5-EPA
United States Office of International and Tribal Affairs
Environmental Protection EPA-1 60-R-l 0-001
Agency December 201 0
U.S.-Mexico Demonstration of Fuel
Switching on Ocean Going Vessels in the
Gulf of Mexico
Prepared for
U.S. Environmental Protection Agency
Office of Global Affairs and Policy
Office of International and Tribal Affairs
Washington, DC
Prepared by
ICF International
San Francisco, CA
EPA Contract Number EP-W-06-045
www.epa.gov/international/fuelswitch.html
-------
Disclaimer
This technical report does not necessarily represent final EPA decisions or positions. It
is intended to present technical analysis of issues using data that are currently available
and were collected through this project. The purpose in the release of such reports is to
facilitate the exchange of technical information and to inform the public of technical
developments which may form the basis for EPA action.
-------
Table of Contents
Table of Contents i
Executive Summary 1
1. Introduction 3
2. Project Goals and Partners 5
3. Benefits of Fuel Switching 7
3.1. Port Emissions Inventories 10
Port of Houston, USA 10
Port ofAlta Mira, Mexico 12
Port of Veracruz, Mexico 14
3.2. Dispersion Modeling 16
Methodology Overview 17
Results 21
4. Fuel Switching Demonstrations 35
4.1. Demonstration Design 35
4.2. Fuel Switching Logistics 37
Ship Operation on HFO 37
Switching from HFO to MGO 39
4.3. Maersk Roubaix Demonstration 40
Fuel Price 41
Operational Issues 42
4.4. Hamburg Sud Demonstration 43
Emission Sampling Methodology 44
Emission Sampling Results 49
Operational Issues 55
Estimated Fuel Switching Emission Reductions 55
5. Summary of Key Findings 59
Appendix A - Port Inventory Methodology 61
General Methodology 61
Port of Houston 67
Mexican Ports 69
Appendix B - Dispersion Modeling Methodology Details 71
Sources of Meteorological Data 71
Meteorological Data Record Details 73
Other Model Inputs 76
Model Execution 77
Appendix C - Monitoring Methodology 79
Appendix D - Related Information 87
List of Tables
Table 1: North American EGA Requirements 7
Table 2: Cost per tonne of emission reduction for NA EGA 9
Table 3: NORMA Oficial Mexicana for Modeled Pollutants 21
Table 4: Estimated Annual Total Deposition of SO2 30
Table 5: Maersk Roubaix Specifications 40
ICF International i
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Table of Contents
Table 6: Estimated Schedule for Maersk Roubaix 41
Table 7: Cap San Lorenzo Specifications 43
Table 8: Estimated Cap San Lorenzo Schedule 44
Table 9: Engine Operating Conditions for the ISO 8178 E-3 Cycle 46
Table 10: Operating Engine Load 47
Table 11: Engine Parameters Measured during Testing 48
Table 12: Detector Method and Concentration Ranges for Gaseous Monitoring 48
Table 13: Oceangoing Vessel Ship Types 62
Table 14: Auxiliary Engine Power Ratios 63
Table 15: Fuel Switching Times 63
Table 16: Vessel Movements and Time-ln-Mode Descriptions 64
Table 17: Auxiliary Engine Load Factor Assumptions 65
Table 18: Emission Factors for OGV Main Engines, g/kWh 66
Table 19: Calculated Low Load Multiplicative Adjustment Factors 67
Table 20: Auxiliary Engine Emission Factors (g/kWh) 67
Table 21: Port of Houston Maneuvering Times per Call 68
Table 22: Port of Houston Hotelling Times per Call 69
Table 23: Average Hotelling Times for Alta Mira and Veracruz 70
Table 24: Meteorological Data Record Sets 72
Table 25: February 2008 through January 2009 Composite Precipitation 76
Table 26: Quality Specifications for the Horiba PG-250 84
List of Figures
Figure 1: North American Emission Control Area 4
Figure 2: 2020 Potential EGA Ozone Reductions 8
Figure 3: 2020 Potential EGA PM25 Reductions 8
Figure 4: 2020 Potential Sulfur Deposition 9
Figure 5: Port of Houston Emissions Assuming a 24 nm Fuel Switching Zone 10
