. 2001. Association between
ozone and hospitalization for acute respiratory diseases in children less than 2 years of age. Am J Epidemiol
153(5):444-452.
29
Moolgavkar, S.H. 2003. "Air Pollution and Daily Deaths and Hospital Admissions in Los Angeles and Cook
Counties." In Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. Boston,
MA: Health Effects Institute.
4-29
-------
30
Ito, K. 2003. "Associations of Particulate Matter Components with Daily Mortality and Morbidity in Detroit,
Michigan." In Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. Health
Effects Institute, Boston, MA.
31 Moolgavkar, S.H. 2000. "Air Pollution and Hospital Admissions for Diseases of the Circulatory System in
Three U.S. Metropolitan Areas." Journal of the Air and Waste Management Association 50:1199-1206.
32 Sheppard, L. 2003. "Ambient Air Pollution and Nonelderly Asthma Hospital Admissions in Seattle,
Washington, 1987-1994." In Revised Analyses of Time-Series Studies of Air Pollution and Health. Special
Report. Boston, MA: Health Effects Institute.
33 Jaffe DH, Singer ME, Rimm AA. 2003. Air pollution and emergency department visits for asthma among
Ohio Medicaid recipients, 1991-1996. Environ Res 91(l):21-28.
34 Peel, J. L., P. E. Tolbert, M. Klein, et al. 2005. Ambient air pollution and respiratory emergency department
visits. Epidemiology. Vol. 16 (2): 164-74.
35 Wilson, A. M., C. P. Wake, T. Kelly, et al. 2005. Air pollution, weather, and respiratory emergency room
visits in two northern New England cities: an ecological time-series study. Environ Res. Vol. 97 (3): 312-21.
36 Norris, G., S.N. YoungPong, J.Q. Koenig, T.V. Larson, L. Sheppard, and J.W. Stout. 1999. "An Association
between Fine Particles and Asthma Emergency Department Visits for Children in Seattle." Environmental
Health Perspectives 107(6) :489-493.
37
Dockery, D.W., J. Cunningham, A.I. Damokosh, L.M. Neas, J.D. Spengler, P. Koutrakis, J.H. Ware, M.
Raizenne, and F.E. Speizer. 1996. "Health Effects of Acid Aerosols On North American Children-Respiratory
Symptoms." Environmental Health Perspectives 104(5):500-505.
38 Pope, C.A., III, D.W. Dockery, J.D. Spengler, and M.E. Raizenne. 1991. "Respiratory Health and PM10
Pollution: A Daily Time Series Analysis." American Review of Respiratory Diseases 144:668-674.
39 Schwartz, J., and L.M. Neas. 2000. "Fine Particles are More Strongly Associated than Coarse Particles with
Acute Respiratory Health Effects in Schoolchildren." Epidemiology 11:6-10.
Ostro, B., M. Lipsett, J. Mann, H. Braxton-Owens, and M. White. 2001. "Air Pollution and Exacerbation of
Asthma in African-American Children in Los Angeles." Epidemiology 12(2):200-208.
41 Vedal, S., J. Petkau, R. White, and J. Blair. 1998. "Acute Effects of Ambient Inhalable Particles in Asthmatic
and Nonasthmatic Children." American Journal of Respiratory and Critical Care Medicine 157(4):1034-1043.
Ostro, B.D. 1987. "Air Pollution and Morbidity Revisited: A Specification Test." Journal of Environmental
Economics Management 14:87-98.
43 Gilliland FD, Berhane K, Rappaport EB, Thomas DC, Avol E, Gauderman WJ, et al. 2001. The effects of
ambient air pollution on school absenteeism due to respiratory illnesses. Epidemiology 12(l):43-54.
44 Chen L, Jennison BL, Yang W, Omaye ST. 2000. Elementary school absenteeism and air pollution. Inhal
Toxicoll2(ll):997-1016.
45
Ostro, B.D. and S. Rothschild. 1989. "Air Pollution and Acute Respiratory Morbidity: An Observational
Study of Multiple Pollutants." Environmental Research 50:238-247.
4-30
-------
46 U.S. Science Advisory Board. 2004. Advisory Plans for Health Effects Analysis in the Analytical Plan for
EPA's Second Prospective Analysis -Benefits and Costs of the Clean Air Act, 1990—2020. EPA-SAB-
COUNCIL-ADV-04-004.
47 National Research Council (NRC). 2002. Estimating the Public Health Benefits of Proposed Air Pollution
Regulations. Washington, DC: The National Academies Press.
48 National Research Council (NRC). 1998. Research Priorities for Airborne Particulate Matter: I. Immediate
Priorities and a Long-Range Research Portfolio. Washington, DC: The National Academies Press.
49 U.S. Environmental Protection Agency, 2004b. Air Quality Criteria for Particulate Matter Volume II of II.
National Center for Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC EPA/600/P-99/002bF
50 Kunzli, N., S. Medina, R. Kaiser, P. Quenel, F. Horak Jr, and M. Studnicka. 2001. "Assessment of Deaths
Attributable to Air Pollution: Should We Use Risk Estimates Based on Time Series or on Cohort Studies?"
American Journal of Epidemiology 153(ll):1050-55.
51 National Research Council (NRC). 2002. Estimating the Public Health Benefits of Proposed Air Pollution
Regulations. Washington, DC: The National Academies Press.
52 Lave, L.B., and E.P. Seskin. 1977. Air Pollution and Human Health. Baltimore: Johns Hopkins University
Press for Resources for the Future.
53 Lipfert, F.W., S.C. Morris, and R.E. Wyzga. 1989. "Acid Aerosols—the Next Criteria Air Pollutant."
Environmental Science & Technology 23(11):1316-1322.
54 Ozkaynak, H., and G.D. Thurston. 1987. "Associations between 1980 U.S. Mortality Rates and Alternative
Measures of Airborne Particle Concentration." Risk Analysis 7(4):449-461.
55 Dockery, D.W., C.A. Pope, X.P. Xu, J.D. Spengler, J.H. Ware, M.E. Fay, B.C. Ferris, and F.E. Speizer. 1993.
"An Association between Air Pollution and Mortality in Six U.S. Cities." New England Journal of Medicine
329(24):1753-1759.
56 Pope, C.A., III, M.J. Thun, M.M. Namboodiri, D.W. Dockery, J.S. Evans, F.E. Speizer, and C.W. Heath, Jr.
1995. "Particulate Air Pollution as a Predictor of Mortality in a Prospective Study of U.S. Adults." American
Journal of Respiratory Critical Care Medicine 151:669-674.
57 Pope, C.A., III, R.T. Burnett, G.D. Thurston, M.J. Thun, E.E. Calle, D. Krewski, and J.J. Godleski. 2004.
"Cardiovascular Mortality and Long-term Exposure to Particulate Air Pollution." Circulation 109: 71-77.
58 Abbey, D.E., N. Nishino, W.F. McDonnell, R.J. Burchette, S.F. Knutsen, W. Lawrence Beeson, and J.X.
Yang. 1999. "Long-term inhalable particles and other air pollutants related to mortality in nonsmokers [see
comments]." American Journal of Respiratory and Critical Care Medicine 159(2):373-382.
59 Lipfert, F.W., H. Mitchell Perry Jr., J. Philip Miller, Jack D. Baty, Ronald E. Wyzg, and Sharon E. Carmody.
2000. "The Washington University-EPRI Veterans' Cohort Mortality Study: Preliminary Results." Inhalation
Toxicology 12:41-74.
60 Lipfert, F.W.; Perry, H.M., Jr.: Miller, J.P.; Baty, J.D.; Wyzga, R.E.; Carmody, S.E. 2003. "Air Pollution,
Blood Pressure, and Their Long-Term Associations with Mortality" Inhalation Toxicology. 15, 493-512.
4-31
-------
61 Lipfert, F.W.; Wyzga, R.E.; Baty, J.D.; Miller, J.P. 2006. "Traffic Density as a Surrogate Measure of
Environmental Exposures in Studies of Air Pollution Health Effects: Long-Term Mortality in a Cohort of US
Veterans" Atmospheric Environment 40: 154-169.
62 Krewski D., R.T. Burnett, M.S. Goldbert, K. Hoover, J. Siemiatycki, M. Jerrett, M. Abrahamowicz, and W.H.
White. July 2000. Reanalysis of the Harvard Six Cities Study and the American Cancer Society Study of
Particulate Air Pollution and Mortality. Special Report to the Health Effects Institute, Cambridge MA.
63 Industrial Economics, Incorporated (lEc). 2006. Expanded Expert Judgment Assessment of the
Concentration-Response Relationship Between PM2.5 Exposure and Mortality. Peer Review Draft. Prepared
for: Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency, Research Triangle
Park, NC. August.
64 Jerrett M, Burnett RT, Ma R, Pope CA 3rd, Krewski D, Newbold KB, Thurston G, Shi Y, Finkelstein N, Calle
EE, Thun MJ. 2005. Spatial analysis of air pollution and mortality in Los Angeles. Epidemiology. 2005
Nov;16(6):727-36.
65 Krewski, D., R. Burnett, M. Jerrett, C.A. Pope, D. Rainham, E. Calle, G. Thurston, M. Thun. 2005. Mortality
and Long-term Exposure to Ambient Air Pollution: Ongoing Analyses Based on the American Cancer Society
Cohort. Journal of Toxicology and Environmental Health, Part A. 68: 1093-1109.
66 Belanger, K., W. Beckett, E. Triche, M.B. Bracken, T. Holford, P. Ren, J.E. McSharry, D.R. Gold, T.A.
Platts-Mills, and B.P. Leaderer. 2003. "Symptoms of Wheeze and Persistent Cough in the First Year of Life:
Associations with Indoor Allergens, Air Contaminants, and Maternal History of Asthma." American Journal of
Epidemiology 158:195-202.
67 Chay, K.Y., and M. Greenstone. 2003. "The Impact of Air Pollution on Infant Mortality: Evidence from
Geographic Variation in Pollution Shocks Induced by a Recession." Quarterly Journal of Economics 118(3).
68 Woodruff, T.J., J. Grillo, and K.C. Schoendorf. 1997. "The Relationship Between Selected Causes of
Postneonatal Infant Mortality and Particulate Air Pollution in the United States." Environmental Health
Perspectives 105(6):608-612.
69 American Lung Association. 1999. "Chronic Bronchitis." Available at
http://www.lungusa.org/diseases/lungchronic.html.
70 Schwartz, J. 1993. "Particulate Air Pollution and Chronic Respiratory Disease." Environmental Research
62:7-13.
71 Abbey, D.E., B.L. Hwang, R.J. Burchette, T. Vancuren, and P.K. Mills. 1995. "Estimated Long-Term
Ambient Concentrations of PM(10) and Development of Respiratory Symptoms in a Nonsmoking Population."
Archives of Environmental Health 50(2): 139-152.
72 Peters, A., D.W. Dockery, J.E. Muller, and M.A. Mittleman. 2001. "Increased Particulate Air Pollution and
the Triggering of Myocardial Infarction." Circulation 103:2810-2815.
73 Poloniecki, J.D., R.W. Atkinson., A.P. de Leon., and H.R. Anderson. 1997. "Daily Time Series for
Cardiovascular Hospital Admissions and Previous Day's Air Pollution in London, UK." Occupational and
Environmental Medicine 54(8):535-540.
4-32
-------
74 Zanobetti, A. and J. Schwartz. 2005. The Effect of Particulate Air Pollution on Emergency Admsissions for
Myocardial Infarction: A Multicity Case-Crossover Analysis. Environmental Health Perspectives. 113: 978-
982.
75 Domenici, F., R.D. Peng, M.L. Bell, L. Pham, A. McDermott, S.L. Zeger, J.M. Samet. 2006. "Fine Particulate
Air Pollution and Hospital Admission for Cardiovascular and Respiratory Diseases. Journal of the American
Medical Association 295: 1127-1134.
76 Samet, J.M., S.L. Zeger, F. Dominici, F. Curriero, I. Coursac, D.W. Dockery, J. Schwartz, and A. Zanobetti.
June 2000. The National Morbidity, Mortality and Air Pollution Study: Part II: Morbidity, Mortality and Air
Pollution in the United States. Research Report No. 94, Part II. Health Effects Institute, Cambridge MA.
77 Moolgavkar, S.H. 2000. "Air Pollution and Hospital Admissions for Diseases of the Circulatory System in
Three U.S. Metropolitan Areas." Journal of the Air and Waste Management Association 50:1199-1206.
78 Liao, D., J. Creason, C. Shy, R. Williams, R. Watts, and R. Zweidinger. 1999. "Daily Variation of Particulate
Air Pollution and Poor Cardiac Autonomic Control in the Elderly." Environ Health Perspect 107:521-5.
79 Gold, D.R., A. Litonjua, J. Schwartz, E. Lovett, A. Larson, B. Nearing, G. Allen, M. Verrier, R. Cherry., and
R. Verrier. 2000. "Ambient Pollution and Heart Rate Variability." Circulation 101(ll):1267-73.
80 Magari, S.R., R. Hauser, J. Schwartz, P.L. Williams, T.J. Smith, and D.C. Christian!. 2001. "Association of
Heart rate Variability with Occupational and Environmental Exposure to Particulate Air Pollution." Circulation
104(9):986-91.
81 Carnethon, M.R., D. Liao, G.W. Evans, W.E. Cascio, L.E. Chambless, W.D. Rosamond, and G. Heiss. 2002.
"Does the Cardiac Autonomic Response to Postural Change Predict Incident Coronary Heart Disease and
Mortality? The Atherosclerosis Risk in Communities Study." American Journal of Epidemiology 155(l):48-56.
82 Dekker, J.M., R.S. Crow, A.R. Folsom, P.J. Hannan, D. Liao, C.A. Swenne, and E.G. Schouten. 2000. "Low
Heart Rate Variability in a 2-Minute Rhythm Strip Predicts Risk of Coronary Heart Disease and Mortality From
Several Causes: The ARIC Study." Circulation 2000 102:1239-1244.
83 Liao, D., J. Cai, W.D. Rosamond, R.W. Barnes, R.G. Hutchinson, E.A. Whitsel, P. Rautaharju, and G. Heiss.
1997. "Cardiac Autonomic Function and Incident Coronary Heart Disease: A Population-Based Case-Cohort
Study. The ARIC Study. Atherosclerosis Risk in Communities Study." American Journal of Epidemiology
145(8):696-706.
84 Tsuji, H., M.G. Larson, F.J. Venditti, Jr., E.S. Manders, J.C. Evans, C.L. Feldman, D. Levy. 1996. "Impact of
Reduced Heart Rate Variability on Risk for Cardiac Events. The Framingham Heart Study." Circulation
94(ll):2850-2855.
85 Moolgavkar, S.H. 2003. "Air Pollution and Daily Deaths and Hospital Admissions in Los Angeles and Cook
Counties." In Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. Boston,
MA: Health Effects Institute.
86 Ito, K. 2003. "Associations of Particulate Matter Components with Daily Mortality and Morbidity in Detroit,
Michigan." In Revised Analyses of Time-Series Studies of Air Pollution and Health. Special Report. Health
Effects Institute, Boston, MA.
87 Moolgavkar, S.H. 2000. "Air Pollution and Hospital Admissions for Diseases of the Circulatory System in
Three U.S. Metropolitan Areas." Journal of the Air and Waste Management Association 50:1199-1206.
4-33
-------
88 Norris, G., S.N. YoungPong, J.Q. Koenig, T.V. Larson, L. Sheppard, and J.W. Stout. 1999. "An Association
between Fine Particles and Asthma Emergency Department Visits for Children in Seattle." Environmental
Health Perspectives 107(6):489-493.
89 American Lung Association. 2002. Trends in Morbidity and Mortality: Pneumonia, Influenza, and Acute
Respiratory Conditions. American Lung Association, Best Practices and Program Services, Epidemiology and
Statistics Unit.
90 Dockery, D.W., J. Cunningham, A.I. Damokosh, L.M. Neas, J.D. Spengler, P. Koutrakis, J.H. Ware, M.
Raizenne, and F.E. Speizer. 1996. "Health Effects of Acid Aerosols On North American Children-Respiratory
Symptoms." Environmental Health Perspectives 104(5):500-505.
91 Schwartz, J., and L.M. Neas. 2000. "Fine Particles are More Strongly Associated than Coarse Particles with
Acute Respiratory Health Effects in Schoolchildren." Epidemiology 11:6-10.
92 Pope, C.A., III, D.W. Dockery, J.D. Spengler, and M.E. Raizenne. 1991. "Respiratory Health and PM10
Pollution: A Daily Time Series Analysis." American Review of Respiratory Diseases 144:668-674.
93 Ostro, B.D. 1987. "Air Pollution and Morbidity Revisited: A Specification Test." Journal of Environmental
Economics Management 14:87-98.
94 Ostro, B.D. and S. Rothschild. 1989. "Air Pollution and Acute Respiratory Morbidity: An Observational
Study of Multiple Pollutants." Environmental Research 50:238-247.
95 Ostro, B., M. Lipsett, J. Mann, H. Braxton-Owens, and M. White. 2001. "Air Pollution and Exacerbation of
Asthma in African-American Children in Los Angeles." Epidemiology 12(2):200-208.
96 Vedal, S., J. Petkau, R. White, and J. Blair. 1998. "Acute Effects of Ambient Inhalable Particles in Asthmatic
and Nonasthmatic Children." American Journal of Respiratory and Critical Care Medicine 157(4):1034-1043.
97 Yu, 0., L. Sheppard, T. Lumley, J.Q. Koenig, and G.G. Shapiro. 2000. "Effects of Ambient Air Pollution on
Symptoms of Asthma in Seattle-Area Children Enrolled in the CAMP Study." Environmental Health
Perspectives 108(12):1209-1214.
98 Whittemore, A.S., and E.L. Korn. 1980. "Asthma and Air Pollution in the Los Angeles Area." American
Journal of Public Health 70:687-696.
99 U.S. Environmental Protection Agency Science Advisory Board. 2005. EPA's Review of the National
Ambient Air Quality Standards for Paniculate Matter (Second Draft PM Staff Paper, January 2005). EPA-SAB-
CASAC-05-007. June.
100 National Research Council (NRC), 2008. Estimating Mortality Risk Reduction and Economic Benefits from
Controlling Ozone Air Pollution. The National Academies Press: Washington, D.C.
101 Burnett, R. T.: Smith-Doiron, M.; Stieb, D.: Raizenne, M. E.; Brook, J. R.: Dales, R. E.: Leech, J. A.:
Cakmak, S.: Krewski, D. (2001) Association between ozone and hospitalization for acute respiratory diseases in
children less than 2 years of age. Am. J. Epidemiol. 153: 444-452.
102 Schwartz J. 1994. Air Pollution and Hospital Admissions for the Elderly in Detroit, Michigan. Am J Respir
Crit Care Med 150(3):648-655.
4-34
-------
103 Moolgavkar SH, Luebeck EG, Anderson EL. 1997. Air pollution and hospital admissions for respiratory
causes in Minneapolis St. Paul and Birmingham. Epidemiology 8(4):364-370.
104 Schwartz]. 1994. PM(10) Ozone, and Hospital Admissions For the Elderly in Minneapolis St Paul,
Minnesota. Arch Environ Health 49(5):366-374.
105 Schwartz]. 1995. Short term fluctuations in air pollution and hospital admissions of the elderly for
respiratory disease. Thorax 50(5):531-538.
106 Peel, J. L; Tolbert, P. E.; Klein, M.; Metzger, K. B.; Flanders, W. D.; Knox, T.; Mulholland, J. A.; Ryan, P.
B.; Frumkin, H. (2005) Ambient air pollution and respiratory emergency department visits. Epidemiology 16:
164-174.
107 Wilson, A. M., C. P. Wake, T. Kelly, et al. 2005. Air pollution, weather, and respiratory emergency room
visits in two northern New England cities: an ecological time-series study. Environ Res. Vol. 97 (3): 312-21.
108 Jaffe DH, Singer ME, Rimm AA. 2003. Air pollution and emergency department visits for asthma among
Ohio Medicaid recipients, 1991-1996. Environ Res 91(l):21-28.
109 Ostro BD, Rothschild S. 1989. Air Pollution and Acute Respiratory Morbidity—an Observational Study of
Multiple Pollutants. Environ Res 50(2):238-247.
110 Gilliland FD, Berhane K, Rappaport EB, Thomas DC, Avol E, Gauderman WJ, et al. 2001. The effects of
ambient air pollution on school absenteeism due to respiratory illnesses. Epidemiology 12(l):43-54.
111 Chen L, Jennison BL, Yang W, Omaye ST. 2000. Elementary school absenteeism and air pollution. Inhal
Toxicoll2(ll):997-1016.
112 National Research Council (NRC). 2002. Estimating the Public Health Benefits of Proposed Air Pollution
Regulations. The National Academies Press: Washington, D.C.
113 HallJV, Brajer V, Lurmann FW. 2003. Economic Valuation of Ozone-related School Absences in the South
Coast Air Basin of California. Contemporary Economic Policy 21(4):407-417.
114 U.S. Environmental Protection Agency. March 2008. Final Ozone NAAQS Regulatory Impact Analysis.
Prepared by: Office of Air and Radiation, Office of Air Quality Planning and Standards.
4-35
-------
5 Costs
The reduction of SOx, NOx, and PM emissions from ships has an associated cost that
reaches not only to the shipping industry but also to marine fuel suppliers and companies who
rely on the shipping industry. Though these cost impacts do exist, analyses presented in this
document indicate that the costs associated with the proposed EGA are expected to have a
minimal economic impact and to be relatively small compared to the resulting improvements in
air quality. This chapter describes the analyses used to evaluate the cost impacts of Tier III NOx
requirements combined with the use of lower sulfur fuel on vessels operating within the U.S.
portion of the proposed EGA; including estimates of lower sulfur fuel production costs, engine
and vessel hardware costs, and the associated differential operating costs. This chapter also
presents cost per ton estimates for EGA-based NOx and fuel sulfur standards and compares these
costs with established land-based control programs.
The costs presented here are based on the application of EGA controls and compliance
with EGA standards in 2020. Consistent with the presentation of the inventory (Chapter 2) and
the benefits (Chapter 4), the estimated costs are reported for the year 2020. In this year, only
new vessels will incur hardware costs, while all vessels (new or existing) will incur additional
operating costs in the proposed EGA (e.g. the use of urea on an SCR equipped vessel built in or
prior to 2020). A separate analysis is provided for the benefit of ship owners, which presents the
estimated one-time hardware costs that may be incurred by some existing vessels to
accommodate the use of lower sulfur fuel. These costs are expected to be incurred by 2015 when
the fuel sulfur standards take effect, and are not included in the 2020 total. All costs are
presented in terms of 2006 U.S. dollars.