Figure 6: Port of Houston Emissions by Mode 11
Figure 7: Port of Houston Emissions by Ship Type 11
Figure 8: Port of Alta Mira Emissions Assuming a 24 nm Fuel Switching Zone 12
Figure 9: Port of Alta Mira Emissions by Mode 13
Figure 10: Port of Alta Mira Emissions by Ship Type 13
Figure 11: Effect of Fuel Switching Zone Distance for Port of Alta Mira 14
Figure 12: Port of Veracruz Emissions Assuming a 24 nm Fuel Switching Zone 15
Figure 13: Port of Veracruz Emissions by Mode 15
Figure 14: Port of Veracruz Emissions by Ship Type 16
Figure 15: Effect of Fuel Switching Zone Distance for Port of Veracruz 16
Figure 16: February 2008 through January 2009 Composite Record Wind Rose 18
Figure 17: Dispersion Modeling Sources and Receptors 20
Figure 18: Estimated 24-hour Average Concentrations of PM25on HFO 22
Figure 19: Estimated 24-hour Average Concentrations of PM25with Fuel Switching 23
Figure 20: Estimated Annual Average Concentrations of PM25on HFO 24
Figure 21: Estimated Annual Average Concentrations of PM25with Fuel Switching 25
Figure 22: Estimated 24-hour Average Concentrations of SO2on HFO 26
Figure 23: Estimated 24-hour Average Concentrations of SO2 with Fuel Switching 27
ICF International N
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Table of Contents
Figure 24: Estimated Annual Average Concentrations of SO2on HFO 28
Figure 25: Estimated Annual Average Concentrations of SO2with Fuel Switching 29
Figure 26: Estimated Annual Deposition of SO2 30
Figure 27: Ship Destinations from Port of Houston 35
Figure 28: Mexican Port Destinations from Port of Houston 36
Figure 29: Schematic of Fuel Switching Demonstration Design 36
Figure 30: Vessel Fuel System 38
Figure 31: Residual Fuel Unheated 39
Figure 32: Maersk Roubaix 40
Figure 33: Estimated Emissions for Fuel Switch at Port of Houston 42
Figure 34: Estimated Emissions for Fuel Switch at Port of Progreso 42
Figure 35: Hamburg Sud Cap San Lorenzo 43
Figure 36: Schematic of the Emission Sampling System 45
Figure 37: Emission Sampling of Main Engine Exhaust 46
Figure 38: Propulsion Engine SO2 Emissions 50
Figure 39: Propulsion Engine NOx Emissions 51
Figure 40: Propulsion Engine PM25 Emissions 51
Figure 41: Propulsion Engine Speciated PM25 Emissions 52
Figure 42: Speciated PM2 5 Emissions Comparisons with Other Ships 52
Figure 43: Auxiliary Engine SO2 Emissions 53
Figure 44: Auxiliary Engine NOx Emissions 54
Figure 45: Auxiliary Engine PM25 Emissions 54
Figure 46: Auxiliary Engine Speciated PM25 Emissions 55
Figure 47: Estimated Emissions for Fuel Switch at Port of Veracruz 56
Figure 48: Estimated Emissions for Fuel Switch at Port of Alta Mira 56
Figure 49: Estimated Emissions for Fuel Switch at Port of Houston 57
Figure 50: Data Sources and their Uses 62
Figure 51: Location of Hourly Meteorological Observations in Veracruz 74
Figure 52: Location of 10-Minute EMA Meteorological Observations near Veracruz 75
Figure 53: HFO Fuel Certificate of Analysis 79
Figure 54: MGO Fuel Certificate of Analysis 80
Figure 55: Fuel Audit Results 82
ICF International ill
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Table of Contents
ICF International iv
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Executive Summary
EPA engaged the U.S. Maritime Administration, the Port of Houston Authority, two maritime
shipping companies and government representatives from Mexico, including local, municipal,
state and federal agencies, such as the State of Veracruz, SEMARNAT (Secretarfa de medio
ambiente y recursos naturales, Mexico's Ministry of Environment and Natural Resources) and
PEMEX (Mexico's state-owned petroleum company) to conduct the first-ever EPA fuel switch
demonstration in the Gulf of Mexico. The project focused on illustrating the effectiveness of fuel