5.1 Fuel Production Costs
This section presents our analysis of the impact of the proposed EGA on marine fuel
costs. Distillate fuel will likely be needed to meet the 0.1 percent fuel sulfur limit, beginning in
2015, for operation in ECAs.A As such, the primary cost of the fuel sulfur limit will be that
associated with switching from heavy fuel oil to higher-cost distillate fuel, when operating in the
EGA. Some engines already operate on distillate fuel and would not be affected by fuel
switching costs. Distillate fuel costs may be affected by the need to further refine the distillate
fuel to meet the 0.1 percent fuel sulfur limit. To investigate these effects, studies were
performed on the impact of a U.S./Canada EGA on global fuel production and costs. These
studies, which are summarized below, include economic modeling to project bunker fuel demand
and refinery modeling to assess the impact of a U.S./Canada EGA on fuel costs.
A As an alternative, an exhaust gas cleaning device (scrubber) may be used. This analysis does not include the
effect on distillate fuel demand of this alternative approach. It is expected that scrubbers would only be used in the
case where the operator determines that the use of a scrubber would result in a cost savings relative to using
distillate fuel. Therefore we are only estimating the cost of compliance using distillate fuel here as we believe this is
the most likely approach.
5-1
-------
5.1.1 Bunker Fuel Demand Modeling
To assess the affect of an EGA on the refining industry, we needed to first understand and
characterize the fuels market and more specifically the demand for the affected marine fuels both
currently and in the future. Research Triangle Institute (RTI) was contracted to conduct a fuels
study using an activity-based economic approach.1 The RTI study established baseline bunker
fuel demand, projected a growth rate for bunker fuel demand, and established future bunker fuel
demand volumes. The basis for this work was the Global Insights economic model which
projects international trade for different categories of commodities. Demand for marine fuels
was derived from the demand of transportation of various types of cargoes by ship, which, in
turn, was derived from the demand for commodities produced in one region of the world and
consumed in another. The flow of commodities was matched with typical vessels for that trade
(characterized according to size, engine power, age, specific fuel consumption, and engine load
factors). Typical voyage parameters were assigned, including average ship speed, round trip
mileage, tonnes of cargo shipped, and days in port. Fuel consumption for each trade route and
commodity type was thus a function of commodity projections, ship characteristics, and voyage
characteristics.
The bunker demand model included operation off the coasts of the contiguous United
States and southeastern Alaska. The bunker demand volumes for this modeling in the Canadian
portion of the EGA was based on fuel consumed by ships en route to and from Canadian ports
based on estimates from Environment Canada.
These affected fuel volumes which are used in the WORLD model described below, are
slightly higher than what we now estimate for the proposed EGA. This difference is because the
RTI evaluation of affected fuel volumes was performed before the EGA was defined and was
performed independently of the emission inventory modeling described in Chapter 2. However,
we believe it is reasonable to use the fuel cost increases, on a per-tonne basis, from the WORLD
modeling to estimate the impact of the proposed EGA. In earlier work,2 EnSys modeled a
number of fuel control scenarios where the volume of affected fuel was adjusted to represent 1)
different EGAs or 2) various penetration scenarios of exhaust gas scrubbers (as an alternative to
fuel switching). This work suggests that the differences in fuel volume between these scenarios
have only a small effect on fuel cost. Although this earlier work was based on the older crude oil
and refinery costs used in the expert group study, it is sufficient for observing the sensitivity of
fuel cost increases to small changes (on a global scale) in affected fuel volume. In addition, the
larger affected fuel volume, used in the WORLD modeling, directionally increases the projected
fuel cost increases, and therefore allows for a conservative analysis.
5.1.2 Bunker Fuel Cost Modeling
5.1.2.1 Methodology
To assess the impacts of the proposed EGA on fuel costs, the World Oil Refining
Logistics and Demand (WORLD) model was run by Ensys Energy & Systems, the owner and
developer of the refinery model. The WORLD model is the only such model currently
developed for this purpose, and was developed by a team of international petroleum consultants.
It has been widely used by industries, government agencies, and OPEC over the past 13 years,
5-2
-------
including the Cross Government/Industry Scientific Group of Experts, established to evaluate the
effects of the different fuel options proposed under the revision of MARPOL Annex VI.3 The
model incorporates crude sources, global regions, refinery operations, and world economics.
The results of the WORLD model have been shown to be comparable to other independent
predictions of global fuel, air pollutant emissions and economic predictions.
WORLD is a comprehensive, bottom-up model of the global oil downstream that
includes crude and noncrude supplies; refining operations and investments; crude, products, and
intermediates trading and transport; and product blending/quality and demand. Its detailed
simulations are capable of estimating how the global system can be expected to operate under a
wide range of different circumstances, generating model outputs such as price effects and
projections of refinery operations and investments.
5.1.2.2 Assessment of the Impact of Marine Fuel Standards
During the development of the amendments to MARPOL Annex VI, a Cross
Government/Industry Scientific Group of Experts was established, by IMO, to evaluate the
effects of the different fuel options that were under consideration at the time. This expert group
engaged the services of EnSys to assess the impact of these fuel options using the WORLD
model. The final report from this study presents great detail on the capabilities of the WORLD
model and provides support for why the WORLD model was chosen as the appropriate tool for
modeling the economic impacts of the different fuel options.4 The following description of the
WORLD model is taken from the expert group study:
WORLD is a linear programming model that simulates the activities and economics of
the world regional petroleum industry against short, medium or long term horizons. It models
and captures the interactions between:
• crude supply;
• non-crudes supply: Natural gas Liquids (NGLs), merchant MTBE, biofuels, petrochemical
returns, Gas To Liquid fuels (GTLs), Coal to Liquid fuels (CTLs);
• refining operations;
• refining investment;
• transportation of crudes, products and intermediates;
• product blending/quality;
• product demand; and
• market economics and pricing.
The model includes a database representing over 180 world crude oils and holds detailed,
tested, state-of-the-art representation of fifty-plus refinery processes. These representations
include energy requirements based on today's construction standards for new refinery units. It
allows for advanced representation of processes for reformulated, ultra-lower sulfur/aromatics
fuels and was extended for detailed modeling of marine fuels for the aforementioned EPA and
API studies. The model contains detailed representations of the blending and key quality
specifications for over 50 different products spread across the product spectrum and including
5-3
-------
multiple grades of gasolines, diesel fuels/gasoils (marine and non-marine) and residual fuels
(marine and non-marine).
The refining industry is a co-product industry. This means that changes in production of
one product also affect production volume and/or production costs of other products. As
necessary, the model will adjust refinery throughputs and operations, crude and product trade
patterns to ensure that a specified product demand slate is met, without surplus or deficit of any
product.
To evaluate the impact of changes to marine fuels specifications as a result of any of the
options under consideration, the model is run with a future demand scenario for all products. The
first run, the base case, assumes marine fuels in line the current Annex VI regulation. The second
run is done with marine fuel specifications in line with the option under consideration. Both runs
are optimized independently. Since the only thing that is altered between the cases is the change
in the projected marine fuels regulation, the difference between both cases is therefore a true
assessment of the actual cost and other implications of the change to the marine fuels
requirements under consideration. Thus, the incremental refining investment costs, incremental
marine fuel costs and incremental refinery/net C02 emissions are all directly attributable to - and
must be allocated to - the change in regulation.
Prior to the expert group study, EnSys made updates to the WORLD model to be able to
perform the analysis of the impacts of different marine fuel options. As part of this effort, the
refinery data, capacity additions, technology assumptions, and costs were reviewed. EnSys
reviewed relevant regulations to ensure that the WORLD model was correctly positioned to
undertake future analyses of marine fuels EGAs. In developing these updates, a number of
issues had to be considered:
• the costs of refining, including the capital expenditures required to reduce bunker fuel sulfur
content and the potential for process technology improvements;
• likely market reactions to increased bunker fuel costs, such as fuel grade availability, impacts
on the overall transportation fuels balance, and competition with land-based diesel and
residual fuels for feedstocks that can upgrade fuels;
• the effects of emissions trading; and
• the potential for low- and high-sulfur grade bunker sources and consumption to partially shift
location depending on supply volume, potential, and economics.
The analytical system thus had to be set up to allow for alternative compliance scenarios,
particularly with regard to (a) adequately differentiating bunker fuel grades; (b) allowing for
differing degrees to which the EGA or other standards in a region were presumed to be met by
bunker fuel sulfur reductions, rather than by other means such as scrubbing or emissions trading;
and (c) allowing for all residual fuel bunker demand to be reallocated to marine diesel. Beyond
any international specifications, the analytical system needed to be able to accommodate future
consideration of regional, national, and local specifications.
5-4
-------
The primary approach taken to manage these issues was to:
• expand the number of bunker grades in the model to three distillates and four residual
grades;B
• allow for variation where necessary in (regional) sulfur standards on specific bunker grades;
and
• enable residual bunker demand to be switched to marine diesel.
Other updates to the WORLD model included product transportation matrices covering
tanker, interregional pipeline, and minor modes were expanded to embody the additional
distillate and residual bunker grades, adjustments to the yield patterns of the residuum
desulfurization, and blocking of paraffinic streams from residual fuel blends. The details of
compliance in any particular region must be estimated external to the main WORLD model. As
discussed above, we provided our estimates of affected fuel volumes to Ensys.
5.1.2.3 Updates for ECA Analysis
To determine the impact of the proposed ECA, the WORLD model was employed using
the same basic approach as for the IMO expert group study. Modeling was performed for 2020
in which the control case included a fuel sulfur level of 0.1 percent in the U.S. and Canadian
EEZs.5 The baseline case was modeled as "business as usual" in which ships continue to use the
same fuel as today. This approach was used for two primary reasons. First, significant emission
benefits are expected in an ECA, beginning in 2015, due to the use of 0.1 percent sulfur fuel.
These benefits, and costs, would be much higher in the early years of the program before the 0.5
percent fuel sulfur global standard goes into effect. By modeling this scenario, we are able to
observe the impact of the proposed ECA in these early years. Second, there is no guarantee that
the global 0.5 percent fuel sulfur standards will begin in 2020. The global standard may be
delayed until 2025, subject to a fuel availability review in 2018. In addition, the 3.5 percent fuel
sulfur global standard, which begins in 2012, is higher than the current residual fuel sulfur
average of 2.7 percent.
In the modeling for the expert group study, crude oil prices were based on projections
released by the U.S. Energy Information Administration (ElA) in 2006.6 Since that time, oil
prices have fluctuated greatly. Using new information, EIA has updated its projections of oil
price for 2020.7'8 In response to this real-world effect, the ECA modeling was conducted using
the updated oil price estimates. Specifically, we used a crude oil price of $51.55 for the reference
case, and $88.14/bbl for the high price case, both expressed in real (2006) dollars. These crude
oil prices were input to the WORLD model which then computed residual and distillate marine
oil prices for 2020. The net refinery capital impacts are imputed based on the differences in the
costs to the refining industry that occur between the Base Cases and ECA cases in 2020. The
B Specifically, the following seven grades were implemented: MGO, plus distinct high- and low-sulfur blends for
MDO and the main residual bunker grades IFO 180 and IFO 380. The latest international specifications applying to
these fuels were used, as were tighter sulfur standards for the low-sulfur grades applicable in SEC As.
5-5
-------
incremental global refining investment over the Base Case is projected to cost an additional
$3.83 billion, with $1.48 billion being used for debottlenecking projects and $1.96 billion used
for new units. For the high priced crude case, the incremental capital investments for an EGA
is $3.44 billion over the base case, with new units accounting for $2.49 billion while
debottlenecking costs are $0.72 billion. For both of the crude oil price cases, refinery
investments represent a marginal increase of about 2 percent over the corresponding total base
case investments required in 2020. Additionally, the majority of these EGA investments occur
in the U.S./Canada refining regions, though smaller amounts also occur in other world regions.
In addition to increased oil price estimates, the updated model accounts for increases in natural
gas costs, capital costs for refinery upgrades, and product distribution costs.
5.1.3 Results of Fuel Cost Study
5.1.3.1 Incremental Refinery Capital Investments Associated with Desulfurization
5.1.3.1.1 General Overview
The primary refining cost of desulfurization is associated with converting IFO bunker oil
into a distillate fuel with a DMA specification. The other significant refining costs are those
related to desulfurizing distillate stocks. The bulk of the refinery investments occur in regions
located outside of the U.S. and Canada, because capital investments in these regions are
approximately 9 and 23 percent of the overall capital for the reference and high priced crude
cases, respectively. Table 5.1-1 summarizes the overall capital investments made for both
conversion of IFO bunker oil into distillate as well as desulfurization in refineries in the various
U.S. regions (East Coast, West Coast and Gulf Coast) and overseas. These cost estimates are
based on the WORLD modeling results.
Table 5.1-1 Incremental Refinery Capital Investment Made in 2020 (2006 dollars)
USEC
USGCCE
USWCCW
Refinery Investments Total USA+Canada
Refinery Investments Total Other Regions
Total World
REFINERY INVESTMENTS ($ BILLION)
Base Case
$52/bbl
Crude
1.4
14.5
1.4
17.3
85.2
102.5
NAECA
$52/bbl
Crude
1.2
14.8
1.6
17.6
88.1
105.7
Delta
-0.2
0.3
0.2
0.3
2.9
3.2
Base Case
$88/bbl
Crude
1.0
26.2
1.4
28.6
110.5
139.1
NAECA
$88/bbl
Crude
0.9
27.3
1.5
29.8
115.0
144.8
Delta
-0.1
1.2
0.2
1.3
4.4
5.7
Type of Modification
Debottleneck
Major New Units
Total World
0.7
97.8
102.5
0.7
100.8
105.7
0.0
3.0
3.2
1.4
132.1
139.1
1.4
138.0
144.8
0.0
6.0
5.7
Note: USEC is United States East Coast, USGCCE is United States Gulf Coast and Eastern Canada, USWCCW is
United States West Coast and Western Canada, $Bn is Billion U.S. Dollars. The results presented are investments
made in 2020 to add new refinery processing capacity to what exists in the 2008 base case plus known projects.
5-6
-------
Refinery investments in North America, Greater Caribbean and South American regions
account for greater than half of all investments for the reference case, while investments made in
China and Middle Eastern Gulf regions account for close to 40 percent of remaining investments.
This accounts for greater than 90 percent of investments for the reference case. For the high
priced EGA case, investments in U.S., Canada, Greater Caribbean and South American refiner
regions again account for greater than half of all investments made, while European north and
China regions account for greater than 44 percent of the remaining investments. Table 5.1-2
summarizes overall incremental investments made in all world refining regions for the reference
and high priced EGA case.
Table 5.1-2 World Region Refining Investments for ECA Made in 2020
USEC
USGICE
USWCCW
GrtCAR
SthAM
AfWest
AfN-EM
Af-E-S
EUR-No
EUR-So
EUR-Ea
CaspRg
RusFSU
MEGulf
Paclnd
PacHi
China
RoAsia
Total
REFERENCE CASE
Capital, $
Billion
-0.167
0.277
0.176
0.253
0.810
0.004
0.143
0.007
0.011
-0.006
0.021
0.157
0.185
0.754
-0.115
0.177
0.490
0.018
3.20
% of Capital
-5.2%
8.7%
5.5%
7.9%
25.4%
0.1%
4.5%
0.2%
0.4%
-0.2%
0.7%
4.9%
5.8%
23.6%
-3.6%
5.5%
15.3%
0.6%
100.0%
HIGH PRICED CASE
Capital, $ Billion
-0.095
1.159
0.224
0.828
0.870
0.002
-0.006
0.006
1.239
-0.035
-0.014
-0.001
0.036
0.119
0.069
0.000
1.305
-0.002
5.70
% of Capital
-1.7%
20.3%
3.9%
14.5%
15.3%
0.0%
-0.1%
0.1%
21.7%
-0.6%
-0.2%
0.0%
0.6%
2.1%
1.2%
0.0%
22.9%
0.0%
100.0%
Note: USEC = US East Coast, USGICE= US Gulf Coast, Interior & Canada East, USWCCW= US
West Coast & Canada West, GrtCAR= Greater Caribbean, SthAM= South America, AfWest=African
West, AFN- EM= North Africa/Eastern Mediterranean, AF-E-S=Africa East and South, Eur-
No=Europe North, EUR-So= Europe South, EUR-EA= Europe East, CaspRg= Caspian Region,
RusFSU= Russia & Other Former Soviet Union, MEGulf= Middle East Gulf, Pac Ind= Pacific
Industrialized, PacHi= Pacific High Growth / Industrialising, RoAsia= Rest of Asia
5-7
-------
5.1.3.1.2 Processing of Residual Stocks
IFO bunker grades are primarily comprised of residual stocks, such as Vacuum
Residuals, Atmospheric Residuals, Visbreaker Residuals, and Fluidized Catalytic Cracking
(FCC) clarified oil. These fuels also contain distillates that are added as cutter stocks, such as
Light Cycle Oil (LCO), Vacuum Gas Oils (VGO), and kerosenes. As such, only the residual fuel
blendstocks in IFO bunkers would need to be replaced or converted into distillate volumes to
provide for additional lower sulfur distillate marine fuel. For converting residuals to distillates,
refiners use two process technologies: Coking Units (Cokers) and Residual Hydrocrackers.
Coking units are used to convert the poorer quality residual feedstocks in IFO bunkers,
such as Vacuum residuals. The coking units crack these residuals into distillates, using heat and
residence time to make the conversion. The process produces petroleum coke and off gas as
byproducts. Residual hydrocrackers are used to convert low and medium sulfur residual streams
into distillates. Residual hydrocracking uses fluidized catalyst, heat and hydrogen to
catalytically convert residual feedstocks into distillates and other light fuel products. The
hydrocracking process upgrades low value residual stocks into high value distillate transportation
fuels consuming large amounts of hydrogen.
For processing of residual blendstocks, vacuum tower distillation capacity is added to
extract gas oils blendstocks that exist in residuals fuels used in current IFO bunker grades. The
extracted gas oils are further processed in either distillate hydrotreaters or gas oil hydrocrackers
to produce a distillate fuel that would meet a 0.1 percent fuel sulfur limit. The use of additional
vacuum towers capacity minimizes the volume of residual stocks which lowers processing costs,
as less volume of fuel is processed in high cost residual coking and residual hydrocracker
processes.
5.1.3.1.3 Distillate Stocks Processing
Conventional distillate hydrotreating technology is used to lower the sulfur levels of high
sulfur distillate stocks. This technology removes sulfur compounds from distillate stocks using
catalyst, heat and hydrogen. Since the EGA sulfur standard is 0.1 percent, conventional distillate
hydrotreating would likely be the technology chosen by refiners to make this distillate, rather
than the ultra lower sulfur technology that is used to remove sulfur to levels below 15 ppm
(0.0015 percent). Conventional distillate hydrotreating refers to the design and conditions in the
process, such as catalyst type, catalyst volume, reactor pressure, feed and reactor flow scheme
used to lower sulfur levels to 0.05 percent or higher.
Although the cutter stocks in IFO bunkers are distillate fuels, they would need to be
desulfurized because the 0.1 percent sulfur limit for the EGA is lower than the nominal sulfur
levels for these blendstocks under the "business as usual" projections. The sulfur levels of
distillate used directly as bunker fuel (MDO and MGO), are greater than 1,000 ppm, and thus
would also need to be treated. Therefore, in addition to converting residuals to distillate fuels,
existing distillates used as bunker fuel in MDO, MGO and IFO would also need to be
hydrotreated. More distillate hydrotreating capacity would be required to lower the sulfur
content of incremental distillate produced from cokers and residual hydrocrackers that do not
meet lower sulfur marine fuel standards.
5-8
-------
For distillate stocks that are highly aromatic and high in sulfur, the use of technology for
hydrocracking lower sulfur gas oil is used to convert these blendstocks into No 2. grade diesel
streams. Gas oil hydrocracking is a high volume gain process which produces diesel blendstocks
that typically meet EGA sulfur standards, eliminating the need for further processing in
hydrotreaters.
5.1.3.1.4 Supportive Processes
The increase in hydrotreating and hydrocracking requires new hydrogen and sulfur plant
capacity. Extra hydrogen is required to react with and remove sulfur compounds in refinery
hydrotreating process. It is also needed to improve the hydrogen to carbon ratio of products
made from converting IFO blend components to distillates, via processing in cokers and
hydrocrackers.
5.1.3.2 Capacity and Throughput Changes for the Reference Case
The WORLD model used a total of 140 thousand barrels per stream day (KBPSD) of
coking capacity to convert residual stocks to distillates. Of this amount, 110 KBPSD is existing
spare or "slack" capacity available in U.S. and Canada refiner regions. This capacity is available
based on projections that refiners add excess coking capacity in the base case. The remaining
balance of coking capacity, or 30 KBPSD, is new capacity added to refiner regions outside of
United States and Canada. In addition to utilizing more coking capacity, the WORLD model
also increased residual hydrocracking capacity by 50 KBPSD to convert residual stocks into
distillates. These hydrocrackers were added to refiner regions located outside of United States
and Canada. Overall, considering the use of cokers and residual hydrocrackers, the total refiner
process capacity is 190 KBPSD for residual stocks processing, mirroring the amount needed to
process the residual volumes contained in IFO 180 and IFO 380 bunker grades. To remove any
gas oils in residual blendstocks such as atmospheric and vacuum tower residuals, the model
utilized 60 KBPSD of existing vacuum tower capacity, 50 KBPSD in U.S. and Canada and 10
KBPSD in other refiner regions.
Crude throughput is increased by 54 KBPSD, primarily to account for increased energy
usage in refinery processes such as hydro crackers and hydrotreaters. Crude throughput is also
increased to offset liquid volume loss from residual stocks that are converted to petroleum coke
in coking units. Table 5.1-3 summarizes overall crude and non crude throughputs for the base
and EGA cases in units of million barrels per stream day (MMBPD).
Table 5.1-3 Refiner Crude and Non Crude Throughputs
Crude Throughput
Non Crude Supply
NGL ETHANE
NGLs C3+
PETCHEM RETURNS
MMBPD
MMBPD
MMBPD
MMBPD
REFERENCE
BASE CASE
86.7
1.7
6.3
1.0
REFERENCE
EGA
CASE
86.7
1.7
6.3
1.0
DELTA
0.1
0.0
0.0
0.0
HIGH
BASE
CASE
75.6
1.7
6.1
0.8
HIGH
EGA
CASE
75.6
1.7
6.1
0.8
DELTA
0.0
0.0
0.0
0.0
5-9
-------
BIOMASS
METHANOL (EXNGS)
GTL LIQUIDS (EXNGS)
CTL LIQUIDS (EX COAL)
HYDROGEN (EXNGS)
Total Non Crude Supply
TOTAL Supply
MMBPD
MMBPD
MMBPD
MMBPD
MMBPD
MMBPD
MMBPD
1.5
0.1
0.3
0.5
1.0
12.3
99.3
1.5
0.1
0.3
0.5
1.0
12.3
99.4
0.0
0.0
0.0
0.0
0.0
0.0
0.1
3.0
0.1
0.6
0.8
0.8
14.0
90.2
3.0
0.1
0.6
0.8
0.9
14.0
90.3
0.0
0.0
0.0
0.0
0.1
0.0
0.1
The model added 70 KBPSD of new ultra lower sulfur gas oil hydrocracking capacity in
refiner regions outside of the U.S. and Canada. The distillate produced from these units has a sulfur
content low enough to meet EGA standards and therefore does not require further processing in
hydrotreaters. The model also reduced throughput by 40 KBPSD in existing base case capacity for
Conventional Gas Oil Hydrocrackers located in U.S. and Canada refiner regions.
The model added 160 KBPSD of new conventional distillate hydrotreating capacity, 140
KBPSD to U.S. and Canada refiner regions and 20 KBPSD in refining regions in other areas of the
world. In addition to new units, the model used 150 KBPSD of "slack" distillate conventional
hydrotreating capacity, 90 KBPSD of this located in U.S. and Canada and 60 KBPSD in other world
refiner regions. Considering this, the total net use of conventional distillate hydrotreating for the
reference case is 310 KBPSD above the base case, mirroring incremental demand of lower sulfur
distillate for EGA. The model used 70 KBPSD of slack capacity for vacuum gas oil/residual
hydrotreating in addition to distillate hydrotreating. Of this amount, 40 KBPSD is in U.S. and
Canada and 30 KBPSD in other world refiner regions.
The increased hydrotreating and hydrocracking capacity requires new hydrogen and sulfur
plant capacity and was added to refiner regions that use more distillate hydrotreating and
hydrocracking. Other minor refinery process modifications were required by the model in 2020,
although these were not substantial (see Table 5.1-4).
Table 5.1-4 Refinery Secondary Processing Capacity Additions in 2020 Reference Case (Million barrels per
stream day)
Total Additions
Over Base
Total Crude
Capacity Used
2020
Vacuum
Distillation
Coking
Catalytic
Cracking
Hydro-Cracking
(TOTAL)
- Gasoil
Conventional
- Gasoil ULS
- Resid LS
USE OF BASE CAPACITY
US/CAN
0.00
0.02
0.05
0.11
(0.07)
(0.04)
(0.04)
0.00
0.00
Rest of
World
0.05
0.04
0.01
0.00
0.01
0.00
0.00
0.00
0.00
Total
0.05
0.05
0.06
0.12
(0.06)
(0.04)
(0.04)
0.00
0.00
NEW CAPACITY
US/CAN
0.00
0.02
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Rest of
World
0.05
0.04
(0.02)
0.02
(0.01)
0.12
0.00
0.07
0.01
Total
0.05
0.05
(0.02)
0.02
(0.01)
0.12
0.00
0.07
0.01
BASE PLUS NEW CAPACITY
US/CAN
0.00
0.017
0.05
0.11
(0.07)
(0.04)
(0.04)
0.00
0.00
Rest of
World
0.05
0.037
(0.01)
0.03
0.00
0.12
0.00
0.07
0.01
Total
0.05
0.054
0.04
0.14
(0.07)
0.08
(0.04)
0.07
0.01
5-10
-------
- Resid MS
Catalytic
Reforming with
Revamp
Hydro treating
(Total)
- Gasoline - ULS
Distillate
-New Conv/LS
- VGO/Resid
Hydrogen,
(MMSCFD)
Sulfur Plant,
(TPD)
0.00
0.01
0.13
0.00
0.09
0.04
0
500
0.00
0.00
0.08
(0.00)
0.06
0.03
70
500
0.00
0.02
0.21
(0.00)
0.15
0.06
70
1000
0.00
0.00
0.11
(0.03)
0.14
0.00
8
10
0.04
0.07
0.05
0.03
0.02
0.00
211
130
0.04
0.07
0.17
(0.00)
0.16
0.00
218
140
0.00
0.01
0.24
(0.03)
0.23
0.04
8
510
0.04
0.07
0.14
0.02
0.08
0.03
280
630
0.04
0.08
0.37
(0.01)
0.31
0.07
288
1140
While coking and hydrocracking (residual and gas oil) processes primarily produce
distillates, to a lesser extent, some low octane gasoline blendstocks are also manufactured, requiring
refiners to install additional catalytic reforming unit capacity. As such, in the U.S. and Canada
regions approximately 10 KPBSD of existing spare catalytic reforming capacity is used while
approximately 70 BPSD of new catalytic reforming capacity is added to other WORLD refiner
regions that added cokers and hydrocrackers.
5.1.3.3 Capacity and Throughput Changes for the High Price Crude Oil Case
For the high priced case, the high cost of crude and high capital costs for processing units
push the model to reduce installation of new processing units. The price of natural gas is also
reduced relative to the price of crude which induces the model to use more natural gas and
reduce the use of crude. Under these conditions, the model uses less crude, more natural gas and
installs less capital for refinery processing units. As a result, the model favors the use of more
hydrocracking processing which adds hydrogen (made from natural gas) to residual and gas oils,
producing lower sulfur distillates stocks that do not require further processing in hydrotreaters.
The model also uses more synthetic crudes and less heavy sour crudes, which reduce the
amounts of residual stocks that need upgrading.
Crude throughput is increased by 29 KBPSD, which is less than the reference case, as the
model preferentially uses natural gas over crude and reduces the use of cokers and hydrotreating.
Table 5.1-5 shows crude and non crude inputs for the high priced case.
The WORLD model used a total of 80 KBPSD of "slack" coking capacity to convert residual
stocks to distillates. Of this amount, 70 KBPSD was used in the U.S. and Canada regions and 10
KBPSD in regions in other areas of the world. The model also added 80 KBPSD of new low and
medium sulfur residual hydrocracking capacity to convert residual stocks into distillates—20 KBPSD
in the U.S. and Canada and 60 KBPSD in other world refiner regions. Overall, considering the use
of cokers and residual hydrocrackers, the total refiner process capacity for residual stocks processing
for use in the EGA is 160 KBPSD for the high priced case.
To extract gas oils from residual blendstocks, the model utilized 90 KBPSD of existing
vacuum tower capacity—80 KBPSD in the U.S. and Canada and 10 KBPSD on other refiner regions.
In addition, the model added 120 KBPSD of new ultra lower sulfur gas oil hydrocracking capacity in
5-11
-------
refiner regions outside of the U.S. and Canada. The distillate fuel produced from these units meet
EGA sulfur standards. The model also used 30 KBPSD of slack capacity in the U.S. and Canada
refiner regions for hydrocracking of conventional gas oil.
The model added 40 KBPSD of new conventional distillate hydrotreating capacity to the U.S.
and Canada refiner regions and 20 KBPSD of new capacity to refining regions in other areas of the
world. While the model also used 40 KBPSD of "slack" conventional distillate hydrotreating
capacity in the U.S. and Canada, other world refiner regions decreased use of base case or slack
capacity by 80 KBPSD. Considering the use of the new and slack capacity, a total net use of
capacity is 20 KBPSD of conventional distillate hydrotreating capacity. The model also used 60
KBPSD of existing slack capacity for vacuum gas oil/residual distillate hydrotreaters, with 20
KBPSD used in the U.S. and Canada refiner regions and 40 KBPSD in other world refining regions.
The use of additional hydrocracking and hydrotreater capacity requires installation of new
hydrogen plant capacity. New sulfur plant capacity is required in refiner regions to process the
offgas produced from incremental use of hydro cracking and hydrotreating (see Table 5.1-5 below).
Table 5.1-5 Refinery Secondary Processing Capacity Additions in 2020 High Priced Case
(Million barrels per stream day)
Total Additions
Over Base Case
Total Crude
Capacity Used in
2020
Vacuum Distillation
Coking
Catalytic Cracking
Hydro-Cracking
(Total)
- Gasoil
Conventional
- Gasoil ULS
- Resid LS
- Resid MS
Catalytic Reforming
with Revamp
Hydrotreating
(Total)
- Gasoline - ULS
Distillate
-New Conv/LS
- VGO/Resid
Hydrogen,
(MMSCFD)
Sulfur Plant, (TPD)
USE OF BASE CAPACITY
US/CAN
0.00
0.05
0.08
0.07
(0.03)
0.03
0.03
0.00
0.00
0.00
0.00
0.06
0.00
0.04
0.02
0
580
Rest of
World
(0.05)
(0.02)
0.10
0.01
(0.05)
0.00
0.00
0.00
0.00
0.00
0.02
(0.04)
0.00
(0.08)
0.03
0
300
Total
(0.05)
0.03
0.18
0.08
(0.09)
0.03
0.03
0.00
0.00
0.00
0.02
0.02
0.00
(0.03)
0.05
0
880
NEW CAPACITY
US/CAN
0.00
0.05
0.00
0.00
0.00
0.02
0.00
0.00
0.02
0.00
(0.05)
0.04
0.00
0.04
0.00
243
0
Rest of
World
(0.05)
(0.02)
0.00
(0.00)
0.00
0.18
0.00
0.12
0.03
0.03
0.02
0.02
(0.01)
0.02
0.00
325
120
Total
(0.05)
0.03
0.00
(0.00)
0.00
0.20
0.00
0.12
0.05
0.03
(0.03)
0.06
(0.01)
0.06
0.00
568
120
BASE PLUS NEW CAPACITY
US/CAN
0.00
0.054
0.08
0.07
(0.03)
0.05
0.03
0.00
0.02
0.00
(0.05)
0.11
0.00
0.08
0.02
243
580
Rest of
World
(0.05)
(0.024)
0.10
0.00
(0.05)
0.18
0.00
0.12
0.03
0.03
0.04
(0.03)
(0.01)
(0.06)
0.04
325
420
Total
(0.05)
0.029
0.18
0.08
(0.09)
0.22
0.03
0.12
0.05
0.03
(0.00)
0.08
(0.01)
0.02
0.06
568
1000
5-12
-------
5.1.3.4 Overall Increases Due to Fuel Switching and Desulfurization
Global fuel use in 2020 by international shipping is projected to be 500 million tonnes/yr.
The main energy content effects of bunker grade shifts were captured in the WORLD modeling
by altering the volume demand and, at the same time, consistency was maintained between the
bunker demand figures in tonnes and in barrels. The result was that partial or total conversion of
IFO to distillate was projected to lead to a reduction in the total global tonnes of bunker fuel
required but also led to a projected increase in the barrels required. These effects are evident in
the WORLD case results. Based on the WORLD modeling, the volume of marine fuel affected
by an EGA encompassing the U.S.C and Canadian EEZs would be about 4 percent of total world
residual volume. As would be expected, since the shift in fuel volumes on a world scale is
relatively small, the WORLD model predicts the overall global impact of an EGA to also be
small.
There are two main components to projected increased marine fuel cost associated with
an EGA. The first component results from the shifting of operation on residual fuel to operation
on higher cost distillate fuel. This is the dominant cost component. The WORLD model
computed costs based on a split between the costs of residual and distillate fuels. However, there
is a small cost associated with desulfurizing the distillate to meet the 0.1 percent fuel sulfur
standard in the EGA. Based on the WORLD modeling, the average increase in costs associated
with switching from marine residual to distillate will be $145 per tonne.0 This is the cost
increase that will be borne by the shipping companies purchasing the fuel. Of this amount, $6
per tonne is the cost increase associated with distillate desulfurization. In other words, we
estimate a cost increase of $6/tonne for distillate fuel used in an EGA.
The above cost estimates are based on EIA's "reference case" projections for crude oil
price in 2020. We also performed a sensitivity analysis using EIA's "high price" scenario.
Under this scenario, the increase in fuel costs for switching from residual to distillate fuel is $237
per tonne. The associated increase in distillate fuel cost is $7 per tonne.
Table 5.1-6 summarizes the reference and high price fuel cost estimates with and without
an EGA. In the baseline case, fuel volumes for operation are 18% marine gas oil (MGO), 7%
marine diesel oil (MDO), and 75% IFO. In the proposed EGA, all fuel volumes are modeled as
MGO.
c For the contiguous U.S. and southeastern Alaska.
D Note that distillate fuel has a higher energy content, on a per tonne basis, than residual fuel. As such, there is an
offsetting cost savings, on a per tonne basis, for switching to distillate fuel. Based on a 5 percent higher energy
content for distillate, the net equivalent cost increase is estimated as $123 for each tonne of residual fuel that is being
replaced by distillate fuel ($200/tonne for the high price case).
5-13
-------
Table 5.1-6: Estimated Marine Fuel Costs
FUEL
MGO
MDO
IFO
UNITS
$/bbl
$/tonne
$/bbl
$/tonne
$/bbl
$/tonne
REFERENCE CASE
Baseline
$ 61.75
$ 464
$ 61.89
$ 458
$ 49.87
$ 322
EGA
$ 62.23
$ 468
$ 62.95
$ 466
$ 49.63
$ 321
HIGH PRICE CASE
Baseline
$ 102.70
$ 772
$ 102.38
$ 757
$ 83.14
$ 538
EGA
$ 103.03
$ 775
$ 103.70
$ 767
$ 82.52
$ 534
5.2 Engine and Vessel Costs
This section presents the analysis of the potential cost impacts that the proposed EGA
may have on new engines and vessels in the year 2020. To assess the potential cost impacts we
must understand: the makeup of the fleet of ships expected to visit the U.S. when these
requirements go into effect, the emission reduction technologies expected to be used, and the
cost of these technologies. The total engine and vessel costs associated with the proposed EGA
are based on a hardware cost per unit value applied to the number of affected vessels, and
include operational costs. This section discusses an overview of the methodology used to
develop a fleet of vessels expected to visit the U.S. portion of the proposed EGA, and presents
the methodology used to develop the hardware and operational costs.
5.2.1 Overview
There are a number of technologies available or expected to be available to meet Tier III
NOx standards and to accommodate the use of lower sulfur fuel. We expect that each
manufacturer will evaluate all possible technology avenues to determine how to best balance
their respective costs while ensuring compliance; however, this analysis makes certain
assumptions regarding how manufacturers will comply with the new emission and fuel
standards. We expect that selective catalytic reduction (SCR) is the emission control technology
most likely to be used to meet Tier III NOx standards in the proposed EGA; therefore, this cost
analysis is based on the use of SCR. With respect to fuel sulfur controls, we expect that
switching to lower sulfur fuel is the most likely method of control to meet the fuel sulfur
requirements when operating in the proposed EGA; therefore, this cost analysis is also based on
switching to the use of lower sulfur fuel.
While fuel sulfur standards will take effect in 2015 and Tier III NOx standards will take
effect in 2016, this cost analysis only presents the hardware and operating costs that are expected
to be incurred in 2020. In order to present the costs associated with the proposed EGA in 2020,
the hardware costs are only applied to new vessels in 2020 expected to visit U.S. ports, while
operating costs apply to all ships operating in the U.S. portion of the proposed EGA in 2020.
The cost estimates presented here assume that all of the hardware costs for new ships in 2020 are
due exclusively to this proposed EGA, and do not include an adjustment accounting for the
potential existence of other EGAs that these ships may visit which would also require Tier III
NOX controls and appropriate fuel sulfur controls. The operational costs described in this section
include those incurred in 2020 within the U.S. portion of the proposed EGA as a result of the use
5-14
-------
of urea on ships built as of 2016 equipped with SCR, and the differential costs associated with
the use of lower sulfur fuel.
5.2.2 Methodology
To project future costs, we needed to first develop estimates of the number of ships that
may visit the proposed EGA in 2020. To develop a future fleet, an approach similar to that used
to estimate the emissions inventory (see Chapter 2) was used here. Specifically, the same inputs
were used to develop a fleet of ships by ship type and engine type that may be expected to visit
U.S. ports in 2020. Next, we needed to develop the estimated technology hardware costs, and
sought input from the regulated community regarding the expected future costs of applying the
emission control technologies associated with the proposed EGA. The U.S. Government
contracted with IGF International to research the fixed and variable costs associated with the
technologies expected to be used to meet Tier III NOx and fuel sulfur standards.9 To assess the
cost of these new technologies, we developed a series of 'typical' engines with varying sizes and
characteristics (e.g. stroke, number of cylinders, etc.) that the technologies would be applied to
for the purposes of performing the cost research. The resulting cost estimates of applying
different technologies to these 'typical' engines formed the basis for this cost analysis; Table
5.2-1 lists these engine configurations.
Table 5.2-1 Average Engine Characteristics Used in this Study
ENGINE TYPE
Engine Power (kW)
Cylinders
Liters/cylinder
Engine Speed (rpm)
BSFC (g/kWh)
MEDIUM-SPEED
4,500
9
35
650
9,500
12
65
550
18,000
16
95
500
210
LOW-SPEED
8,500
6
380
130
15,000
8
650
110
48,000
12
1400
100
195
After initial cost estimates were developed, IGF provided surveys to several engine and
emission control technology manufacturers to determine the reasonableness of the approach and
cost estimates. Input received from those surveyed was incorporated into the final cost estimates
used in this analysis. The resulting costs for the 'typical' engines were plotted and a curve-fit
was used to determine an equation to estimate the dollar-per-kilowatt ($/kW) cost for each
technology. The hardware costs per vessel were based on average vessel characteristics (e.g.
engine type and propulsion power) determined for various ship types. The per vessel costs were
used with the estimated number of new vessels in 2020 expected to visit U.S. ports to evaluate
the total hardware costs associated with the U.S. portion of the proposed EGA. The total
operational costs were determined from the differential fuel cost estimates presented in Section
5.1 and the regional fuel consumption values presented in Chapter 2. For vessels equipped with
SCR, urea consumption is expected to be 7.5 percent of the fuel consumption.
Operating costs per vessel vary depending on what year the vessel was built, for example,
in 2020, vessels built as of 2016 will incur operating costs associated with the use of urea
necessary when using SCR as a Tier III NOx emission control technology, while vessels built
prior to 2016 will only incur operating costs associated with the differential cost of using of
5-15
-------
lower sulfur fuel. To develop the costs associated with the proposed EGA in 2020, an
approximation of the number of ships by age that may visit the proposed EGA in 2020 had to be
constructed. To develop this future 2020 fleet, the data from ship calls to U.S. ports in the
baseline year of 2002 were used to estimate how many ships would visit U.S. ports in 2020.E'10
5.2.2.1 2020 Fleet Development
The U.S. port data from 2002 used in the inventory port analysis and the regional growth
rates presented in Chapter 2 were used to estimate how many ships by ship type and engine type
may visit U.S. ports in the future. The ships that called on the U.S. in 2002 were cross
referenced with Lloyd's database using their IMO numbers to determine the propulsion power,
engine type, and ship type of each ship.11 This allowed for all ships without Category 3 engines
to be removed from the analysis. In order to separate slow speed engines (SSD) from medium
speed engines (MSD) where that information was not explicitly available, 2-stroke engines were
assumed to be SSD, and 4-stroke engines were assumed to be MSD. The research performed for
this cost analysis differentiated between SSD and MSD engines, and separate $/kW values were
developed for each of these engine types.
The ship type information gathered from this baseline data, for the purposes of both this
analysis and the inventory, was categorized into one of the following ship types: Auto Carrier,
Bulk Carrier, Container, General Cargo, Miscellaneous, Passenger, Refrigerated Cargo (Reefer),
Roll-On Roll-Off (RoRo), and Tankers. The 2002 baseline fleet was also used to develop
average ship characteristics shown in Table 5.2-2. These values were used to represent the
characteristics of new (and future existing) vessels for the purposes of this cost analysis.
The 2002 port call data were sorted by IMO number to determine the total number of
unique ships that visited all included U.S. ports in 2002. Table 5.2-3 shows the breakout by ship
type of these approximately 6,700 ships. Next, in order to be consistent with the inventory
analysis which presents growth rates by region, the original port call data was separated into the
same regions used by the inventory (South Pacific (SP), North Pacific (NP), East Coast (EC),
Gulf Coast (GC), Alaska East (AE), Alaska West (AW), Hawaii East (HE), and West Hawaii
(HW)). This was done by matching each port-of-call entry in the original port call data file with
the corresponding region containing that port as per the inventory analysis.12 This resulted in a
fleet of ships for each region, each with a unique IMO number as shown in Table 5.2-3.
E The 2002 U.S. ship call data used to determine the 2002 baseline fleet was also used to construct port inventories,
as discussed in the Emissions Inventory Chapter. As such, this fleet includes the same ports and limitations as the
inventory analysis (e.g. military vessels are excluded, as are ships powered by engines <30 L/cyl.)
5-16
-------
Table 5.2-2 Average Ship Characteristics used in this Cost Analysis
SHIP TYPE
Auto Carrier
Bulk Carrier
Container
General Cargo
Passenger
Reefer
RoRo
Tanker
Misc.
ENGINE
SPEED
Slow Speed
Medium Speed
Slow Speed
Medium Speed
Steam Turbine
Slow Speed
Medium Speed
Steam Turbine
Slow Speed
Medium Speed
Steam Turbine
Slow Speed
Medium Speed
Steam Turbine
Gas Turbine
Slow Speed
Medium Speed
Slow Speed
Medium Speed
Gas Turbine
Steam Turbine
Slow Speed
Medium Speed
Gas Turbine
Steam Turbine
Slow Speed
Medium Speed
Steam Turbine
AVERAGE
PROPULSION
POWER (KW)
11,000
9,600
8,400
6,300
6,400
27,000
14,000
21,000
7,700
5,200
18,000
24,000
24,000
27,000
44,000
10,000
7,400
16,000
8,600
47,000
22,000
9,800
6,700
7,600
21,000
4,700
9,400
13,000
AVERAGE
AUXILIARY
POWER (KW)
3,000
2,600
1,900
1,400
1,400
6,000
3,000
4,700
2,000
1,300
4,600
6,600
6,600
7,600
12,000
4,200
3,000
4,000
2,200
12,000
5,800
2,100
1,400
1,600
4,400
1,300
2,500
3,500
SERVICE
SPEED
(KNOTS)
19
17
15
14
15
22
19
21
15
15
21
210
20
19
24
20
18
18
16
24
25
15
15
15
18
14
13
21
AVERAGE
DWT
17,000
13,000
47,000
27,000
19,000
45,000
19,000
30,000
26,000
8,700
23,000
6,200
6,200
13,000
12,000
11,000
7,600
30,000
8,400
37,000
19,000
61,000
27,000
40,000
59,000
8,800
6,000
17,000
Some ships may have visited ports in more than one region which could result in an
overestimate of the hardware costs (which are applied to each unique vessel) if the number of
vessels in each region were grown, summed together and used for the total costs. To prevent
over-counting of vessels visiting U.S. ports, a factor was developed (see Equation 1) to account
for this overlap. The number of unique ships in each region (identified by unique IMO numbers)
was summed together to produce a total number of "unique" ships visiting all regions, this value
was reduced by the total number of actual unique ships that visited U.S. ports in 2002 (from the
original baseline data) to provide a factor representing the original number of unique ships
visiting U.S. ports. This factor was then applied to the vessel count in each region to provide a
regional total that would coincide with the baseline total, and eliminate the over-counting of
ships that had visited multiple regions.