switching on ocean going vessels to reduce impacts to the Gulf of Mexico and its coastal
populations. Vessels switched from heavy fuel oil (HFO) to marine gas oil (MGO) within 24
nautical miles (nm) of one U.S. and several Mexican Gulf ports. EPA also sought to raise
awareness about the environmental benefits of the upcoming North American Emission Control
Area (NA EGA) effective in August 2012, which will require that ships use lower sulfur fuels
within 200 nautical miles of the majority of U.S. and Canadian Atlantic and Pacific coastal
waters, French territories off the Canadian Atlantic coast, the U.S. Gulf Coast, and the main,
populated islands of Hawaii. The NA EGA phases in lower sulfur fuels starting in 2012,
requiring 0.1 per cent sulfur fuel content by 2015. The NA EGA was established under the
auspices of Annex VI of the International Convention for the Prevention of Pollution from Ships
(MARPOL Annex VI), a treaty developed by the International Maritime Organization.
This project demonstrated the benefits of the fuel sulfur provision of the NA EGA. It showed that
fuel switching to MGO with a fuel sulfur content of less than 0.1 percent in the Gulf of Mexico on
two ocean going vessels leads to large emission reductions of sulfur oxide (SOx) and particulate
matter (PM) emissions and small emission reductions in nitrous oxide (NOx), as observed
through on-board emission sampling corroborated by calculated emission reductions. Human
exposure to these pollutants results in serious health impacts such as premature mortality and
aggravation of heart and lung disease. Atmospheric inputs related to emissions from fossil fuel
combustion and other sources of strong acids (such as nitric (HNO3) and sulfuric (H2SO4) acids)
alter surface seawater alkalinity, pH, and inorganic carbon storage which can disrupt natural
biogeochemical cycles. This is expected to have the greatest impact in near-coastal waters,
where the ecosystem responses to ocean acidification most affect human populations.
During the demonstrations, the test vessels encountered no operational issues of concern due
to fuel switching.
Emission measurements were taken on one test vessel while steaming between, approaching,
and hotelling at the Ports of Houston, Veracruz and Alta Mira. It was found that switching from
HFO (with a 3.79 % sulfur content) to MGO (with a 0.01% sulfur content) achieved significant
reductions in emissions of SOx and PM (2.5 micron in size) and small reductions in NOx- 89,
80 and 5 percent respectively -at a 2 percent increase in vessel operating costs, due to the
higher cost of lower-sulfur fuel.
Ship emission inventories were developed for the Ports of Houston, Veracruz and Alta Mira
using vessel port call data together with Lloyd's Register of Ships data. Emissions were
ICF International 1
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Executive Summary
estimated on both HFO with a sulfur content of 3.0 percent1 and MGO with a sulfur content of
0.1 percent. Emission calculations were based on EPA's Best Practice Guidance Document2.
Annual emissions by ship type, ship operating mode (e.g., maneuvering, hotelling, etc.), fuel
type and fuel switching zone boundary were calculated for each port. Tankers contributed most
to annual emissions in Houston, whereas containers were the largest sources of annual
emissions for Veracruz and Alta Mira. At all ports, the "cruise" operating mode contributed the
most to total annual ship emissions. At all ports calculated annual emissions reductions of NOx,
PM and SOx achieved through fuel switching within a 24 nm fuel switching zone were over 5,
75 and 80 per cent respectively. EPA found a three to five-fold increase in emissions reductions
using a 200 nm fuel switching zone boundary versus a 24 nm boundary.