5-17
-------
Equation 1 Regional Fleet Overlap Reduction Factor Example
# Unique Auto Carriers in Total Port Call Data ,., . , ,, . „ . , . „
= %_ Actual _ Unique _ Re gional_ Auto_ Carriers
2^ Unique _ Auto_ Carriers _ by _ Re gion
For example, a total of 300 unique auto carriers visited all included U.S. ports in 2002,
yet when looking at unique ships on a regional basis and totaling all regions, 650 auto carriers
appeared to visit. This implied that only 46 percent of the regional auto carriers were "unique"
and that the additional 350 auto carriers were ships that had visited multiple regions. Therefore,
only 46 percent of all auto carriers within each regional fleet were assumed to be "unique." The
growth rates were only applied to this corrected count of "unique" ships in each region to
estimate the regional fleet makeup in future years.
Table 5.2-3 2002 Baseline Fleet of Ships and Regional Overlap Factor
SHIP TYPE
Auto Carrier
Bulk
Container
Gen. Cargo
Misc
Pass
Reefer
RoRo
Tanker
Total
TOTAL UNIQUE
SHIP VISITS TO
U.S. PORTS IN
2020
300
2,500
1,000
980
24
110
280
120
1,400
6,700
REGIONAL
UNIQUE SHIPS
VISITING U.S.
PORTS IN 2020
650
3,600
1,600
1,700
50
200
400
200
2,700
11,000
REGIONAL
OVERLAP
FACTOR
46%
68%
63%
57%
49%
57%
71%
58%
52%
62%
Within each region, the ship types were further broken down by engine type. The unique
ship fleet within each region was then grown by ship type and engine type using the appropriate
growth rate to estimate the makeup of the future fleet in 2020. Table 5.2-4 shows the estimated
2020 fleet of ships expected to visit U.S. ports.
Table 5.2-4 Estimated 2020 Fleet by Ship Type and Engine Type
SHIP TYPE
Auto Carrier
Bulk Carrier
Container
ENGINE TYPE
SSD
MSD
SSD
MSD
ST
SSD
MSD
ST
NUMBEROF
NEW VESSELS
45
4
440
8
3
210
8
9
NUMBEROF
EXISTING VESSELS
570
55
5500
110
21
2600
95
72
5-18
-------
SHIP TYPE
General Cargo
Passenger
Reefer
RoRo
Tanker
Misc.
ENGINE TYPE
SSD
MSD
ST
SSD
MSD
ST
GT
SSD
MSD
SSD
MSD
GT
ST
SSD
MSD
GT
ST
SSD
MSD
Total:
NUMBEROF
NEW VESSELS
100
57
0
1
8
1
1
35
6
7
3
0
0
220
16
0
8
0
0
1,200
NUMBEROF
EXISTING VESSELS
1300
95
3
9
110
5
8
440
80
78
38
3
2
2700
200
5
59
1
5
14,000
5.2.2.2 Existing Fleet That May Require Retrofit to Use Low Sulfur Fuel
Although most ships primarily operate on residual fuel, they typically carry some amount
of distillate fuel as well. This distillate fuel is available for use in emergencies such as
mechanical breakdown, off-spec bunker delivery, or prior to an extended engine shut-down to
clear the residual fuel out of the heaters and piping. Switching to the use of lower sulfur
distillate fuel is the compliance strategy assumed here to be used by both new and existing ships
when the new fuel sulfur standards go into effect. To estimate the potential cost of this
compliance strategy, we first evaluated the distillate storage capacity of the current existing fleet
to estimate how many ships may require additional hardware to accommodate the use of lower
sulfur fuel. We performed this analysis on the entire global fleet listed in Lloyd's database as of
2008. Of the nearly 43,000 vessels listed, approximately 20,000 vessels had provided Lloyd's
with fuel tankage information, cruise speed, and propulsion engine power data. Using this
information, we were able to estimate how far each vessel could travel on its existing distillate
carrying capacity.
The cruise speed provided by Lloyd's was used to determine the vessel's maximum
speed using Equation 2 while transit speed was assumed to be 12 knots, and maneuver speed 5.8
knots.13 The load factor used at cruise speed was 83 percent; while both the transit and
maneuver load factors were estimated by cubing the ratio their respective speeds to the ship's
maximum speed. The same low load factors used in the inventory (for loads less than 20
percent) were used here to adjust brake specific fuel consumption (BSFC) because diesel engines
are less efficient at low loads and the BSFC tends to increase. It was also assumed that ships
5-19
-------
spent a total of four hours per call in both transit and maneuver speeds. The fuel consumption
values used here were the same as reported in the inventory section, 195 g/kWh for SSD, 210
g/kWh for MSD, and 305 g/kWh for steam and gas turbines. The fuel consumed by auxiliary
engines was also taken into account and the same auxiliary power ratios used in the inventory
analysis were used here to estimate the total installed auxiliary engine power, as were the
auxiliary engine load factors appropriate for when the vessel is at cruise, transit, and maneuver
speeds for each ship.
Equation 2: Maximum Speed
Lloyds _ speed
094
0.83 = max imum_ speed
In order to determine if the current distillate capacity of a particular ship was sufficient to
call on a U.S. EGA without requiring additional hardware, we evaluated whether or not each ship
could travel 1,140 nm, the distance between the Port of Los Angeles and the Port of Tacoma.
This distance was selected because it represents one of the longer trips a ship could travel
without stopping at another port, and should overestimate the number of vessels that would
require such a modification. The amount of fuel a ship would consume calling on a port and
travelling a total distance of 1,140 nm was determined using the methodology described above.
The total fuel used in each mode (cruise, transit and maneuver) by both main and auxiliary
engines was summed and compared to the total amount of distillate fuel carried onboard. This
provided an estimate of the number of ships that had sufficient distillate capacity onboard, shown
in Table 5.2-5. The resulting percentages of ships determined to require a retrofit were then
applied to the number of existing ships in the 2015 fleet to estimate the total cost of this
compliance strategy for existing ships. The same percentages were also applied to all new ships
projected to be built in 2020 to determine the number of ships that may require additional
hardware and to estimate the cost of this compliance strategy for new vessels in 2020.
Table 5.2-5 Ships that Can Travel 1,140 nm on Existing Distillate Carrying Capacity
SHIP
TYPE
General
Cargo
Tanker
Container
Bulk
Cargo
RoRo
Auto
Carrier
Misc.
Passenger
Reefer
Total
TOTAL
#OF
SHIPS
4600
5900
1900
3600
510
360
1600
710
530
19,710
TOTAL* OF
SHIPS THAT
ONLY
CARRY
DISTILLATE
1900
740
45
230
70
20
1100
170
60
4,335
TOTAL # OF
SHIPS THAT
CARRY
DISTILLATE
+ ANOTHER
FUEL
2300
4900
1700
3000
380
310
210
460
440
13,700
SHIPS THAT
CARRY
DISTILLATE +
ANOTHER FUEL
THAT MAY
NEED A
MODIFICATION
#
200
1600
910
1600
30
20
70
270
20
4,720
%
8.9%
33%
53%
53%
7.6%
7.1%
34%
59%
4.1%
24%
TOTAL # OF
SHIPS THAT
CARRY NO
DISTILLATE
370
280
140
400
60
40
210
85
25
1,610
%NO
DISTILLATE
8.2%
4.7%
7.3%
11%
12%
10%
14%
12%
4.8%
8%
TOTAL OF ALL
SHIPS THAT MAY
NEED A
MODIFICATION
#
580
1900
1000
2000
90
60
280
360
40
6,310
%
13%
33%
55%
55%
18%
16%
18%
51%
8.2%
32%
5-20
-------
5.2.3 Tier III NOX Emission Reduction Technologies
The Selective Catalytic Reduction (SCR) process involves injecting a reagent, such as
ammonia or urea, into an exhaust flow, upstream of a reactor, to reduce NOx compounds into
nitrogen and water. Main system components are: an SCR reactor, aqueous urea
injection/dosing, and monitoring/control systems. The SCR system does require storage of urea
solution onboard in a separate tank. In addition to SCR, it is expected that manufacturers will
also use compound or two-stage turbocharging as well as electronic valving to enhance
performance and emission reductions to meet Tier III NOx standards. Engine modifications to
meet Tier III emission levels may also include a higher percentage of common rail fuel injection
coupled with two-stage turbocharging and electronic valving.
5.2.4 SOx/PM Emission Reduction Technology
In addition to Tier III NOx standards, the IMO EGA standards also include reductions in
fuel sulfur limits that will result in reductions in SOx and PM. While there are many existing
ships that already have the capacity to operate on both heavy fuel oil and distillate fuel and have
separate fuel tank systems to support each type of fuel, some ships may not have sufficient
onboard storage capacity to accommodate temporary fuel switching to operate both main and
auxiliary engines on lower sulfur fuel, since the minimum space practical is devoted to fuel and
machinery to maximize cargo space. If additional capacity is required, installation and use of a
fuel cooler, associated piping, and viscosity meters to the fuel treatment system may be required
to ensure viscosity matches between the fuel and injection system. If a new or segregated tank is
desired, ancillary equipment such as pumps, piping, vents, filling pipes, gauges, and access
would be required, as well as tank testing.14
5.2.5 NOx Emission Reduction Technology per Unit Hardware Costs
Tier III NOx standards are approximately 80 percent lower than the existing Tier I NOx
standards set by the IMO. To meet these standards, it is expected that SCR will be used along
with additional migration from either mechanically controlled mechanical fuel injection systems
(MFI) or electronically controlled fuel injection systems (EFI) to common rail, and engine
modifications. The methodology used here to estimate the capacity of the SCR systems is based
on the power rating of the propulsion engines only. Auxiliary engine power represents about 20
percent of total installed power on a vessel; however, it would be unusual to operate both
propulsion and auxiliary engines at 100 percent load. Typically, ships operate under full
propulsion power only while at sea when the SCR is not operating; when nearing ports the
auxiliary engine is operating at high loads while the propulsion engine is operating at very low
loads. It is estimated that the remaining 20 percent of SSD engines (5 percent MFI and 15
percent EFI) that have not already been upgraded to common rail to meet global Tier II NOx
standards will receive that upgrade for Tier III, and 40 percent of MSD (10 percent MFI and 30
percent EFI) will get common rail for Tier III as well. The fixed and variable costs of the six
'typical' engines developed for the migration to common rail from MFI are shown in Table
5.2-6.
5-21
-------
Table 5.2-6 Fixed and Variable Costs for MFI to Common Rail Fuel Injection Systems
SPEED
Engine Power (kW)
Cylinders
Liters/cylinder
Engine Speed (rpm)
MEDIUM
4,500
9
35
650
MEDIUM
9,500
12
65
550
MEDIUM
18,000
16
95
500
LOW
8,500
6
380
130
LOW
15,000
8
650
110
LOW
48,000
12
1400
100
VARIABLE COSTS
Component Costs
Electronic Control Unit
Common Rail Accumulators
(each)
Number of Accumulators
Low Pressure Pump
High Pressure Pump
Modified injectors (each)
Number of injectors
Wiring Harness
Total Component Cost
$3,500
$2,000
3
$2,000
$3,500
$2,500
9
$2,500
$40,000
$3,500
$2,000
6
$3,000
$4,500
$2,500
12
$2,500
$55,500
$3,500
$2,000
8
$4,000
$6,000
$2,500
16
$2,500
$72,000
$5,000
$2,000
9
$2,500
$4,500
$3,500
18
$3,000
$96,000
$5,000
$2,000
12
$3,500
$6,000
$3,500
24
$3,000
$125,500
$5,000
$2,000
18
$4,500
$8,000
$3,500
36
$3,000
$182,500
Assembly
Labor (hours)
Cost ($23.85/hr)
Overhead @ 40%
Total Assembly Cost
Total Variable Cost
Markup @ 29%
Total Hardware RPE
120
$2,900
$1,100
$4,000
$44,000
$12,800
$56,800
160
$3,800
$1,500
$5,300
$60,800
$17,700
$78,500
200
$4,800
$1,900
$6,700
$78,700
$22,800
$101,500
200
$4,800
$1,900
$6,700
$102,700
$29,800
$132,500
250
$5,900
$2,400
$8,300
$133,800
$38,800
$172,600
300
$7,100
$2,900
$10,000
$192,500
$55,800
$248,300
FIXED COSTS
R&D Costs (1 year R&D)
Retooling Costs
Marine Society Approval
Engines/yr.
Years to recover
Fixed cost/engine
$688,000
$1,000,000
$5,000
40
5
$8,500
$688,000
$1,000,000
$5,000
40
5
$8,500
$688,000
$1,000,000
$5,000
40
5
$8,500
$688,000
$1,000,000
$5,000
40
5
$8,500
$688,000
$1,000,000
$5,000
40
5
$8,500
$688,000
$1,000,000
$5,000
40
5
$8,500
The fixed and variable costs associated with the migration from EFI to common rail are
shown in Table 5.2-7. A curve-fit to estimate the variable cost of each technology was then used
to determine a $/kW equation applicable to other engine sizes and types, Figure 5-1 shows the
curve-fit for MFI to common rail variable costs and Figure 5-2 shows the curve fit for EFI to
common rail variable costs.
5-22
-------
Costs of Migrating from Mechanical Fuel Injection
to Common Rail
$300,000
$250,000 -
$200,000 -
•^ $150,000 -
$100,000 -
$50,000 -
• "Slow Speed - Mechanical Injection"
• Medium Speed Mechanical Injection
48,000, $248,000
y= 67,OOOLn(x) -470,000
15,000, $173,C
8,500,$132,000
18,000,$101,000
^500, $78,000
4,500, $57,000
y= 32,OOOLn(x) -215,000
10,000
20,000 *kW' 30,000
40,000
50,000
60,000
Figure 5-1 Variable Cost Curve-Fit for MFI to Common Rail Fuel Injection Systems
Table 5.2-7 Fixed and Variable Costs for EFI to Common Rail Fuel Injection Systems
SPEED
Engine Power (kW)
Cylinders
Liters/cylinder
Engine Speed (rpm)
MEDIUM
4,500
9
35
650
MEDIUM
9,500
12
65
550
MEDIUM
18,000
16
95
500
LOW
8,500
6
380
130
LOW
15,000
8
650
110
LOW
48,000
12
1400
100
Hardware Costs to the Manufacturer
Component Costs
Electronic Control Unit
Common Rail Accumulators
(each)
Number of Accumulators
Low Pressure Pump
High Pressure Pump
Modified injectors (each)
Number of injectors
Wiring Harness
Total Component Cost
$500
$2,000
3
$1,000
$1,500
$500
9
$500
$14,000
$500
$2,000
6
$1,000
$1,500
$500
12
$500
$21,500
$500
$2,000
8
$1,000
$1,500
$500
16
$500
$27,500
$500
$2,000
9
$1,500
$2,000
$750
18
$650
$36,150
$500
$2,000
12
$1,500
$2,000
$750
24
$650
$46,650
$500
$2,000
18
$1,500
$2,000
$750
36
$650
$67,650
Assembly
Labor (hours)
Cost ($23.85/hr)
40
$950
60
$1,430
80
$1,910
40
$950
60
$1,430
80
$1,910
5-23
-------
Overhead @ 40%
Total Assembly Cost
Total Variable Cost
Markup @ 29%
Total Hardware RPE
$380
$1,330
$15,300
$4,400
$19,700
$570
$2,000
$23,500
$6,800
$30,300
$760
$2,670
$30,200
$8,800
$39,000
$380
$1,330
$37,500
$10,900
$48,400
$570
$2,000
$48,700
$14,100
$62,800
$760
$2,670
$70,300
$20,400
$90,700
FIXED COSTS
R&D Costs (0.5 year R&D)
Retooling Costs
Marine Society Approval
Engines/yr.
Years to recover
FIXED COST/ENGINE
$344,000
$500,000
$5,000
40
5
$4,200
$344,000
$500,000
$5,000
40
5
$4,200
$344,000
$500,000
$5,000
40
5
$4,200
$344,000
$500,000
$5,000
40
5
$4,200
$344,000
$500,000
$5,000
40
5
$4,200
$344,000
$500,000
$5,000
40
5
$4,200
Costs of Migrating from Electronic Fuel Injection to
Common Rail
$100,000
"Slow Speed - Electronic Injectipn"
Medium SpeedJEIectronic Injection
$90,000 -
$80,000 -
$70,000 -
$60,000 -
$50,000 -
$40,000 -
$30,000 -
$20,000 -
$10,000 -
48,000, $91,000
15,000, $63,000
*-?4,000*Ln(x) -170,000
18,000, $39,000
),000
4,500, $20,000 y = 14,000*Ln(x) - 96,000
0
10,000
20,000
(kW)
30,000
40,000
50,000
60,000
Figure 5-2 Cost Curve-Fit for EFI to Common Rail Fuel Injection Systems
The variable costs associated with the use of engine modifications for Tier III include the
use of two stage turbochargers and electronic valve actuation, and are shown with the estimated
fixed costs in Table 5.2-8, Figure 5-3 shows the variable cost curve-fit used to determine a $/kW
equation applicable to other engine sizes and types. Table 5.2-9 shows the variable costs
associated with the use of SCR, these costs include the urea tank, the reactor, dosage pump, urea
injectors, piping, bypass valve, the acoustic horn, a cleaning probe and the control unit and
wiring. Detailed costs for the urea tank are shown in Table 5.2-10 and are based on estimated
storage of urea sufficient for up to 250 hours of normal operation of the SCR. It is envisioned
that the urea tank is constructed of 304 stainless steel, 1 mm thick due to the corrosive nature of
5-24
-------
urea, at a cost of approximately $2,700 per metric tonne.F The cost of Tier III technology as
presented here was developed using Tier II as a baseline. Figure 5-4 shows the shows the cost
curve used to determine a $/kW equation applicable to other engine types and sizes. The total
variable hardware costs of Tier III estimated here include the fuel injection changes, engine
modifications, and SCR.
Table 5.2-8 Fixed and Variable Costs for Engine Modifications Associated with Tier III
SPEED
Engine Power (kW)
Cylinders
Liters/cylinder
Engine Speed (rpm)
MEDIUM
4,500
9
35
650
MEDIUM
9,500
12
65
550
MEDIUM
18,000
16
95
500
LOW
8,500
6
380
130
LOW
15,000
8
650
110
LOW
48,000
12
1400
100
Hardware Costs to the Manufacturer
Component Costs
2 Stage Turbochargers (Incremental)
Electronic Intake Valves (each)
Intake Valves per Cylinder
Electronic Exhaust Valves (each)
Exhaust Valves per Cylinder
Controller
Wiring
Total Component Cost
Markup @ 29%
Total Hardware RPE
$16,250
$285
2
$285
2
$3,750
$2,800
$33,000
$10,000
$43,000
$20,900
$285
2
$285
2
$3,750
$2,800
$41,000
$12,000
$53,000
$46,750
$285
2
$285
2
$3,750
$2,800
$72,000
$21,000
$93,000
$28,000
$425
4
$3,750
$2,800
$45,000
$13,000
$58,000
$42,000
$425
4
$3,750
$2,800
$62,000
$18,000
$80,000
$61,000
$425
4
$3,750
$2,800
$88,000
$25,000
$113,000
Fixed Costs
R&D Costs (1 year R&D)
Retooling Costs
Marine Society Approval
Engines/yr.
Years to recover
Fixed cost/engine
$688,000
$1,000,000
$5,000
40
5
$8,500
$688,000
$1,000,000
$5,000
40
5
$8,500
$688,000
$1,000,000
$5,000
40
5
$8,500
$688,000
$1,320,000
$5,000
40
5
$10,000
$688,000
$1,320,000
$5,000
40
5
$10,000
$688,000
$1,320,000
$5,000
40
5
$10,000
F http://www.metalprices.com/FreeSite/metals/stainless product/product.asp#Tables for 2006.
5-25
-------
Engine Modifications for Tier III Costs
$140,000
Slow Speed - Engine Modifications
Medium Speed Engine Modifications
$120,000 -
$100,000 -
$80,000 -
$60,0008
$40,000 -
$20,000 -
$0
y = 32,000*l_n(x) - 227,000 48,ooo, $113,000
18,000, $92,000
15,000, $80,00'
500, $58,000^
' 9,500, $53,000
4,50"0, $43,000
y = 3.8x + 22,000
10,000
20,000 'kW* 30,000
40,000
50,000
60,000
Figure 5-3 Variable Cost Curve-Fit for Engine Modifications Associated with Tier III
Table 5.2-9 Fixed and Variable Costs Associated with the Use of SCR
SPEED
Engine Power (kW)
Cylinders
Liters/cylinder
Engine Speed (rpm)
MEDIUM
4,500
9
35
650
MEDIUM
9,500
12
65
550
MEDIUM
18,000
16
95
500
LOW
8,500
6
380
130
LOW
15,000
8
650
110
LOW
48,000
12
1400
100
Hardware Costs to the Supplier
Component Costs
Aqueous Urea Tank
Reactor
Dosage Pump
Urea Injectors (each)
Number of Urea Injectors
Piping
Bypass Valve
Acoustic Horn
Cleaning Probe
Control UnitAA/iring
Total Component Cost
$1,200
$200,000
$9,500
$2,400
3
$4,700
$4,700
$9,500
$575
$14,000
$251,000
$1,900
$295,000
$11,300
$2,400
6
$5,600
$5,600
$11,300
$575
$14,000
$360,000
$2,800
$400,000
$13,000
$2,400
8
$6,600
$6,600
$13,000
$575
$14,000
$476,000
$1,700
$345,000
$11,300
$2,400
12
$5,600
$5,600
$11,700
$700
$19,000
$429,000
$2,400
$560,000
$13,000
$2,400
16
$7,500
$6,600
$14,000
$700
$19,000
$662,000
$4,600
$1,400,000
$15,000
$2,400
24
$9,500
$7,500
$16,400
$700
$19,000
$1,530,000
Assembly
Labor (hours)
1000
1200
1500
1200
1600
2000
5-26
-------
Cost ($23.85/hr)
Overhead @ 40%
Total Assembly Cost
$23,900
$9,500
$33,400
$28,600
$11,400
$40,000
$35,800
$14,300
$50,100
$28,600
$11,400
$40,000
$38,200
$15,300
$53,500
$47,700
$19,100
$66,800
Total Variable Cost
Markup @ 29%
Total Hardware RPE
$284,800
$82,600
$367,400
$399,700
$115,900
$515,600
$525,800
$152,500
$678,300
$469,400
$136,100
$605,500
$715,000
$207,300
$922,300
$1,597,100
$463,200
$2,060,300
Fixed Costs
R&D Costs (1 year R&D)
Retooling Costs
Marine Society Approval
Engines/yr.