Dispersion modeling was conducted for the Port of Veracruz using the calculated emission
inventory. The modeling showed a large reduction in impacts of ship emissions on port area air
quality and sensitive reefs due to fuel switching within 24 nm of the port. Only emissions from
ships were modeled. The study did not include the impact of other sources on air quality, such
as those from all other activities at the port as well as all other regional sources. Air quality
modeling showed a seven-fold reduction in 24-hour average and annual average PM2.s
concentrations and a 24- to 25-fold reduction in 24-hour average and annual average SO2
concentrations. This study has indicated that local concentrations of PM2.s pollution could be
reduced as much as 43 to 88 percent over the entire modeling domain by moving to a fuel-
switching mode for ships calling on the Port of Veracruz. Deposition modeling showed a 99 per
cent reduction of SO2 deposition to sensitive reef areas off the coast of Veracruz.
While acknowledging that this study has not quantified the effects of fuel switching on overall
concentrations or deposition of air pollutants, the reductions of PM and SOx concentrations
associated with fuel switching imply that similar results could be achieved in Mexico through
reduced use of HFO fuel in shipping.
1 This sulfur content for HFO was used for inventory calculations for the Gulf Region because SEMARNAT used 3.0% in their
inventory calculations for Mexican ports. 3.0% is assumed to be conservative for the Gulf Region based upon the two
demonstration projects. HFO had a sulfur content of 3.37% and 3.79% for the Maersk and Hamburg Sud demonstrations,
respectively. In addition, SEMARNAT indicates average HFO used in Mexico is 3.8% sulfur. Larger reductions should be
expected if the sulfur fuel levels are greater than the 3.0% assumed here..
2ICF International, Current Methodologies in Preparing Mobile Source Port Related Emission Inventories, Final Report, April
2009. Available at http://www.epa.qov/sectors/sectorinfo/sectorprofiles/ports/ports-emission-inv-
april09.pdf.
ICF International 2
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
1. Introduction
Ocean going vessels (OGVs) are used to transport the majority of goods (measured by weight
and value) globally. These vessels are a significant source of air pollution, affecting populations
and ecosystems especially near coastal areas34. EPA's modeling also shows potential impacts
far inland5. This project focused on the impact of OGV emissions in the Gulf of Mexico, where
they contribute to air pollution at ports throughout the Gulf region, and also adversely affect Gulf
ecosystems. One method of significantly reducing emissions from OGVs is to switch from a
high sulfur marine heavy fuel oil (HFO) (also known as bunker fuel or residual oil) to lower sulfur
marine gas oil (MGO) (also known as marine distillate fuel or marine diesel oil). . Switching from
HFO to MGO can dramatically reduce ship particulate matter (PM) and sulfur oxides (SOx)
emissions as well as achieving moderate reductions in nitrous oxide (NOx) emissions. These
and other pollutants emitted from ships are related to human and environmental health impacts,
including asthma, increased cancer risk, regional haze/smog, and, via aquatic deposition,
acidification and hypoxia. The Port of Houston, three key Gulf Ports in Mexico - Progreso, Alta
Mira and Veracruz - and the Port of Houston's Sister port in Brazil - Santos - have all been
targeted through this project, which involved switching to lower sulfur marine fuel in ships
approaching the U.S., Mexican and Brazilian coasts. EPA did not estimate or measure
emissions reductions at the Port of Santos for this report.
The United States Government, together with Canada and France, has established a North
American Emission Control Area (NA EGA) that will put in place lower sulfur marine fuel
standards and other requirements beginning in August 2012. The EGA was established under
the auspices of Annex VI of the International Convention for the Prevention of Pollution from
Ships ((MARPOL Annex VI), a treaty developed by the International Maritime Organization.