Years to recover
Fixed cost/engine
$1,376,000
$2,000,000
$5,000
40
5
$16,900
$1,376,000
$2,000,000
$5,000
40
5
$16,900
$1,376,000
$2,000,000
$5,000
40
5
$16,900
$1,376,000
$2,000,000
$5,000
40
5
$16,900
$1,376,000
$2,000,000
$5,000
40
5
$16,900
$1,376,000
$2,000,000
$5,000
40
5
$16,900
$2,500,000 -,
$2,000,000 -
$1,500,000 -
$1,000,000 -
$500,000-)
Tier III Selective Catalytic Reduction Costs
• Slow Speed -SCR • Medium Speed SCR
y= -0.0004X2 + 57.2x + 145,000
8,500, $606,000
15,000, $922,000
18,000, $678,000
4,500, $367,00
y= 22.6x+ 279,000
10,000
20,000 (kW) 30,000
40,000
48,000, $2,100,001)
50,000
60,000
Figure 5-4 Variable Cost Curve-Fit for SCR Systems
5-27
-------
Table 5.2-10 Detailed Urea Tank Variable Costs
SPEED
Engine Power (kW)
Cylinders
Liters/cylinder
Engine Speed (rpm)
MEDIUM
4,500
9
35
650
MEDIUM
9,500
12
65
550
MEDIUM
18,000
16
95
500
LOW
8,500
6
380
130
LOW
15,000
8
650
110
LOW
48,000
12
1400
100
Urea Tank Costs
Urea Amount (kg)
Density (kg/mA3)
Tank Size (mA3)
Tank Material (mA3)
Tank Material Cost ($)
12,910
1,090
14
0.04
$758
27,255
1,090
30
0.06
$1,248
51,642
1,090
57
0.09
$1,909
22,645
1,090
21
0.05
$977
39,961
1,090
37
0.07
$1,426
127,875
1,090
117
0.14
$3,093
Assembly
Labor (hours)
Cost ($/hr)
Overhead @ 40%
Total Assembly Cost
Total Variable Cost
Markup @ 29%
Total Hardware RPE
5
$119
$48
$167
$925
$268
$1,194
6
$143
$57
$200
$1,448
$420
$1,868
7
$167
$67
$234
$2,143
$621
$2,765
10
$238
$95
$334
$1,310
$380
$1,690
12
$286
$114
$401
$1,826
$530
$2,356
15
$358
$143
$501
$3,594
$1,042
$4,636
5.2.6 SOx and PM Emission Reduction Technology per Unit Hardware Costs
As discussed above, this cost analysis is based on the use of switching to lower sulfur
fuel to meet the EGA fuel sulfur standards when operating in the U.S portion of the proposed
EGA. This section discusses the costs that may be incurred by some newly built ships if
additional fuel tank equipment, beyond that installed on comparable new ships, is required to
meet lower sulfur fuel standards in the proposed EGA. We estimate that nearly one-third of new
vessels in 2020 may need additional equipment installed to accommodate additional lower sulfur
fuel storage capacity. The size of the tank is dependent on the frequency with which the
individual ship owner prefers to fill the lower sulfur fuel tank. The size of the tanks as estimated
here will carry capacity sufficient for 250 hours of propulsion and auxiliary engine operation
while within an EGA. Similar to the urea tank size estimation presented in this analysis, this is
most likely an overestimate of the amount of lower sulfur fuel a ship owner would need to call
on the proposed EGA. The hardware costs include additional distillate fuel storage tanks
assumed to be constructed of cold rolled steel 1 mm thick and double walled, an LFO fuel
separator, an HFO/LFO blending unit, a 3-way valve, an LFO cooler, filters, a viscosity meter,
and various pumps and piping. These costs are shown in Table 5.2-11. This cost analysis does
not reflect other design options such as partitioning of a residual fuel tank to allow for lower
sulfur fuel capacity which would reduce the amount of additional space required, nor does this
analysis reflect the possibility that some ships may have already been designed to carry smaller
amounts of distillate fuel in separate tanks for purposes other than continuous propulsion.
5-28
-------
Table 5.2-11 Fuel Switching Hardware Costs (New Construction)
SPEED
Engine Power (kW)
Cylinders
Liters/cylinder
Engine Speed (rpm)
MEDIUM
4,500
9
35
650
MEDIUM
9,500
12
65
550
MEDIUM
18,000
16
95
500
LOW
8,500
6
380
130
LOW
15,000
8
650
110
LOW
48,000
12
1400
100
Hardware Cost to Supplier
Component Costs
Additional Tanks
LFO Separator
HFO/LFO Blending Unit
3-Way Valve
LFO Cooler
Filters
Viscosity Meter
Piping/Pumps
Total Component Cost
Assembly
Labor (hours)
Cost ($23.85/hr)
Overhead @ 40%
Total Assembly Cost
Total Variable Cost
Markup @ 29%
Total Hardware RPE
$3,400
$2,800
$4,200
$950
$2,400
$950
$1,400
$2,000
$18,100
240
$5,700
$2,300
$8,000
$26,100
$7,600
$33,700
FIXED COSTS
R&D Costs (0.25 year
R&D)
Marine Society Approval
Engines/yr.
Years to recover
Fixed cost/engine
$172,000
$5,000
40
5
$880
$5,500
$3,300
$4,700
$1,400
$2,800
$950
$1,400
$2,000
$22,100
320
$7,600
$3,100
$10,700
$32,700
$9,500
$42,200
$172,000
$5,000
40
5
$880
$8,300
$3,800
$5,600
$1,900
$3,300
$950
$1,400
$2,000
$27,300
480
$11,400
$4,600
$16,000
$43,300
$12,600
$55,900
$172,000
$5,000
40
5
$880
$4,600
$3,800
$4,700
$1,400
$2,800
$950
$1,400
$2,000
$21,600
320
$7,600
$3,100
$10,700
$32,300
$9,400
$41,700
$172,000
$5,000
40
5
$880
$6,500
$4,200
$5,600
$1,900
$3,800
$950
$1,400
$2,000
$26,400
480
$11,400
$4,600
$16,000
$42,400
$12,300
$54,700
$172,000
$5,000
40
5
$880
$13,700
$4,700
$6,600
$2,800
$4,700
$950
$1,400
$2,000
$36,900
600
$14,300
$5,700
$20,000
$56,900
$16,500
$73,400
$172,000
$5,000
40
5
$880
In order to apply the hardware costs associated with the installation of equipment
required to use lower sulfur fuel in the proposed EGA, we needed to generate an equation in
terms of $/kW that could be applied to other engine sizes. The $/kW value hardware cost values
for the six data points corresponding to the six different engine types and sizes used in this
analysis were plotted. A curve fit was determined for the slow-speed engine as well as for the
medium speed engines, see Figure 5-5.
5-29
-------
Fuel Switching Hardware Costs - NEW Vessels
• Slow Speed - Fuel Switching Hardware Costs Existing Vessels
• Medium Speed Fuel Switching Costs-Existing Vessels
$80,000
$70,000 -
$60,000 -
$50,000 -
$40,000 -
$30,000 -
$20,000 -
$10,000 -
y= 18,OOOLn(x)- 120,000
48,000, $73,000
8,500, $42,000,
4,500, $34,000
9,500, $42,000
y= 1.64x+ 26,000
o
10,000
20,000
(kW)
30,000
40,000
50,000
60,000
Figure 5-5 $/kW Estimated Hardware Costs Associated with the use of Low Sulfur Fuel
5.2.7 Total Hardware Costs to New Ships in 2020
Total hardware costs associated with the proposed EGA were developed from the number
of new ships by ship and engine type estimated to enter the fleet in 2020 as presented earlier in
Table 5.2-4. All new vessels were considered to have the average characteristics (including
propulsion power) shown in Table 5.2-2. Hardware costs associated with switching to lower
sulfur fuel were applied to the percentage0 of new vessels in 2020 that may require additional
tankage, regardless of engine or ship type. The cost estimates developed for the 'typical' engines
discussed in Section 5.2.2 were used to develop $/kW equations that could be applied to other
engine sizes and types (e.g. SSD and MSD engines). The estimated hardware cost ranges for
new vessels, on a per-vessel basis, to meet Tier III NOx and lower sulfur fuel standards are
shown below in Table 5.2-12.
G Section 5.1.5 discusses the estimated percentage of the existing fleet that may require modifications to a retrofit,
the same percentages were applied to new vessels as it was assumed not all new vessels would require extra
hardware to accommodate the use of lower sulfur fuel.
5-30
-------
Table 5.2-12 Range of Technology Hardware Costs by Engine Type in $/kW
TECHNOLOGY
SOX/PM
Reductions
Tier III NOX
Reductions
Fuel Switching
Hardware Costs -
New Vessels
SCR Hardware Costs
ENGINE
SPEED
Medium
Slow
Medium
Slow
ENGINE SIZE
RANGE (KW)
4,500-18,000
8,500 - 48,000
4,500-18,000
8,500-48,000
$/KW
$3. 10 -$7.50
$1.50 -$4.90
$41.00 -$83. 00
$46.00 -$76.00
Table 5.2-13 Total Estimated Variable Hardware Costs per ShipH
SHIP TYPE
Auto Carrier
Bulk Carrier
Container
General Cargo
Passenger
Reefer
RoRo
Tanker
Misc.
Auto Carrier
Bulk Carrier
Container
General Cargo
Passenger
Reefer
RoRo
Tanker
Misc.
ENGINE
SPEED
MSD
MSD
MSD
MSD
MSD
MSD
MSD
MSD
MSD
SSD
SSD
SSD
SSD
SSD
SSD
SSD
SSD
SSD
AVERAGE
PROPULSION POWER
(KW)
9,600
6,400
13,900
5,200
23,800
7,400
8,600
6,700
9,400
11,300
8,400
27,500
7,700
23,600
10,400
15,700
9,800
4,700
NEW VESSEL
FUEL SWITCHING
HARDWARE3
$42,300
$36,900
$49,200
$34,900
$65,400
$38,500
$40,500
$37,400
$41,900
$48,000
$42,700
$63,900
$41,000
$61,200
$46,500
$53,900
$45,300
$32,000
AVERAGE PER
VESSEL COST OF
TIER IIIb
$573,200
$483,500
$687,800
$450,300
$952,500
$511,000
$543,800
$492,800
$566,800
$825,000
$672,600
$1,533,100
$632,900
$1,385,300
$781,000
$1,042,100
$744,200
$453,600
a Assumes 32 percent of new vessels would require the fuel switching equipment
b The cost estimates presented here represent the average cost per vessel, given that to meet Tier III not all
engines are expected to require the same hardware. The costs are determined using the following formula:
(5%*($/SHIPJVIECH^CR))+(15%*($/SHIP_ELEC->CR))+(T3 ENGINE MODS)+(T3 SCR))
5.2.8 Operational Costs Associated with SCR
In addition to the SCR hardware costs discussed above, ships built as of 2016 would
also incur the operating costs associated with SCR's use of urea. The urea operational costs
are based on a price of $1.52 per gallon with a density of 1.09 g/cc. The cost per gallon was
H Note that not all vessels will need these modifications - it is estimated that only 32% of all vessels will require
such additional hardware.
5-31
-------
estimated for a 32.5 percent urea solution delivered in bulk to the ship through research
completed by ICF combined with historical urea price information.15'16'17'18 This cost
analysis used a urea dosing rate of 7.5 percent that of the brake specific fuel consumption
value to estimate how much urea would be used by different engine types and sizes. The
total operational costs associated with the proposed EGA are based on the amount of fuel
consumed within the proposed EGA in the year 2020. Fuel consumption estimates for 2020
are presented in Chapter 2 of this report including how the amount of fuel used in this area
was determined and the fuel costs associated with a U.S. EGA. Based on the U.S. portion of
the proposed EGA, the operational costs associated with the use of urea by ships built as of
2016 in 2020 are based on total urea consumption of nearly 100 million gallons are shown in
Table 5.2-14 and estimated to be approximately $0.14 billion.
Table 5.2-14 Urea Operational Costs Associated with the use of SCR
SPEED
Engine Power (kW)
Cylinders
Liters/cylinder
Engine Speed (rpm)
Urea Costs
BSFC (g/kWh)
Load factor
Aequous Urea Rate
Aqueous Urea (kg/hr)
Aqueous Urea Cost per kg
Aqueous Urea Cost per hour
MEDIUM
4,500
9
35
650
210
73%
7.5%
52
$0.3684
$19
MEDIUM
9,500
12
65
550
210
73%
7.5%
109
$0.3684
$40
MEDIUM
18,000
16
95
500
210
73%
7.5%
207
$0.3684
$76
LOW
8,500
6
380
130
195
73%
7.5%
91
$0.3684
$33
LOW
15,000
8
650
110
195
73%
7.5%
160
$0.3684
$59
LOW
48,000
12
1400
100
195
73%
7.5%
512
$0.3684
$188
5.2.9 Existing Vessel Hardware Cost Estimates
This analysis also includes cost estimates for retrofitting existing vessels with additional
tankage and related fuel system components, see Table 5.2-15. These hardware costs include
additional distillate fuel storage tanks, an LFO fuel separator, an HFO/LFO blending unit, a 3-
way valve, an LFO cooler, filters, a viscosity meter, and various pumps and piping as well as
additional labor to install the systems on a ship and additional R&D to test systems on existing
ships. Similar to the lower sulfur fuel tank analysis discussed above, this existing vessel
hardware cost analysis assumes 250 hours of operation, which may be an overestimate of the
amount of fuel that is necessary to call on U.S. ports in the EGA. The total estimated hardware
costs of retrofitting the portion of the existing fleet estimated to require these modifications is
$327 million, these costs would be incurred by 2015.
5-32
-------
Table 5.2-15 Fuel Switching Hardware Costs - Existing Vessels
SPEED
Engine Power (kW)
Cylinders
Liters/cylinder
Engine Speed (rpm)
MEDIUM
4,500
9
35
650
MEDIUM
9,500
12
65
550
MEDIUM
18,000
16
95
500
LOW
8,500
6
380
130
LOW
15,000
8
650
110
LOW
48,000
12
1400
100
Hardware Cost to Supplier
Component Costs
Additional Tanks
LFO Separator
HFO/LFO Blending Unit
3-Way Valve
LFO Cooler
Filters
Viscosity Meter
Piping/Pumps
Total Component Cost
Assembly
Labor (hours)
Cost ($23.85/hr)
Overhead @ 40%
Total Assembly Cost
$3,400
$2,800
$4,200
$950
$2,400
$950
$1,400
$2,000
$18,100
480
$11,400
$4,600
$16,000
$5,500
$3,300
$4,700
$1,400
$2,800
$950
$1,400
$2,000
$22,100
640
$15,300
$6,100
$21,400
$8,300
$3,800
$5,600
$1,900
$3,300
$950
$1,400
$2,000
$27,300
960
$22,900
$9,200
$32,100
$4,600
$3,800
$4,700
$1,400
$2,800
$950
$1,400
$2,000
$21,600
640
$15,300
$6,100
$21,400
$6,500
$4,200
$5,600
$1,900
$3,800
$950
$1,400
$2,000
$26,400
960
$22,900
$9,200
$32,100
$13,700
$4,700
$6,600
$2,800
$4,700
$950
$1,400
$2,000
$36,900
1200
$28,700
$11,400
$40,100
Total Variable Cost
Markup @ 29%
Total Hardware RPE
$34,100
$9,900
$44,000
$43,400
$12,600
$56,000
$59,300
$17,200
$76,500
$43,00
$12,500
$55,500
$58,400
$17,000
$75,400
$77,000
$22,300
$99,300
Fixed Costs
R&D Costs (0.33 year R&D)
Marine Society Approval
Engines/yr.
Years to recover
Fixed cost/engine
$227,000
$5,000
40
5
$1,160
$227,000
$5,000
40
5
$1,160
$227,000
$5,000
40
5
$1,160
$227,000
$5,000
40
5
$1,160
$227,000
$5,000
40
5
$1,160
$227,000
$5,000
40
5
$1,160
5.3 Total Estimated ECA Costs in 2020
The total costs associated with improving ship emissions from current performance to
ECA standards in 2020 include both the hardware and operational costs as discussed above. The
hardware costs include those of SCR systems and equipment that may be installed on ships built
in 2020 to accommodate the use of switching to lower sulfur fuel which together total $1.04
billion in 2020. The operational costs associated with the use of urea are estimated to be $0.14
and the additional fuel costs for the U.S. portion of the proposed ECA will be $1.64 billion in
2020. Therefore, the total costs associated with the U.S. portion of the proposed ECA in 2020
are expected to be $2.78 billion, Table 5.3-1 summarizes these costs.
5-33
-------
Table 5.3-1 Total Estimated U.S. ECA Costs in 2020
Operating Costs (all ships
built as of 20 16)
Operating Costs (all ships
operating in ECA in 2020)
Hardware Costs
(ships built in 2020)
TECHNOLOGY
Urea Consumption
Fuel Switching
Fuel Tank
Modifications
SCR
Total Costs
COST IN 2020
(BILLIONS)
$0.14
$1.64
$0.02
$1.02
$2.78
5.4 Cost Effectiveness
As discussed in Chapters 3, 4 and 5, the proposed ECA is expected to bring many human
health and environmental benefits. Sections 5.1 through 5.3, above, summarize the various costs
of the proposed ECA. However, this does not shed light on how cost effective the proposed
ECA will be, compared to other control programs, at providing the expected emission reductions.
One tool that can be used to assess the value of the proposed ECA is the measure of cost
effectiveness; a ratio of engineering costs incurred per tonne of emissions reduced. The U.S.
Government has compared the ECA cost effectiveness to the ratio of costs per tonne of
emissions reduced for other control programs. As is shown in this section, the NOx, SOx and
PM emissions reductions from the proposed ECA compare favorably—in terms of cost
effectiveness—to other land-based control programs that have been implemented.
5.4.1 ECA Cost Effectiveness
Chapter 2 of this document summarizes the inventory analyses from which the U.S.
projections of pollutant reductions are drawn. The projected U.S. emission reductions due to the
proposed ECA are presented above in Table 2-46.
Note that PM2.5 is estimated to be 92 percent of the more inclusive PMio emission
inventory for marine vessels. In Chapter 2, we generate and present PM2.5 inventories since
recent research has determined that these are of greater health concern. Traditionally, we have
used PMio in our cost effectiveness calculations. Since cost effectiveness is a means of
comparing control measures to one another, we use PMio in our cost effectiveness calculations
for comparisons to past control measures.
Using the costs associated with NOx, SOx and PM control described in sections 5.1
through 5.3 above, and the emission reductions shown in Table 2-46, we calculated the cost per
tonne, or cost effectiveness, of the proposed ECA. As described above, the costs of the proposed
ECA include costs to refiners to produce additional distillate fuel, as well as costs for engine
controls, catalysts and reductants to reduce NOx emissions and costs for additional tankage for
5-34
-------
distillate oil. The timing of costs incurred varies, as some costs (i.e. capital expenditures) will be
near-term, while others, such as operational costs, are incurred over time in small increments.
The resultant cost per tonne numbers depend on how the costs are allocated to each
pollutant. We have allocated costs as closely as possible to the pollutants for which they are
incurred. The costs to apply engine controls to meet Tier III NOx standards, including catalysts
and reductants, have been allocated to NOx. In our analyses, we have allocated half of the costs
of fuel switching, including production and tankage, to PM and half to SOx because the costs
incurred for control measures to reduce SOx emissions directly reduce emissions of PM as well.
The resultant estimated cost effectiveness numbers are shown in Table 5.4-1. These
include costs and emission reductions that are expected to occur due to compliance with the U.S.
portion of the proposed EGA.
Table 5.4-1 Aggregate Long Term ECA Cost per Tonne (2006 U.S. Dollars)
POLLUTANT
NOX
SOX
PM2.5
30-YR NET PRESENT VALUE
DISCOUNTED AT 3%
2,600
1,200
11,000'
5.4.2 Land-Based Control Program Cost Effectiveness
The U.S. Government has already imposed restrictions on emissions of NOx, SOx, PM
and other air pollutants, from a wide range of land-based industrial (stationary) and
transportation (mobile) sources as well as consumer and commercial products. We have applied
a wide range of programmatic approaches to achieve significant air pollution reductions.
Regulatory regimes typically either mandate or incentivize emissions aftertreatment, cleaner
fuels or raw materials, improved practices, as well as new processes or technologies.
Significant emission reductions of NOx and SOx in the U.S. have been achieved via
performance standards for new combustion sources and market-based programs that cap
emissions at the regional level. Since 1996, the Acid Rain Program and NOx Budget Trading
Program have been highly successful at drastically reducing both NOX and SOX from power
plants in the Eastern U.S. Since 2004, NOx, SOx and PM emissions from highway and nonroad
heavy duty trucks and equipment have been decreasing with performance and emission standards
that will be completely phased in by 2010. To allow technology to advance, diesel fuel for use in
vehicles in the U.S. and Canada has been reduced to less than 0.0015 percent sulfur (15 parts per
million by weight), and diesel fuel for use in off-road equipment, locomotives and domestic
marine vessels will be reduced to this level by 2012.
Advanced technology is already required on stationary sources in the U.S., including
electricity generation produced by combustion; oil and gas; forest products (including pulp and
Converting to PMi0 the cost per tonne would be 10,000. This figure is used in Table 5.4-2 below.
5-35
-------
paper and wood products); smelting and refining (including aluminum, alumina, and base metal
smelting); iron and steel; iron ore pelletizing; potash; cement; lime; and chemicals production,
including fertilizers. On mobile sources, advanced technology to reduce NOx is phasing in by
2010 for engines on heavy duty trucks and by 2015 for engines on harborcraft.
Programs that are designed to capture the efficiency of designing and building new
compliant sources tend to have better cost-effectiveness than programs that principally rely on
retrofitting existing sources. Even considering the retrofitting programs, the control measures
that have been implemented on land-based sources have been well worthwhile when considering
the benefits of the programs. An early example of a highly effective NOx reduction program is
the regional NOx Budget Program. In 1998, the U.S. Government concluded that NOx
emissions reductions from retrofitting power plants that can be made for less than $3,400 per
tonne (in 2006 dollars) are "highly cost effective," considering the emissions reduced by the
advanced control technology, not including societal benefits.
The cost of reducing air pollution from these land-based sources has ranged greatly,
depending on the pollutant, the type of control program and the nature of the source. A selection
of programs and their cost effectiveness is presented in Table 5.4-2. Unless otherwise noted, the
programs named in the table address newly built sources only.
Table 5.4-2 Land-Based Source Control Program Cost Per Tonne" Comparisons
SOURCE CATEGORY19
Highway Diesel Fuel Program d
55 Fed Reg 34120, August 21, 1990
Stationary Diesel (CI) Engines c
71 Fed Reg 39 154, July 11, 2006
Locomotives and Harborcraft (Both
New and Retrofits) d
73 Fed Re^ 25097, May 6, 2008
Heavy Duty Nonroad Diesel Enginesd
69 Fed Reg 38957, June 29, 2004
Heavy Duty Onroad Diesel Engines d
66 Fed Reg 5001, January 18, 2001
International Shipping (ECA)
(Both New and Retrofits) d
Light Duty Gasoline/Diesel Engines d
65 FedReg6697, February 10, 2000
Fossil Fuel Fired Power Plants
(Retrofits) c
58 Fed Reg 3590, January 11, 1993;
63 Fed Reg 57356, October 27, 1998
Other Stationary Sources
(Both New and Retrofits) c
67 Fed Reg 80186, December 31, 2002
IMPLEMENTATION
DATE
1993
2006
2015
2015
2010
2016
2009
2000 to 2010
Ongoing
NOX
COST/TONNE
-
600 - 22,000
800b
1,200"
2,400 b
2,600
2,800"
3,400
4,000- 12,000
SOX
COST/TONNE
-
-
900
6,400
1,200
6,600
300
300 - 6,000
PM10
COST/TONNE
11,000
4,000 - 46,000
9,300 (New)
50,000
(Retrofit) c
14,000
16,000
10,000
14,000
Variable
Notes:
a Units are 2006 U.S. dollars per metric ton. To convert to $/short ton, multiply by 0.907.
b Includes NOX plus non-methane hydrocarbons (NMHC). NMHC are also ozone precursors, thus some rules set
combined NOX+NMHC emissions standards. NMHC are a small fraction of NOX so aggregate cost/ton
comparisons are still reasonable.
5-36
-------
c Annualized costs of control for individual sources, except SOX for Power Plants is a typical auction price.
d Aggregate program-wide cost/tonne over 30 years, discounted at 3%, except Light Duty and Highway Fuel
aggregate costs were discounted at slightly higher rates, yielding slightly lower cost estimates.
Another example of one of the earlier programs is the 1990 regulation promulgated by
the U.S. Government to reduce the sulfur content of highway diesel fuel. The cost effectiveness
of PM reductions from that program varied depending on how the benefit of reduced wear on the
engines was credited. Because the cleaner fuel with 0.05% sulfur (500 ppm) lengthened the
useful life of the engines, the program could be characterized as having negative costs (with
savings up to $100,000 per tonne) if the maximum engine wear credit was attributed to the
program. If no engine wear credit was included, the program was estimated to cost a maximum
of $ 11,000 per tonne of PM reduced.
As shown above, the projected cost per tonne of the proposed EGA falls well within the
respective ranges of the other programs. The proposed EGA cost-effectiveness is comparable to
the cost per tonne of current programs for new land-based sources, and has favorable cost
effectiveness compared to land-based retrofit programs.
5-37
-------
1 Research Triangle Institute, 2008. "Global Trade and Fuels Assessment—Future Trends and Effects of
Designating Requiring Clean Fuels in the Marine Sector"; Research Triangle Park, NC; EPA420-R-08-021;
November. (Available at http://www.epa.gov/otaq/regs/nonroad/marine/ci/420r08021.pdf)
2 Research Triangle Institute, 2008. "Global Trade and Fuels Assessment—Future Trends and Effects of
Designating Requiring Clean Fuels in the Marine Sector"; Research Triangle Park, NC; EPA420-R-08-021;
November. (Available at http://www.epa.gov/otaq/regs/nonroad/marine/ci/420r08021.pdf)
3 International Maritime Organization, Note by the Secretariat, "Revision of MARPOL Annex VI and NOX
Technical Code; Input from the four subgroups and individual experts to the final report of the Informal Cross
Government/Industry Scientific Group of Experts," Subcommittee on Bulk Liquids and Gases, 12th Session,
Agenda Item 6, BLG 12/INF.10, December 28, 2007.
4 International Maritime Organization, Note by the Secretariat, "Revision of MARPOL Annex VI and NOX
Technical Code; Input from the four subgroups and individual experts to the final report of the Informal Cross
Government/Industry Scientific Group of Experts," Subcommittee on Bulk Liquids and Gases, 12th Session,
Agenda Item 6, BLG 12/INF.10, December 28, 2007.
5 EnSys Energy & Systems, Inc. and RTI International 2009. Global Trade and Fuels Assessment—Additional EGA
Modeling Scenarios, prepared for the U.S. Environmental Protection Agency.
6 Energy Information Administration, 2006. "Annual Energy Outlook 2006" (DOE/EIA-0383(2006)); Washington,
DC. (Available at: http://www.eia.doe.gov/oiaf/aeo/archive.html)
7 Energy Information Administration, 2008. "Annual Energy Outlook 2008" (DOE/EIA-0383(2008)); Washington,
DC. (Available at: http://www.eia.doe.gov/oiaf/aeo/)
8 Energy Information Administration, 2008. "International Energy Outlook 2008" (DOE/EIA-0484(2008));
Washington, DC. (Available at: http://www.eia.doe.gov/oiaf/ieo/)
9ICF International, "Costs of Emission Reduction Technologies for Category 3 Marine Engines" prepared for the
U.S. Environmental Protection Agency, March 2009.
10 ICF International, "Commercial Marine Port Inventory Development," prepared for the U.S. Environmental
Protection Agency, EPA Report Number EPA420-R-07-012c, September 2007.
11 Lloyd's Register of Ships Online, Lloyd's Register, Fairplay. September, 2008 can be found at www.sea-
web.com.
12 "Matched Typical Ports to Modeled Ports" Table 2-33 of section 2.5.2 of "The Commercial Marine Port Inventory
Development, 2002 and 2005 Draft Inventories" report to EPA from ICF International, September 2007.
13 ICF International, "Commercial Marine Port Inventory Development 2002 and 2005 Draft Inventories" prepared
for the U.S. Environmental Protection Agency, September 2007.
14 Entec UK Limited, Quantification of Emissions from Ships Associated with Ship Movements between Ports in the
European Community, July 2002, pps. 86-87.
5-38
-------
15 "Nonroad SCR-Urea Study Final Report" July 29, 2007 TIAX for Engine Manufacturers Association (EMA) can
be found at:http://www.enginemanufacturers.org/admin/content/upload/198.pdf
16 http://www.adblueonline.co.uk/air_l/bulk_delivery
17 http://www.factsaboutscr.com/documents/IntegerResearch-Ureapricesbackto20051evels.pdf
18 http://www.fertilizerworks.com/fertreport/index.html
19 Regulation of Fuels and Fuel Additives: Fuel Quality Regulations for Highway Diesel Fuel Sold in 1993 and
Later Calendar Years, 55 Fed Reg 34120, August 21, 1990.
Standards of Performance for Stationary Compression Ignition Internal Combustion Engines, 71 FedReg3Q154,
July 11,2006.
Control of Emissions of Air Pollution from Locomotives and Marine Compression-Ignition Engines Less Than 30
Liters per Cylinder, 73 FedReg25097, May 6, 2008.
Control of Emissions of Air Pollution From Nonroad Diesel Engines and Fuel 69 FedReg38957, June 29, 2004.
Control of Air Pollution from New Motor Vehicles: Heavy-Duty Engine and Vehicle Standards and Highway
Diesel Fuel Sulfur Control Requirements 66 Fed Reg 5001, January 18, 2001.
Control of Air Pollution From New Motor Vehicles: Tier 2 Motor Vehicle Emissions Standards and Gasoline
Sulfur Control Requirements 65 FedRegQQQJ, February 10, 2000.
Acid Rain Program: General Provisions and Permits, Allowance System, Continuous Emissions Monitoring,
Excess Emissions and Administrative Appeals, 58 FedReg3590, January 11, 1993: Finding of Significant
Contribution and Rulemaking for Certain States in the Ozone Transport Assessment Group Region for Purposes of
Reducing Regional Transport of Ozone, 63 FedReg57356, October 27, 1998.
Prevention of Significant Deterioration (PSD) and Nonattainment New Source Review (NSR): Baseline Emissions
Determination, Actual-to-Future-Actual Methodology, Plantwide Applicability Limitations, Clean Units, Pollution
Control Projects, 67 Fed Reg 80186, December 31, 2002
5-39
-------
6 Economic Impacts
Chapter 5 provides the engineering costs associated with complying with the Tier III
NOx limits and the EGA fuel sulfur limits for all ships operating in the U.S. portion of the
proposed EGA in 2020. In this chapter, we examine the economic impacts of these costs on
shipping engaged in international trade. We look at two aspects of the economic impacts:
estimated social costs and how they are shared across stakeholders, and estimated market
impacts in terms of changes in prices and quantities produced for directly affected markets. All
costs are presented in terms of 2006 U.S. dollars.
The total estimated social costs associated with the U.S. portion of the proposed EGA in
2020 are equivalent to the estimated compliance costs of the program, at approximately $2.78
billion. These costs are expected to accrue initially to the owners and operators of affected
vessels. These owners and operators are expected to pass their increased costs on to the entities
that purchase their transportation services in the form of higher freight rates. Ultimately, these
costs will be borne by the final consumers of goods transported by ocean-going vessels in the
form of higher prices for those goods.
The compliance costs associated with the U.S. portion of the proposed EGA are described
earlier in this chapter. We estimate that these costs added to the total cost of shipping goods to
or from a U.S. origin or destination will result in only a modest increase in the costs of goods
transported by ship. We estimate that the cost to comply with the EGA requirements would
increase the price of a new vessel by 2 percent or less. With regard to operating costs, analysis
of a ship in liner service between Singapore, Seattle, and Los Angeles/Long Beach, which
includes about 1,700 nm of operation in the proposed EGA, suggests that improving from current
performance to EGA standards would increase the operating costs by about 3 percent. For a
container ship, this represents a price increase of about $18 per container, assuming the total
increase in operating costs is passed on to the purchaser of marine transportation services. This
would be about a 3 percent price increase. The per passenger price of a seven-day Alaska cruise
operating entirely within the EGA is expected to increase about $7 per day. For ships that spend
less time in the EGA, the expected increase in total operating costs would be smaller.
It should be noted that this economic analysis holds all other aspects of the market
constant except for the designation of the proposed EGA. It does not attempt to predict the
equilibrium market conditions for 2020, particularly with respect to how excess capacity in
today's market due to the current economic downturn will be absorbed. This approach is
appropriate because the goal of an economic impact analysis is to explore the impacts of a
specific program; allowing changes in other market conditions would confuse the impacts due to
the proposed regulatory program.
The remainder of this chapter provides detailed information on the methodology we used
to estimate these economic impacts and the results of our analysis.
6-1
-------
6.1 The Purpose of an Economic Impact Analysis
An Economic Impact Analysis (EIA) is prepared to provide information about the
potential economic consequences of a regulatory action. Such an analysis consists of estimating
the social costs of a regulatory program and the distribution of these costs across stakeholders.
In an economic impact analysis, social costs are the value of the goods and services lost
by society resulting from a) the use of resources to comply with and implement a regulation and
b) reductions in output. There are two parts to the analysis. In the economic welfare analysis,
we look at the total social costs associated with the program and their distribution across key
stakeholders. In the market analysis, we estimate how prices and quantities of goods and directly
affected by the emission control program can be expected to change once the program goes into
effect.
6.2 Economic Impact Analysis Methodology
Economic impact analysis is rooted in basic microeconomic theory. We use the laws of
supply and demand to simulate how markets can be expected to respond to increases in
production costs that occur as a result of the new emission control program. Using that
information, we construct the social costs of the program and identify how those costs will be
shared across the markets and, thus, across stakeholders. The relevant concepts are summarized
below and are presented in greater detail in Appendix 6A to this chapter.
Before the implementation of a control program, a market is assumed to be in
equilibrium, with producers producing the amount of a good that consumers desire to purchase at
the market price. The implementation of a control program results in an increase in production
costs by the amount of the compliance costs. This generates a "shock" to the initial equilibrium
market conditions (a change in supply). Producers of affected products will try to pass some or
all of the increased production costs on to the consumers of these goods through price increases,
without changing the quantity produced. In response to the price increases, consumers will
decrease the quantity they buy of the affected good (a change in the quantity demanded). This
creates surplus production at the new price. Producers will react to the decrease in quantity
demanded by reducing the quantity they produce, and they will be willing to sell the remaining
production at a lower price that does not cover the full amount of the compliance costs.
Consumers will then react to this new price. These interactions continue until the surplus is
removed and a new market equilibrium price and quantity combination is achieved.
The amount of the compliance costs that will be borne by stakeholders is ultimately
limited by the price sensitivity of consumers and producers in the relevant market, represented
by the price elasticities of demand and supply for each market. An "inelastic" price elasticity
(less than one) means that supply or demand is not very responsive to price changes (a one
percent change in price leads to less than one percent change in quantity). An "elastic" price
elasticity (more than one) means that supply or demand is sensitive to price changes (a one
percent change in price leads to more than one percent change in quantity). A price elasticity of
one is unit elastic, meaning there is a one-to-one correspondence between a percent change in
price and percent change in quantity.
6-2
-------
On the production side, price elasticity of supply depends on the time available to adjust
production in response to a change in price, how easy it is to store goods, and the cost of
increasing (or decreasing) output. In this analysis we assume the supply for engines, vessels, and
marine transportation services is elastic: an increase in the market price of an engine, vessel or
freight rates will lead producers to want to produce more, while a decrease will lead them to
produce less (this is the classic upward-sloping supply curve). It would be difficult to estimate
the slope of the supply curve for each of these markets given the global nature of the sector.
However, it is reasonable to assume that the supply elasticity for the ocean marine transportation
services market is likely to be greater than one. This is because output can more easily be
adjusted due to a change in price. For the same reason, the supply elasticity for the new
Category 3 engine market is also likely to be greater than one, especially since these engines are
often used in other land-based industries, especially in power plants. The supply elasticity for
the vessel construction market, on the other hand, may be less than or equal to one, depending on
the vessel type, since it may be harder to adjust production and/or store output if the price drops,
or rapidly increase production if the price increases. Because of the nature of this industry, it
would not be possible to easily switch production to other goods, or to stop or start production of
new vessels.
On the consumption side, we assume that the demand for engines is a function of the
demand for vessels, which is a function of the demand for international shipping (demand for
engines and vessels is derived from the demand for marine transportation services). This makes
intuitive sense: Category 3 engine and ocean-going vessel manufacturers would not be expected
to build an engine or vessel unless there is a purchaser, and purchasers will want a new
vessel/engine only if there is a need for one to supply marine transportation services. Deriving
the price elasticity of demand for the vessel and engine markets from the international shipping
market is an important feature of this analysis because it provides a link between the product
markets.
In this analysis, the price elasticity of demand is nearly perfectly inelastic. This stems
from the fact that, that, for most goods, there are no reasonable alternative shipping modes. In
most cases, transportation by rail or truck is not feasible, and transportation by aircraft is too
expensive. Approximately 90 percent of world trade by tonnage is moved by ship, and ships
provide the most efficient method to transport these goods on a tonne-mile basis. Stopford
notes that "shippers need the cargo and, until they have time to make alternative arrangements,
must ship it regardless of cost... The fact that freight generally accounts for only a small portion
of material costs reinforces this argument."2 A nearly perfectly inelastic price elasticity of
demand for marine transportation services means that virtually all of the compliance costs can be
expected to be passed on to the consumers of marine transportation services, with no change in
output for engine producers, ship builders, or owners and operators of ships engaged in
international trade.
The economic impacts described below rely on the estimated engineering compliance
costs presented in Chapter 5. These include the cost of hardware for new vessels to comply with
the Tier III engine standards, and the cost of fuel switching equipment for certain new and
existing vessels. Also included are expected increases in operating costs for vessels operating in
the EGA. These increased operating costs include changes in fuel consumption rates, increases
6-3
-------
in fuel costs, and the use of urea for engines equipped with SCR, as well as a small increase in
operating costs for operation outside the EGA due to the fuel price impacts of the program.
6.3 Expected Economic Impacts of the Proposed ECA
6.3.1 Engine and Vessel Market Impacts
The assumption of nearly perfectly inelastic demand for marine transportation services
means that the amount of these services purchased is not expected to change as a result of costs
of complying with the ECA requirements in the U.S. portion of the proposed ECA. As a result,
the demand for vessels and engines would also not change compared to the no-control scenario,
and the quantities produced would stay the same in 2020.
Also due to the assumption of nearly perfectly inelastic demand for marine transportation
services, the price impacts would be equivalent to the engineering compliance costs for the new
engine and vessel markets. Estimated price impacts for a sample of engine and vessel
combinations are set out in Table 6.3-1, for medium speed engines, and Table 6.3-2, for slow
speed engines.
Table 6.3-1 Summary of Estimated Market Impacts - New Medium Speed Engines and Vessels
(2020; $2006)
SHIP TYPE
Auto Carrier
Bulk Carrier
Container
General Cargo
Passenger
Reefer
RoRo
Tanker
Misc.
AVERAGE
PROPULSION
POWER
9,600
6,400
13,900
5,200
23,800
7,400
8,600
6,700
9,400
NEW VESSEL ENGINE
PRICE IMPACT (NEW
TIER III ENGINE
PRICE IMPACT)A
$573,200
$483,500
$687,800
$450,300
$952,500
$511,000
$543,800
$492,800
$566,800
NEW VESSEL FUEL
SWITCHING
EQUIPMENT PRICE
IMPACT8
$42,300
$36,900
$49,200
$34,900
$65,400
$38,500
$40,500
$37,400
$41,900
NEW VESSEL
TOTAL PRICE
IMPACT
$615,500
$520,400
$736,000
$475,200
$1,107,900
$549,500
$584,300
$530,200
$608,700
a Medium speed engine price impacts are estimated from the cost information presented in Chapter 5 using the
following formula: (10%*($/SHIPJVIECH^CR))+(30%*($/SHIP_ELEC->CR))+(T3 ENGINE MODS)+(T3SCR))
b Assumes 32 percent of new vessels would require the fuel switching equipment.
These price impacts reflect the impacts of the costs that will be incurred when the most
stringent ECA standards are in place in 2020. These estimated price impacts are small when
compared to the price of a new vessel.
6-4
-------
Table 6.3-2 Summary of Estimated Market Impacts - Slow Speed Engines and Vessels (2020; $2006)
SHIP TYPE
Auto Carrier
Bulk Carrier
Container
General Cargo
Passenger
Reefer
RoRo
Tanker
Misc.
AVERAGE
PROPULSIO
N POWER
11,300
8,400
27,500
7,700
23,600
10,400
15,700
9,800
4,700
NEW VESSEL ENGINE
PRICE IMPACT (NEW
ENGINE PRICE
IMP ACT) A
$825,000
$672,600
$1,533,100
$632,900
$1,385,300
$781,000
$1,042,100
$744,200
$453,600
NEW VESSEL FUEL
SWITCHING
EQUIPMENT PRICE
IMPACT8
$48,000
$42,700
$63,900
$41,000
$61,200
$46,500
$53,900
$45,300
$32,000
NEW VESSEL
TOTAL PRICE
IMPACT
$873,000
$715,300
$1,597,000
$673,900
$1,446,500
$827,500
$1,096,000
$789,500
$485,600
a Slow speed engine price impacts are estimated from the cost information presented in Chapter 5 using the
following formula: (5%*($/SHIPJV[ECH^CR))+(15%*($/SHIP_ELEC->CR))+(T3 ENGINE MODS)+(T3 SCR))
b Assumes 32 percent of new vessels would require the fuel switching equipment
A selection of new vessel prices is provided in Table 6.3-3, and range from about $40
million to $480 million. The program price increases range from about $600,000 to $1.5 million.
A price increase of $600,000 to comply with the EGA requirements would be an increase of
approximately 2 percent for a $40 million vessel. The largest vessel price increase noted above,
for a passenger vessels, is about $1.5 million; this is a price increase of less than 1 percent for a
$478 million passenger vessel. Independent of the nearly perfect inelasticity of demand, price
increases of this magnitude would be expected to have little, if any, effect on the quantity sales
of new vessels, all other economic conditions held constant.
Table 6.3-3 Newbuild Vessel Price by Ship Type and Size, Selected Vessels (Millions, $2008)
VESSEL
TYPE
Bulk Carrier
Container
Gas carrier
General
cargo
VESSEL SIZE
CATEGORY
Handy
Handymax
Panamax
Capesize
Feeder
Intermediate
Panamax
Post Panamax
Midsize
LGC
VLGC
Coastal Small
Coastal Large
Handy
Panamax
SIZE RANGE (MEAN)
(DWT)
10,095 - 39,990 (27,593)
40,009-54,881 (47,616)
55,000-78,932 (69,691)
80,000-364,767 (157,804)
1,000-13,966 (9,053)
14,003-36,937 (24,775)
37,042-54,700 (45,104)
55,238-84,900 (67,216)
1,001-34,800 (7,048)
35,760-59,421 (50,796)
62,510-122,079 (77,898)
1,000-9,999 (3,789)
10,000-24,912 (15,673)
25,082-37,865 (29,869)
41,600-49,370 (44,511)
NEWBUILD
$56.00
$79.00
$97.00
$175.00
$38.00
$70.00
$130.00
$165.00
$79.70
$37.50
$207.70
$33.00
$43.00
$52.00
$58.00
6-5
-------
VESSEL
TYPE
Passenger
Reefer
Ro-Ro
Tanker
VESSEL SIZE
CATEGORY
All
All
All
Coastal
Handymax
Panamax
AFRAmax
Suezmax
VLCC
SIZE RANGE (MEAN)
(DWT)
1,000-19,189(6,010)
1,000-19,126 (6,561)
1,000-19,126(7,819)
1,000-23,853 (7,118)
25,000-39,999 (34,422)
40,000-75,992 (52,300)
76,000-117,153 (103,112)
121,109-167,294 (153,445)
180,377-319,994 (294,475)
NEWBUILD
$478.40
$17.30
$41.20
$20.80
$59.00
$63.00
$77.00
$95.00
$154.00
Sources: Lloyd's Shipping Economist (2008), Informa (2008), Lloyd's Sea-Web (2008)
6.3.2 Fuel Market Impacts
The market impacts for the fuel markets were estimated through the modeling performed
to estimate the fuel compliance costs for the coordinated strategy. In the WORLD model, the
total quantity of fuel used is held constant, which is consistent with the assumption that the
demand for international shipping transportation would not be expected to change due to the lack
of transportation alternatives.
The expected price impacts of the coordinated program are set out in Table 6.3-4. Note
that on a mass basis, less distillate than residual fuel is needed to go the same distance (5 percent
less). The prices in Table 6.3-4 are adjusted for this impact.
Table 6.3-4 shows that the coordinated strategy is expected to result in a small increase in
the price of marine distillate fuel, about 1.3 percent. The price of residual fuel is expected to
decrease slightly, by less than one percent, due to a reduction in demand for that fuel.
Table 6.3-4 Summary of Estimated Market Impacts - Fuel Markets
FUEL
Distillate
Residual
Fuel
Switching
UNITS
$/tonne
$/tonne
$/tonne
BASELINE
PRICE
$462
$322
$322
CONTROL
PRICE
$468
$321
$468
ADJUSTED FOR
ENERGY
DENSITY
N/A
N/A
$444
% CHANGE
+1.3%
-0.3%
+38.9%
Because of the need to shift from residual fuel to distillate fuel in the EGA, ship owners
are expected to see an increase in their total cost of fuel. This increase is because distillate fuel is
more expensive than residual fuel. Factoring in the higher energy content of distillate fuel,
relative to residual fuel, the fuel cost increase would be about 39 percent.