This EGA will require use of lower sulfur fuels in ships operating within 200 nautical miles of the
majority of the U.S. and Canadian coastline, including the U.S. Gulf Coast (see Figure 1). The
fuel switching demonstration project sought to demonstrate the benefits of the NA EGA
provision requiring 0.1 per cent fuel sulfur by 2015. This project also sought to raise awareness
throughout the Gulf of Mexico about the environmental and human health benefits associated
with implementing lower sulfur fuel content requirements, such as those of the NA EGA.
3 Corbett, J. etal. (2007), Mortality from Ship Emissions: A Global Assessment, Environ. Sci. Technol. 41(24):8512-8.
4 Dals0ren, S. B., et al. (2009), Update on emissions and environmental impacts from the international fleet of ships: the
contribution from major ship types and ports, Atmos. Chem. Phys., 9, 2171-2194.
5 U.S. Environmental Protection Agency, Regulatory Impact Analysis: Control of Emissions of Air Pollution from Category 3
Marine Diesel Engines, EPA Report EPA-420-R-09-019, December 2009. Available at
ICF International 3
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Introduction
Figure 1: North American Emission Control Area
o
to
§
o
to
o
CM
Saint-Pierre
SMiquelon
[France)
United States (48 states)
Hawaii (U.S
160°
140°
120°
1 DO-
60°
ICF International 4
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
2. Project Goals and Partners
This international project was the result of a partnership between the U.S. EPA, the Port of
Houston Authority, the Mexican federal government, the U.S. Maritime Administration, Maersk
Line, a Danish-based shipping company, and Hamburg Sud, a German-based shipping
company. Additionally, ICF International managed the technical elements of the program with
the University of California-Riverside performing the emission measurements on the Hamburg
Sud vessel.
EPA's fuel switch demonstration engaged the maritime shipping industry and government
representatives from Mexico, to raise awareness about the feasibility of fuel switching and the
environmental benefits of implementing fuel sulfur marine fuel requirements and the upcoming
North American EGA in August 2012. The fuel switching demonstration along with the emission
reduction estimates and dispersion modeling were intended, in particular, to inform policy
makers in the Gulf of Mexico of the potential health and environmental benefits of fuel switching.
EPA and the Mexican federal government conducted a technical workshop in April 2010 at the
Port of Veracruz in Mexico to launch the fuel switching demonstration. The workshop also
provided Mexican government and industry stakeholders an opportunity to learn first-hand about
this issue and to gather information on how to address marine emissions. It was well attended
by officials from local, municipal, state and federal agencies, including the State of Veracruz,
SEMARNAT6 and PEMEX7. This report presents the results of the fuel switching
demonstration, emission inventory development and emissions dispersion modeling. The fuel
switching demonstration enabled the documentation of any operational issues related to fuel
switching, the calculation of emissions reductions based on fuel use, and the direct
measurement of air pollutant reductions. The emission inventory was developed using port call
data at the Ports of Veracruz, Alta Mira and Houston. The dispersion modeling used the
emission inventory data to calculate air concentrations and loadings to the Gulf. This report and
a fuel switching outreach video are tools to help raise the awareness of stakeholders of the
benefits of fuel switching. In 2011 the video will also be available via the Gulf Coastal
Ecosystem Learning Centers and the National Oceanic and Atmospheric Administration's
Oceans Today Kiosk through the Smithsonian Institution. For resources and more information
see the project web site: www.epa.gov/international/fuelswitch.html.
6 The Ministry of Environment and Natural Resources (Secretaria de Medio Ambiente y Recursos Maturates, Semarnat) is a
federal government agency which main purpose is to promote the protection, restoration and conservation of ecosystems and
natural resources, as well as environmental goods and services, in order to promote their sustainable use and development.
7 Petroleos Mexicanos or Pemex is a Mexican state-owned petroleum company.
ICF International 5
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Project Goals and Partners
ICF International 6
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
3. Benefits of Fuel Switching
Fuel switching can produce significant emission reductions in coastal areas with benefits
potentially extending to inland areas. To quantify these reductions in the Gulf of Mexico, port
emission inventories were developed for the Port of Houston as well as the Ports of Alta Mira
and Veracruz in Mexico. In addition, dispersion modeling of PM and SOx emissions was done
at the Port of Veracruz to see the reduction in deposition on the city of Veracruz and the
surrounding sensitive reef areas.