6-6
-------
6.3.3 Marine Transportation Market Impacts
We used the above information to estimate the impacts on the prices of marine
transportation services. This analysis, presented in Appendix 6B to this chapter, is limited to the
impacts of increases in operating costs due to the fuel and emission requirements of the
coordinated strategy. Operating costs would increase due to the increase in the price of fuel, the
need to switch to fuel with a sulfur content not to exceed 1,000 ppm while operating in the EGA,
and due to the need to dose the aftertreatment system with urea to meet the Tier III standards.
Estimates of the impacts of these increased operating costs were performed using a
representative fleet, fuel cost, actual operational parameters, and sea-route data for three types of
ocean going vessels: container, bulk carrier, and cruise liner. The representative fleet values
used were obtained from the Lloyd's of London Sea-Web Database, and were based on actual
vessel size (Dead Weight Tonnes (DWT)) and engine power (kilowatt - hour (kW-hr)) of each
vessel type. Additionally, to develop a representative sea-route for our price estimations, we
created two theoretical trips, a 'circle route' occurring in the Pacific Ocean and an Alaskan
cruise. The total nautical mileage (nm) for the 'circle route' was determined to be 15,876 nm,
with approximately 1,700 nm occurring within the proposed U.S. EGA boundary, while the
Alaskan voyage travelled up the Canadian / Alaskan coastline for seven days, stopping at five
destinations, and operating completely in the proposed EGA for a total of 2,000 nm. We also
estimated the impacts for a trip to the port at Montreal (1,000 nm).
To conduct our price increase estimations, we calculated the average fuel operational
costs of the theoretical 'circle route' for the container and bulk carrier, and the Alaskan voyage
for the cruise liner as they would function today, completely on residual fuel. We then
calculated the operational fuel costs for the vessels if they were to travel the route with the U.S.
EGA in place. This EGA calculation was conducted assuming that the vessel would continue to
operate on residual fuel when outside of the EGA, and that approximately 33 percent of these
vessels would also use an exhaust aftertreatment technology that would require urea usage.
The overall price differences for each of these hypothetical trips were obtained by
subtracting the residual fuel operational costs from the calculated EGA operational fuel / urea
costs. Table 6.3-5 summarizes these price increases as they relate to goods shipped and per-
passenger impacts. Additionally, the table lists the vessel and engine parameters that were used
in the calculations.
Table 6.3-5 Summary of Impacts of Operational Fuel / Urea Cost Increases
VESSEL TYPE
Container
North Pacific Circle Route
Bulk Carrier
North Pacific Circle Route
Cruise Liner
(Alaska)
VESSEL AND ENGINE
PARAMETERS
36,540 kW
50,814 DWT
3,825 kW
16,600 DWT
31,500kW
226,000 DWT
1,886 passengers
OPERATIONAL PRICE
INCREASES
$17.53/TEU
$0.56 /tonne
$6.60 / per passenger per
day
6-7
-------
This information suggests that the increase in marine transportation service prices would
be small, both absolutely and when compared to the price charged by the ship owner per unit
transported. For example, Stopford notes that the price of transporting a 20 foot container
between the UK and Canada is estimated to be about $1,500; of that, $700 is the cost of the
ocean freight; the rest is for port, terminal, and other charges.3 An increase of about $18
represents an increase of less than 3 percent of ocean freight cost, and about one percent of
transportation cost. Similarly, the price of a 7-day Alaska cruise varies from $100 to $400 per
night or more. In that case, this price increase would range from 1.5 percent to about 6 percent.
Our analysis also suggests that increases in operational costs of the magnitude expected
to occur for vessels operating in the EGA are within the range of historic price variations for
bunker fuel. This is illustrated in Figure 6.3-1. This figure is based on variation in fuel price
among the ports of Singapore, Houston, Rotterdam, and Fujairah.
$675
$575
$475
$375
$275
$175
Baseline Value (Cheapest)
Most Expensive Fuel
3% Increase due to EGA
Cr
Cr
Date
Figure 6.3-1 Range of Bunker Fuel Prices
This graph illustrates the price differential between these ports, comparing the estimated
3% EGA increase to the cheapest fuel for each month. We then plotted these calculated EGA
increases (the 3% increases), the cheapest fuel (as a baseline) and the most expensive fuel for the
same six month period. As can be observed from the previous calculations and the trends in
Figure 1, there are both spatial and temporal price fluctuations in fuel prices. During this period
(granted, a period of above-average fluctuations), the price of fuel varied both spatially and
temporally. The variation over time is higher than the variation over ports; however, by either
form of variation, the 3% increase in bunker fuel price due to the EGA is smaller than the normal
price variation of the fuel.
6-8
-------
6.3.4 Social Costs of the Proposed EGA and Distribution Across Stakeholders
The total social costs associated with complying with the Tier III NOx limits and the
EGA fuel sulfur limits for all ships operating in the U.S. portion of the proposed EGA are
estimated to be the same as the total engineering costs presented in Chapter 5, or about $2.78
billion in 2020. For the reasons described above and explained more fully in the Appendix to
this chapter, these costs are expected to be borne fully by consumers of international shipping
services.
These social costs are small when compared to the total value of U.S. waterborne foreign
trade. In 2007, waterborne trade for government and non-government shipments by vessel into
and out of U.S. foreign trade zones, the 50 states, the District of Columbia, and Puerto Rico was
about $1.4 trillion. Of that, about $1 trillion was for imports.4
6-9
-------
Appendices
Appendix 6A
The methodology used in this Economic Impact Analysis (EIA) is rooted in applied
microeconomic theory and was developed following U.S. EPA's recommended procedures.5
This appendix describes the economic theory underlying the analysis and how it was applied to
the problem of estimating the economic impacts of the proposed EGA on shipping engaged in
international trade.
The Economic Theory Used to Estimate Economic Impacts
The approach used to estimate the economic impacts of the proposed EGA relies on the
basic relationships between production and consumption in competitive markets.
Multi-Market, Partial-Equilibrium Approach
The approach is behavioral in that it builds on the engineering cost analysis by
incorporating economic theory related to producer and consumer behavior to estimate changes in
market conditions. As Bingham and Fox note, this framework provides "a richer story" of the
expected distribution of economic welfare changes across producers and consumers. In
behavioral models, manufacturers of goods affected by a regulation are economic agents who
can make adjustments, such as changing production rates or altering input mixes, which will
generally affect the market environment in which they operate. As producers change their
production levels in response to a new regulation, consumers of the affected goods are typically
faced with changes in prices that cause them to alter the quantity that they are willing to
purchase. These changes in price and output resulting from the market adjustments are used to
estimate the distribution of social costs between consumers and producers.
This is also a multi-market, partial equilibrium approach. It is a multi-market approach
in that more than one market is examined: the markets for marine engines, vessels, and
international shipping transportation services. It is a partial-equilibrium approach in that rather
than explicitly modeling all of the interactions in the global economy that are affected by
international shipping, the individual markets that are directly affected by the EGA requirements
are modeled in isolation. This technique has been referred to in the literature as "partial
equilibrium analysis of multiple markets."7
This EIA does not examine the economic impact of the proposed EGA on finished goods
that use ocean transportation services as inputs. This is because international shipping
transportation services are only a small part of the total inputs of the final goods and services
produced using the materials shipped. A change in the price of marine transportation services on
the order anticipated by this program would not be expected to significantly affect the markets
for the finished goods. So, for example, while we look at the impacts of the program on ocean
transportation costs, we do not look at the impacts of the controls on gasoline produced using
crude oil transported by ship, or on manufactured products that use petroleum products as inputs.
6-10
-------
It should also be noted that this EIA estimates the aggregate economic impacts of the
control program at the market level. This is not intended to be a firm-level analysis; therefore
compliance costs facing any particular ship operator may be different from the market average,
and the impacts of the program on particular firms can vary significantly. The difference can be
important, particularly where the rule affects different firms' costs over different activity rates.
Competitive Markets
The methodology used in this EIA relies on an assumption of perfect competition. This
means that consumers and firms are price takers and do not have the ability to influence market
prices. Perfect competition is widely accepted for this type of analysis and only in rare cases are
other approaches used.8 Stopford's description of the shipping market and how prices are set in
this market supports this assumption.9
In a perfectly competitive market at equilibrium with no externalities, the market price
equals the value society (consumers) places on the marginal product, as well as the marginal cost
to society (producers). Producers are price takers, in that they respond to the value that
consumers put on the product. It should be noted that the perfect competition assumption is not
primarily about the number of firms in a market. It is about how the market operates: whether or
not individual firms have sufficient market power to influence the market price. Indicators that
allow us to assume perfect competition include absence of barriers to entry, absence of strategic
behavior among firms in the market, and product differentiation.J'10 Finally, according to
contestable market theory, oligopolies and even monopolies will behave very much like firms in
a competitive market if it is possible to enter particular markets costlessly (i.e., there are no sunk
costs associated with market entry or exit). This would be the case, for example, when products
are substantially similar (e.g., a recreational vessel and a commercial vessel).
Intermediate-Run Impacts
This EIA explores economic impacts on affected markets in the intermediate run. In the
intermediate run, some factors of production are fixed and some are variable. A short-run
analysis, in contrast, imposes all compliance costs on producers, while a long-run analysis
imposes all costs on consumers. The use of the intermediate run means that some factors of
production are fixed and some are variable, and illustrates how costs will be shared between
producers and consumers as the markets adjust to the new compliance program. The use of the
intermediate time frame is consistent with economic practices for this type of analysis.
Short-Run Analysis
In the very short run, all factors of production are assumed to be fixed, leaving producers
with no means to respond to the increased costs associated with the regulation (e.g., they cannot
adjust labor or capital inputs). Within a very short time horizon, regulated producers are
constrained in their ability to adjust inputs or outputs due to contractual, institutional, or other
J The number of firms in a market is not a necessary condition for a perfectly competitive market. See Robert H.
Frank, Microeconomics and Behavior, 1991, McGraw-Hill, Inc., p 333.
6-11
-------
factors and can be represented by a vertical supply curve, as shown in Figure 6A-1. Under this
time horizon, the impacts of the regulation fall entirely on the regulated entity. Producers incur
the entire regulatory burden as a one-to-one reduction in their profit. This is referred to as the
"full-cost absorption" scenario and is equivalent to the engineering cost estimates. Although
there is no hard and fast rule for determining what length of time constitutes the very short run, it
is inappropriate to use this time horizon for this type of analysis because it assumes economic
entities have no flexibility to adjust factors of production. Note that the BAF is a way to avoid
this scenario. Additionally, the fact that liner price schedules are renegotiated at least annually,
and that individual service contracts may be negotiated more frequently, suggests that a very
short-run analysis would not be suitable.
Price
Q
Figure 6A-1 Short-Run: All Costs Borne by Producers
Output
Long-Run Analysis
In the long run, all factors of production are variable, and producers can be expected to
adjust production plans in response to cost changes imposed by a regulation (e.g., using a
different labor/capital mix). Figure 6A-2 illustrates a typical, if somewhat simplified, long-run
industry supply function. The supply function is horizontal, indicating that the marginal and
average costs of production are constant with respect to output. This horizontal slope reflects
the fact that, under long-run constant returns to scale, technology and input prices ultimately
determine the market price, not the level of output in the market.
6-12
-------
Price
Increase
S.,: With Regulation
Unit Cost Increase
S • Without Regulation
Q1 Q0 Output
Figure 6A-2 Long-Run: Full Cost Pass-Through
Market demand is represented by the standard downward-sloping curve. The market is
assumed here to be perfectly competitive; equilibrium is determined by the intersection of the
supply and demand curves. In this case, the upward shift in the market supply curve represents
the regulation's effect on production costs and is illustrated in Figure 6A-2. The shift causes the
market price to increase by the full amount of the per-unit control cost (i.e., from PO to PI). With
the quantity demanded sensitive to price, the increase in market price leads to a reduction in
output in the new with-regulation equilibrium (i.e., Q0 to Qi). As a result, consumers incur the
entire regulatory burden as represented by the loss in consumer surplus (i.e., the area P0ac PI). In
the nomenclature of El As, this long-run scenario is typically referred to as "full-cost pass-
through."
Taken together, impacts modeled under the long-run/full-cost-pass-through scenario
reveal an important point: under fairly general economic conditions, a regulation's impact on
producers is transitory. Ultimately, the costs are passed on to consumers in the form of higher
prices. However, this does not mean that the impacts of a regulation will have no impact on
producers of goods and services affected by a regulation. For example, the long run may cover
the time taken to retire today's entire capital equipment, which could take decades. Therefore,
transitory impacts could be protracted and could dominate long-run impacts in terms of present
value. In addition, to evaluate impacts on current producers, the long-run approach is not
appropriate. Consequently a time horizon that falls between the very short-run/full-cost-
absorption case and the long-run/full-cost-pass-through case is most appropriate for this EIA.
Intermediate Run Analysis
The intermediate run time frame allows examination of impacts of a regulatory program
during the transition between the very short run and the long run. In the intermediate run, there
is some resource immobility which may cause producers to suffer producer surplus losses.
Specifically, producers may be able to adjust some, but not all, factors of production, and they
6-13
-------
therefore will bear some portion of the costs of the regulatory program. The existence of fixed
production factors generally leads to diminishing returns to those fixed factors. This typically
manifests itself in the form of a marginal cost (supply) function that rises with the output rate, as
shown in Figure 6A-3.
: With Regulation
Cost Increase
Price
Increase
: Without Regulation
Qi Qo
Figure 6A-3 Intermediate-Run: Partial-Cost Pass-Through
Output
Again, the regulation causes an upward shift in the supply function. The lack of resource
mobility may cause producers to suffer profit (producer surplus) losses in the face of regulation;
however, producers are able to pass through some of the associated costs to consumers, to the
extent the market will allow. As shown, in this case, the market-clearing process generates an
increase in price (from PO to PI) that is less than the per-unit increase in costs, so that the
regulatory burden is shared by producers (net reduction in profits) and consumers (rise in price).
In other words, there is a loss of both producer and consumer surplus.
Economic Impacts of a Control Program - Single Market
A graphical representation of a general economic competitive model of price formation,
as shown in Figure 6A-4(a), posits that market prices and quantities are determined by the
intersection of the market supply and market demand curves. Under the baseline scenario, a
market price and quantity (p,Q) are determined by the intersection of the downward-sloping
market demand curve (DM) and the upward-sloping market supply curve (SM). The market
supply curve reflects the sum of the domestic (Sd) and import (Sf) supply curves.
6-14
-------
= p
Q
Domestic Supply
Foreign Supply
a) Baseline Equilibrium
Market
P'
P
P'
P
P'
P
I"
q'd qo
Domestic Supply
q'f
Foreign Supply
b) With-Regulation Equilibrium
Q' Q
Market
Figure 6A-4 Market Equilibrium Without and With Regulation
With the regulation, the costs of production increase for suppliers. The imposition of
these regulatory control costs is represented as an upward shift in the supply curve for domestic
and import supply by the estimated compliance costs. As a result of the upward shift in the
supply curve, the market supply curve will also shift upward as shown in Figure 6A-4(b) to
reflect the increased costs of production.
At baseline without the new standards, the industry produces total output, Q, at price, p,
with domestic producers supplying the amount qd and imports accounting for Q minus qd, or qf.
With the regulation, the market price increases from p to p', and market output (as determined
from the market demand curve) decreases from Q to Q'. This reduction in market output is the
net result of reductions in domestic and import supply.
As indicated in Figure 6A-4, when the new standards are applied the supply curve will
shift upward by the amount of the estimated compliance costs. The demand curve, however,
does not shift in this analysis. This is explained by the dynamics underlying the demand curve.
The demand curve represents the relationship between prices and quantity demanded. Changes
in prices lead to changes in the quantity demanded and are illustrated by movements along a
6-15
-------
constant demand curve. In contrast, changes in consumer tastes, income, prices of related goods,
or population would lead to change in demand and are illustrated as shifts in the position of the
demand curve.K'u For example, an increase in the number of consumers in a market would
cause the demand curve to shift outward because there are more individuals willing to buy the
good at every price. Similarly, an exogenous increase in average income would also lead the
demand curve to shift outward or inward, depending on whether people choose to buy more or
less of a good at a given price.
Economic Impacts of a Control Program - Multiple Markets
The above description is typical of the expected market effects for a single product
market considered in isolation (for example, the ocean transportation service market). However,
the markets considered in this EIA are more complicated because they are linked: the market for
engines is affected by the market for vessels, which is affected by the market for international
marine transportation services. In particular, it is reasonable to assume that the input-output
relationship between the marine diesel engines and vessels is strictly fixed and that the demand
for engines varies directly with the demand for vessels. Similarly, the demand for vessels varies
directly with the demand for marine transportation services. A demand curve specified in terms
of its downstream consumption is referred to as a derived demand curve. Figure 6A-5 illustrates
how a derived demand curve is identified.
Price
Equipment
($/Q)
APE
AQE
Q .Equipment
Price
Engines
($/Q)
APei
it Cost Increase
Derived
Demand
AQ.
Figure 6A-5 Derived-Demand Curve for Engines
K An accessible detailed discussion of these concepts can be found in chapters 5-7 of Nicholson's (1998)
intermediate microeconomics textbook.
6-16
-------
Consider an event in the engine market, such as a new technology requirement, that
causes the price of an engine to increase by APeng. This increase in the price of an engine will
cause the supply curve in the engine market to shift up, leading to a decreased quantity (AQeng).
The change in engine production leads to a decrease in the demand for equipment (AQs). The
difference between the supply curves in the equipment market, S'E - SE, is the difference in price
in the engine market, APeng, at each quantity. Note that the supply and demand curves in the
equipment market are needed to identify the derived demand in the engine market.
In the market for vessels and engines, the derived demand curves are expected to be
vertical. The full costs of the engines will be passed into the cost of vessels, and the cost of
vessels will be passed into the cost of ocean transportation.
Using Economic Theory to Estimate the Social Costs of a Control Program
The economic welfare implications of the market price and output changes with the
regulation can be examined by calculating consumer and producer net "surplus" changes
associated with these adjustments. This is a measure of the negative impact of an environmental
policy change and is commonly referred to as the "social cost" of a regulation. It is important to
emphasize that this measure does not include the benefits that occur outside of the market, that
is, the value of the reduced levels of air pollution with the regulation. Including this benefit will
reduce the net cost of the regulation and even make it positive.
The demand and supply curves that are used to project market price and quantity impacts
can be used to estimate the change in consumer, producer, and total surplus or social cost of the
regulation (see Figure 6A-6).
6-17
-------
$/Q
Q2 Q,
(a) Change in Consumer Surplus with
Regulation
Q/t
$/Q
Q2 Q,
(b) Change in Producer Surplus with
Regulation
Q/t
$/Q
Q2 Q,
(c) Net Change in Economic Welfare with
Regulation
Q/t
Figure 6A-6 Economic Welfare Calculations: Changes in Consumer, Producer, and Total Surplus
The difference between the maximum price consumers are willing to pay for a good and
the price they actually pay is referred to as "consumer surplus." Consumer surplus is measured
as the area under the demand curve and above the price of the product. Similarly, the difference
between the minimum price producers are willing to accept for a good and the price they actually
receive is referred to as "producer surplus." Producer surplus is measured as the area above the
supply curve below the price of the product. These areas can be thought of as consumers' net
benefits of consumption and producers' net benefits of production, respectively.
6-18
-------
In Figure 6A-6, baseline equilibrium occurs at the intersection of the demand curve, D,
and supply curve, S. Price is PI with quantity Qi. The increased cost of production with the
regulation will cause the market supply curve to shift upward to S'. The new equilibrium price
of the product is P2. With a higher price for the product there is less consumer welfare, all else
being unchanged. In Figure 6A-6(a), area A represents the dollar value of the annual net loss in
consumer welfare associated with the increased price. The rectangular portion represents the
loss in consumer surplus on the quantity still consumed due to the price increase, Q2, while the
triangular area represents the foregone surplus resulting from the reduced quantity consumed, Qi
-Q2.
In addition to the changes in consumers' welfare, there are also changes in producers'
welfare with the regulatory action. With the increase in market price, producers receive higher
revenues on the quantity still purchased, Q2. In Figure 6A-6(b), area B represents the increase in
revenues due to this increase in price. The difference in the area under the supply curve up to the
original market price, area C, measures the loss in producer surplus, which includes the loss
associated with the quantity no longer produced. The net change in producers' welfare is
represented by area B - C.
The change in economic welfare attributable to the compliance costs of the regulations is
the sum of consumer and producer surplus changes, that is, -(A) + (B-C). Figure 6A-6(c) shows
the net (negative) change in economic welfare associated with the regulation as area D.
How the Economic Theory Applied in This EIA
In the above explanation of how to estimate the market and social welfare impacts of a
control action, the price elasticities of supply and demand were nonzero. This was reflected in
the upward-slope of the supply curve and the downward slope of the demand curve. In the
derived demand analysis, a nonzero price elasticity of demand in the vessel market yielded a
nonzero price elasticity of demand in the engine market.
However, the price elasticity of demand in the international shipping market is expected
to be nearly perfectly inelastic (demand curve with near-infinite slope - a vertical demand
curve). This is not to say that an increase in price has no impact on quantity demanded; rather, it
means that the price increase would have to be very large before there is a noticeable change in
quantity demanded.
The price elasticity of demand is expected to be near perfectly inelastic because there are
no reasonable alternatives to shipping by vessel for the vast majority of products transported by
sea to the United States and Canada. It is impossible to ship goods between these countries and
Asia, Africa, or Europe by rail or highway. Transportation of goods between these countries and
Central and South America by rail or highway would be inefficient due to the time and costs
involved. As a result, over 90% of the world's traded goods are currently transported by sea.12
While aviation may be an alternative for some goods, it is impossible for goods shipped in bulk
or goods shipped in large quantities. There are also capacity constraints associated with trans-
continental aviation transportation, and the costs are higher on a per tonne basis.
6-19
-------
A nearly perfectly inelastic price elasticity of demand simplifies the analysis described
above. Figure 6A-7 reproduces the relationships in a multi-level market but this time with a
nearly perfectly inelastic demand curve in the international shipping market. The relationships
between this market and the markets for vessels and engines means that the derived demand
curves for engines and vessels are also expected to be nearly perfectly inelastic. Specifically, if
demand for transportation services is not expected to be affected by a change in price, then the
demand for vessels will also remain constant, as will the demand for engines.
Qn
(a) The vertical demand curve for
ocean transportation market
1 ,ship
>
o.shlp
(b) The vertical demand curve for
ocean vessel market
(c) The vertical demand curve for
C-3 engine market
Figure 6A-7 Market Impacts in Markets with Nearly Perfectly Inelastic Demand
As indicated in Figure 6A-7, a change in unit production costs due to compliance with the
engine emission and fuel sulfur requirements in the proposed EGA shifts the supply curves for
engines, vessels, and ocean transportation services. The cost increase causes the market price to
increase by the full amount of per unit control cost (i.e. from P0 to PI) while the quantity
demanded for engines, vessels, and transportation services remains constant. Thus, engine
manufacturers are expected to be able to pass on the full cost of producing Tier III compliant
6-20
-------
engines to the vessel builders, who are expected to be able to pass the full cost of installing the
engines and fuel switching equipment on to the vessel owners. The vessel owners, in turn, are
expected to be able to pass on these cost increases, as well as the additional operating costs they
incur for the use of SCR reductant (urea) and low sulfur fuel while operating in the EGA.
Note that the fuel and urea costs affect the ocean transportation services market directly,
but affect the vessel and engine markets only through the derived demand curves. That is, the
equilibrium prices and quantities for vessels and engines will change only if the quantity of
ocean transportation services demanded changes due to fuel and urea costs. Because the changes
in fuel and urea prices are expected to be too small to affect the quantity of ocean transportation
services demanded, the markets for vessels and engines are not expected to be affected by fuel
changes.
The sole exception for the assumption of nearly perfectly price elasticity of demand is the
cruise market. Clearly, the consumers in that market, tourists and holiday-makers, have
alternatives available for their recreational activities. If the cost of a cruise increases too much,
they may decide to spend their vacation in other activities closer to home, or may elect to fly
somewhere instead. As a result, the costs of compliance for the cruise industry are more likely to
be shared among stakeholders. If the price elasticity of demand is larger (in absolute value) than
the price elasticity of supply, ship owners will bear a larger share of the costs of the program; if
the price elasticity of demand is smaller (in absolute value) than the price elasticity of supply,
consumers will bear a larger share of the program. Similarly, the vessel builders and engine
manufacturers will also bear a portion of the costs. If the quantity demanded for cruises
decreases, the derived quantity demanded for vessels will decrease, as will the derived quantity
demanded for engines. If the supply curves for these industries are not perfectly elastic (i.e.,
horizontal), then the downward-sloping derived demand curves will lead to shared impacts
among the sectors.
As described in section 6.3.3 of this chapter, the impacts on the cruise market are
expected to be small, with total engine and vessel costs increasing about one percent and
operating costs increasing between 1.5 and 6 percent. These increases are within the range of
historic variations in bunker fuel prices. The impact on the cruise market, then, may be similar
in effect to the market's response to those changes.
Finally, it may be possible for cruise ships to offset some of these costs by advertising the
environmental benefits of using engines and fuels that comply with the EGA requirements.
Many cruise passengers enjoy this form of recreational because it allows them a personal-level
experience with the marine environment, and they may be willing to pay an increased fee to
protect that nature. If people prefer more environmentally friendly cruises, then the demand
curve for these cruises will shift up. Consumers will be willing to bear more of the costs of the
changes. If the demand shift for environmentally friendly cruises is large enough, both the
equilibrium price and quantity of cruises might increase.
6-21
-------
Appendix 6B
Estimation of Transportation Market Impacts
The U.S. and Canada have submitted a joint proposal to IMO to designate an emission
control area in which ships would need to comply with stringent fuel sulfur limits and Tier III
NOx standards. To characterize the increase in vessel operating costs due to the proposed EGA,
and therefore the impacts on transportation market prices, calculations were performed for three
types of ocean going vessels, container, bulk carrier, and cruise liner. Our estimates were
developed using typical vessel characteristics, projected fuel and urea costs, and worst case sea-
route data. This appendix presents the methodology used for these calculations.
Container Vessel
A typical container vessel was derived using data obtained from the Lloyd's of London
Sea-Web Database. This data base includes information on actual vessel size (Dead Weight
Tonnes (DWT)) and engine power (kilowatt - hour (kW-hr)) for a wide range of vessel types.
Operating costs included those associated with switching from residual fuel to 0.1%
sulfur distillate fuel and urea consumption for vessels equipped with selective catalytic reduction
(SCR). The fuel and urea costs are based on projections that are presented in the EGA proposal.
These fuel costs estimates are $322/tonne for residual fuel and $468/tonne for 0.1% sulfur
distillate fuel. We use a urea consumption rate of 7.5% of fuel consumption, at $1.52/gallon.
To develop a representative sea-route for our price estimations, we created a 'circle route'
for a theoretical trip. Since the Port of Los Angeles1 , one of the largest ports in the U.S., lists
the majority of its cargo as traveling from South Asia, our route had a vessel hypothetically
travel from Singapore to the Port of Seattle, then down the West Coast of the United States
(U.S.) to the Port of Los Angeles, then back to Singapore. To map this route, we divided it into
three "legs." The first leg has the vessel traveling from Singapore to the Port of Seattle; the
second part travels down the West Coast of the U.S. to the Port of Los Angeles/Long Beach
(POLA/LB); the third leg continues from Los Angeles to Singapore. The total distance for this
route was determined from http://nauticaldistance.com/. and is described below.
We understand that it will take some additional time and distance to switch vessel
operations from one fuel to another. Additionally, we acknowledge that vessels may enter the
EGA at an angle relative to the port in question, and would be operating in the EGA for a slightly
longer distance than the 200 nautical miles of the EGA. Therefore, to make our fuel usage
estimates as accurate as possible, we included some additional EGA traversing distances in our
circle route calculations, adding 183 nm to the distance for reaching the Port of Seattle, and 35
nm to the distance from POLA/LB.
Baseline Operating Costs
In order to begin our estimated fuel cost increases, we needed to establish the fuel usage
and prices for our baseline route (i.e. the price of the route operating on residual fuel). We
determined average operational values for our hypothetical vessel by selecting the mid-point of
6-22
-------
the operational ranges used today by OGV. Therefore, our baseline estimations for the fuel
usage for the first leg were determined by multiplying the engine power for the average sized
containership (in kilowatts (kW)) by the average estimated engine efficiency (80 percent) as well
as the average residual fuel consumption (195 grams fuel per kilowatt hour (g/kW-hr)).
(Equation 6B-1) This value was then multiplied by the nautical miles (nm) for the first leg of the
trip (the distance from Singapore to Seattle (7,064 nm)), and divided by the average engine speed
(16 knots). To obtain the correct units for the calculation, a unit conversion was also included.
(Equation 6B-2) As average values are represented here, it is possible that these values could
fluctuate slightly depending on the vessel's speed, engine efficiency, and specific fuel
consumption, but we believe that these estimates provide a reasonable forecast for the majority
of container vessels in operation today.
Equation6B-l 36,5401H/x0.8xl95^re_ = 5,700,240 gresid
hr
5,700,240^^x7,064/7/77 tonne
Equation 6B-2 - / 1U . - x - =2,517 tonneresid
16 knots/ 1,000,000^
The same determinations were conducted for the second leg of the trip (1,143 nm,
Equation 6B-3) and the third leg (7,669 nm, Equation 6B-4).
5,700, 240 §resid/hr x 1,1 43/7/77 tonne
hr
Equation 6B-3 - , ' nr. - x - = 407to/7/7erp . ,
\§ knots/ 1,000,0005
5,700, 240 Sresid/ x 7,66972/77 tonne
Equation 6B-4 - / n r, - x - = 2,732tonneresid
\§ knots/ 1,000,0005
Total fuel usage for each leg of the trip was multiplied by the price of the fuel (2006 U.S.
dollars per tonne ($/tonne) which provided the baseline cost of fuel for each leg. These costs
were then summed to produce an aggregate estimation of fuel cost for the entire circle trip
(Equation 6B-5). This calculation provides the baseline cost of about $1.8M for an average sized
container ship to traverse the theoretical circle route.
Equation 6B-5
(2,5l7tonneresid + 407 tonne resid + 2,732tonneresid] x $322.487 tonneresid = $1,823,947
Operating Costs with an ECA
Operating cost increases due to an ECA are due to increased fuel costs and urea
consumption within the ECA. Operating costs are assumed to remain unchanged outside the
ECA. In addition, the ECA is assumed to have no impact on the route travelled.
6-23
-------
Increased Fuel Costs
To determine the fuel usage and price increase caused by the EGA on our vessel traveling
our theoretical circle route, we conducted the same analysis as our baseline using the appropriate
distillate fuel properties. Since the distillate fuel will most likely only be used in the EGA, the
remainder of the trip will continue operating on residual fuel. Therefore, we adjusted our trip
section distances accordingly, using residual fuel over the first leg for 6,679 nm and over 7,434
nm for the third leg, while the remainder of the trip was determined using a distillate fuel.
Equation 6B-6 provides the approximation for engine power and fuel consumption using
distillate fuel and Equation 6B-7, 8, and 9 calculate the corresponding trip segment fuel usages.
Due to the chemical properties of the two marine fuels, there is approximately a five percent
(5%) increase in energy, on a mass basis, when operating on the distillate fuel instead of the
residual fuel, and this increase is accounted for in Equation 6B-6.
Equation 6B-6
Equation 6B-7a Residual Fuel Estimation
36,540Wx0.8x - / kW - hr = 5 428,800^%
1 + 0.05 /hr
5,700,240 £"*% x 6,679/7/77 fnnnp
/ hr LUllllC „ oTrvi
^ x = 2,379 tonneresid
16 knots
hr
1,000,0005
Equation 6B-7b Distillate Fuel Estimation
5,428,800 gdisti/hr x 385/7/77 tonne
1,000,000s-
= 131 tonne*
Equation 6B-8
Equation 6B-9a Residual Fuel Estimation
5,428,800^^x1,143/7/77 tonne
- , / m , - x - = 388to/7/7e,,,,,
IS knots/ 1,000,0005-
,,,,
5,700, 240 Sresid/ x 7,
- hr
16 knot
x
Equation 6B-9b Distillate Fuel Estimation
tonne nCAO +
mam~g = 2'mtome""'
5,428,800 §dml/hr x 235/7/77 tonne
- , / - x - = 80 tonne .f
1 6 knots/ 1,000,0005
6-24
-------
Urea Costs
Switching to a distillate marine fuel will achieve reductions only in sulfur and participate
emissions. In order to meet the required Nitrogen Oxides (NOX) emission reductions, vessel
owners/operators would need to install a Selective Catalytic Reduction (SCR) device, or similar
technologies, on new vessels built in 2016 and later. Using an SCR requires dosing exhaust
gases with urea to aid with the emission reductions - which adds some additional costs to the
operation of the vessel. In an SCR on a marine engine, the average dosage of urea is seven and a
half percent (7.5%) per gallon of distillate fuel used. Subsequently, to estimate the volume of
urea required for our circle route, we multiplied the distillate quantity determined above by this
urea percentage. (Equation 6B-10) As we expect these costs to be incurred several years in the
future, we used the analysis preformed for the EPA by EnSys14 which predicted that in 2020,
33.2% of the fuel used in EGAs will be on vessels equipped SCR. The urea costs below are
adjusted to reflect this prediction.
Equation 6B-10
To determine the additional price of our vessel's operation through the EGA, we then
multiplied the fuel and urea quantities by their corresponding prices ($322.48/tonne for residual,
$467.92/tonne for distillate, and $1.52/gal for the urea). We then summed these values to
determine the aggregate price for fuel and urea required for our container vessel to travel our
circle route with the proposed EGA in place (Equation 6B-1 1).
[(2,379 tonneresid + 2,648toMeresJ x $322.487 tonneresid] +
Equation 6B-11 [ (1 3 1 tonnedisti} + 388tonnedisti} + 8Qtonnedisti}) x $467.927 tonnedisti}] +
= $1,908,549^
The total estimated price for an average sized containership traversing the circle with the
EGA in place is just over $1.9M. The cost increase of this trip caused by the fuel and urea prices
used in the EGA came from subtracting the baseline (residual fuel) trip price from the EGA price
(Equation 6B-12). The price differential between the baseline trip and the EGA trip is
demonstrated in Equation 6B-13 and takes into consideration the fuel cost portion of the
operational cost for a vessel, which is typically around 60 percent of the total. As can be seen,
by operating in the EGA for our theoretical circle route it is estimated that the operational costs
due to the distillate fuel is approximately three percent (3%) .
Equation 6B-12 $1,908,549£G4 -$l,823,947fea5etoe = $84,602
Equation 6B-13 0.60 x $1'908'549^ -$1.823.947
,,
To put this price increase in some perspective, we assumed our average sized
containership was hauling goods, such Twenty-foot Equivalent Units (TEU) , and estimated the
6-25
-------
increase per each TEU. Estimating these prices required the cargo weight of the vessel.
Literature shows that approximately 93-97% of a container vessel's DWT is used for hauling
cargo, with the remaining weight composing the crew, vessel engines and hull, and fuel15.
Equation 6B-14 shows the calculation used to convert the vessel's DWT to cargo weight using
the middle value of 95%.
Equation 6B-14 5Q,8UDWTx 0.95 = 48,273cargo_ tonnes
Dividing the difference between the baseline fuel price and the EGA fuel price we
calculated previously by the cargo tonnes as established in Equation 6B-14 provided the price
increase per tonne of good shipped for the entire route (Equation 6B-15).
($l,908,549£C4-$l,823,947&JsetoJ t1 „,.
Equation 6B-15 ^ basdmei = $ 1J 5 / car a o tonneinrrplKP
48,273cargo_to/7/7es 5 ~ ™e
Using this value and the weight of a full TEU (10 metric tonnes)16, we determined the
cost increase for shipping a fully loaded TEU across our circle route (Equation 6B-16).
$1.75 IQtonnes
-------
Equation 6B-18
no
0.8 x
195 r
1 + 0.05
kW-hr 1,763/im to/we
-^ — 1JL x — ; - -r- x - = 62.6 tonne
1$ knots/ l.OOO.OOOg-
Equation 6B-19
3,825^x0.8x195^^
- — — — —
; — —
IS knots/
-
1,000,0003-
= 52Q tonne rp,irl
"*"*
Equation 6B-20
Q2.Qtonnesrh,fi, x ^ x x 264-17^a/ x 0.075 = l,483^a/yrra x 0.332 = 492gallirea
dlst" Q.QQltonne 836.6^tf; nf * "rea * "rea
Equation 6B-21
[(62.6taMe^ x $467.921 tonne^} + (526tonneresicl x $322.487 tonne^) + (W2galurea x$\.52l galurj]
- [mtonneresid x $322.487tonne^] = $8,756toease
To establish this price increase in terms of bulk cargo shipped, the value from Equation
6B-21 was divided by the available cargo weight for the bulk carrier which was determined from
the actual vessel weight (16,600 tonnes) as was performed in Equation 6B-14. (Equation 6B-22)
Equation 6B-22
(to
(16,600«_cargo_toMesx0.95)
= $0.56/ bulk cargo tonnein
- & -
As can be seen, for an average bulk carrier that would travel from Singapore to Seattle,
LA/LB, and then back out to Singapore, the price increase caused by operation in the EGA
would be around $0.56 per tonne of good shipped. As with the other vessels, this price would
fluctuate depending on the distance traveled within the EGA, the vessel's speed, and the engine
power used.
Cruise Ship
We also conducted an analysis on a typical Alaskan cruise liner. These vessels tend to
operate close to shore and would be within the EGA for the majority of their routes. As such,
this analysis presents worst case cost impacts for this type of vessel.
To conduct this analysis, a series of average vessel characteristics were chosen along with
a typical 7 day Alaskan cruise route. The characteristics used below are the main engine power
(31,500 kW), auxiliary engine power (18,680 kW), base specific residual fuel consumption (178
gfuei/kW-hr for main engines, 188 gfuei/kW-hr for auxiliary engines), distance between voyage
destinations (5 destinations with a distance ranging between 230 to 700 nm), maximum vessel
speed (21.5 knots), and the average number of passengers on-board the vessel (1,886 people).
6-27
-------
Additionally, the arrival and departure times at the various ports of call along the cruise route
were used to calculate the average speed travelled between each destination. The required power
for a given journey segment was calculated using the relationship shown in Equation 6B-23.
This relationship was developed for the "2005-2006 BC Ocean-Going Vessel Emissions
Inventory"17 and was shared with several cruise ship operators for their input and validation.
Equation 6B-23
Required engine power = 0.8199 x(avg speed/max speed)3 - 0.0191 x (avg speed/max speed)2
+ 0.0297 x (avg speed/max speed) +0.1682
This relationship was developed to approximate effective power given cruise ships'
diesel-electric operation. The auxiliary engines reported within the Lloyd's of London 'Seaweb'
database18, and are presumably operated independently of the vessels main diesel-electric power
generation, as well as assumed to operate at an average of 50% power for the entire voyage.
To demonstrate the price increase for the cruise liner that would operate within the EGA,
calculations for one leg of the Alaskan voyage are shown in Equation 6B-24-27, the entire trip
operational cost increase per person in Equation 6B-28, and with Table 6B-1 depicting the total
increases over the entire trip broken out by destination.
Equation 6B-24
31,500kWx 0.5683 x 1? §fud x IWknots x — x —tome = mtonneresid
kW-hr \6.76knots 1,000,000^ resid
mtonneresid $322.48 .99QO/
Equation 6B-25 rj^ x = $22.897 person .,
l,m people tonneresid p resid
Equation 6B-26
Wx 0.5683 x—I78gfuel—x7Wknotsx — x tOme = 127tonnedjsti]
(l.05)kW-hr 16.76Jtnofc 1,000,000^ distil
\Tl tonnedlstll $467.92 M1G9/
Equation 6B-27 d-^- x = $31.627
1,886 people tonnedistil
Equation 6B-28 $31.62 - $22.89 = $8.737 personmain incr
6-28
-------
Table 6B-1 Alaskan Cruise Liner Destinations and the Corresponding Operational Price Increases
DESTINATION
ORIGIN
Vancouver
Sitka
Hubbard Glacier
Juneau
Ketchilkan
DESTINATION
CONCLUSION
Sitka
Hubbard Glacier
Juneau
Ketchilkan
Vancouver
Total
ESTIMATED PRICE INCREASE /
PERSON ($)
$8.73
$3.06
$2.67
$2.42
$6.13
J>Zj.UZm.|j|, increase
Additionally, the operational cost increases for the auxiliary engines were estimated
(Equation 6B-29-33), as well as the cost increases caused by dosing the engine exhaust with urea
(Equation 6B-34& 35), and the total price increase for the cruise (Equation 6B-36) divided by
the length of the cruise (Equation 6B-37).
Equation 6B-29
Equation 6B-30
Equation 6B-31
Equation 6B-32
Equation 6B 33
Equation 6B-34
616.75tonnesdjstax
18.680W x 0.50 x mgf»el x 168Ars x tonne = 29 5 to/we resj d
resid
kW-hr
1,000,000^
295toMe_w $322.48
tonne
= $50.441 personres
18,680Wx0.50x,i88£M . x!68Ar5x_^fl = 281tonnedisti}
(1.05)kW-hr
281tonnedistil $467.92
1,000,000^
person^
^
1,886 people tonnedistil
$69.71 -$50.44 = $19.277personaux in
nf 2G4.17gal
Q.QQltonne S36.6kgdjsti] nf
urea x 0.332 = 4,
Equation 6B-35
Equation6B-36 $23.02ma/D_AraB
Equation 6B-37
$46.207 persontotal_increase
ldayscr
= UQ.20/persontota}_incn
= $6.60/ person I day
-ruise _ length
6-29
-------
To put this price increase in perspective of the additional cost for a typical seven-day
Alaskan cruise, we also determined the % increase for the various stateroom types available on
the vessel. These values were established as shown in Equation 6B-38 and Table 6B-2 lists the
four main stateroom types used on a typical Alaskan cruise liner.
Equation 6B-38
$46.20
Stateroom _ /W7'ce($599)
x 100 = 7.7%
Table 6B-2 Representative Alaskan Cruise Liner Stateroom Price Increases
STATEROOM TYPE
Interior
Ocean View
Balcony
Suite
ORIGINAL AVERAGE
PRICE PER NIGHT ($)
$100
$200
$300
$400
PERCENTAGE
INCREASE
6.6%
3.3%
2.2%
1.7%
As can be seen from all the above price increase estimations, the additional costs of the
distillate fuel and the urea required to operate in the proposed EGA will not be a significant
monetary increase to the overall operation of the vessel, regardless of vessel type.
1 Harrould-Koleib, Ellycia. Shipping Impacts on Climate: A Source with Solutions. Oceana, July 2008. A copy of
this report can be found at
http://www.oceana.org/fileadmin/oceana/uploads/Climate_Change/Oceana_Shipping_Report.pdf
2 Stopford, Martin. Maritime Economics, 3rd Edition. Routledge, 2009. p. 163.
3 Stopford, Martin, Maritime Economics, 3rd Edition. Routledge, 2009. Page 519.
4 Census Bureau's Foreign Trade Division, U.S. Waterborne Foreign Trade by U.S. Custom Districts, as reported by
the Maritime Administration at
http://www.marad.dot.gov/library_landing_page/data_and_statistics/Data_and_Statistics.htm , accessed April 9,
2009.
5 U.S. EPA. "OAQPS Economic Analysis Resource Document." Research Triangle Park, NC: EPA 1999. A copy
of this document can be found at http://www.epa.gov/ttn/ecas/econdata/6807-305.pdf: U.S. EPA "EPA Guidelines
for Preparing Economic Analyses." EPA 240-R-00-003. September 2000. A copy of this document can be found
at http://yosemite.epa.gov/ee/epa/eed.nsf/webpates/guidelines.html
6 Bingham, T.H., and T.J. Fox. "Model Complexity and Scope for Policy Analysis." Public Administration
Quarterly, 23(3), 1999.
7 Berck, P., and S. Hoffman. "Assessing the Employment Impacts." Environmental and Resource Economics
22:133-156. 2002.
6-30
-------
8 U.S. EPA "EPA Guidelines for Preparing Economic Analyses." EPA 240-R-00-003. September 2000, p. 113. A
copy of this document can be found at http://yosemite.epa.gov/ee/epa/eed.nsf/webpates/guidelines.html
9 Stopford, Martin. Maritime Economics, 3rd Edition. Routledge, 2009. See Chapter 4.
10 Robert H. Frank, Microeconomics and Behavior, 1991, McGraw-Hill, Inc., p 333.
11 Nicholson, W., Microeconomic Theory: Basic Principles and Extensions, 1998, The Dryden Press, Harcourt
Brace College Publishers.
12 UN Conference on Trade and Development (UNCTAD), Trade and Development Report, 2008, Geneva.
13 http ://www. portoflosangeles. org
14 EnSys Navigistics, "Analysis of Impacts on Global Refining & C02 Emissions of Potential MARPOL
Regulations for International Marine Bunker Fuels," Final Report for the U.S. Environmental Protection Agency, 26
September 2007,
15 Wellmer, F.W., Dalheimer, M., Wagner, M. 2008. Economic Evaluations in Exploration. New York, NY:
Springer-Verlag Berlin Heidelberg
16 http://www.imo.org/includes/blastDataOnly.asp/data id%3D12740/471.pdf
http://www.cosbc.ca/index.php?option=com docman&task=doc view&gid=3&tmpl=component&format=raw&Ite
mid=53
18 http://www.sea-web.com/seaweb welcome.aspx
6-31
-------
Emission Control Area Designation
6-4
------- |