The North American EGA will come into effect in August 2012 and will require the NOx and fuel
sulfur reductions shown in Table 1. This project focused on demonstrating the benefits of the
fuel sulfur provision.
Table 1: North American EGA Requirements
Requirements
NOx
Fuel Sulfur (%)
Outside EGA
20% reduction in new
vessels by 2011
2072: 3.50%
2020: 0.50%
The 2020 fuel
standard could be
delayed to 2025;
subject to 2018 fuel
availability review
Inside EGA
80% reduction in new
vessels by 2016
- 2010-14: 1.00%
- 2015: 0.10%
These NOx and fuel sulfur reductions will lead to substantial reductions in ozone and PM2.5
emissions and Sulfur depositions well into the interior of the U.S. as shown in Figure 2, Figure 3,
and Figure 4, respectively.
ICF International 7
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 2: 2020 Potential EGA Ozone Reductions
| <= -0.05 ppb
> -0.05 to < 0.05
>= 0.05 to < 0.10
| >=0.10to<0.20
>= 0.20 to < 0.50
| >=0.5to< 1.00
>= 1.00 to < 2.00
>= 2.00 to < 3.00
| >=3.00to< 5.00
I >= 5.00
Legend
<»00i ug/m3
>001lo<*00
>003to<=OC
>005to<* 0 1
> 0 1 to « 0 25
> 0 25 lo <* 0 5
> 0 5 !o <* 1 0
> 1 0 to <= 2 0
> 2 0 to <- 4 1
outsid*
Figure 3: 2020 Potential EGA PMa.s Reductions
ICF International 8
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 4: 2020 Potential Sulfur Deposition
Percent Change in Annual Total Sulfur Deposition
As a result of these reductions, the health benefits in the United States are substantial. In 2020,
EPA expects to save 5,500 to 14,000 lives and provide respiratory relief for 5 million people.
The monetized health benefits exceed $47 to $110 billion dollars annually. The cost per tonne8
of emission reduction from ships compares favorably with land-based emission control
programs as shown in Table 2.9
Total costs for EGA implementation in 2020 were estimated at $3.2 billion. These costs
included hardware costs for NOx controls, fuel system modifications and operating costs for
using lower sulfur fuel. Taking the monetized health benefits (as cited in the above paragraph)
and comparing to these total costs, the health benefit to cost ratio is substantial - ranging from
15:1 to 30:1.
Table 2: Cost per tonne of emission reduction for NA EGA
Pollutant EGA Land-Based
NOx
PM2.5
$2,600/tonne
$11,000/tonne
SOx $1,200/tonne
$200-$12,000/tonne
$2,000-$50,000/tonne
$200 - $6,000/tonne
8 Tonne is used here to denote metric tons.
9 U.S. Environmental Protection Agency, Proposal to Designate an Emission Control Area for Nitrogen Oxides, Sulfur Oxides and
Paniculate Matter, Report EPA-420-R-10-013, August 2010. Available at
http://www.epa.gov/otaq/regs/nonroad/marine/ci/420r10013.pdf
ICF International 9
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
3.1. Port Emissions Inventories
Port emission inventories of three ports were developed using vessel port call data together with
Lloyd's Register of Ships data. Emissions were estimated on both residual fuel with a sulfur
content of 3.0 percent10 (HFO) and distillate fuel with a sulfur content of 0.1 percent (MGO).
Emission calculations were done following EPA's Best Practice Guidance Document11 and are
discussed in detail in Appendix A.
Port of Houston, USA
Using the methodology described in Appendix A, emissions on HFO and MGO were calculated
for 2007 if all ships entering or leaving the Port of Houston used that fuel within 24 nm of the
U.S. coastline. Fuel switching was assumed to occur prior to the 24 nm boundary.
Comparisons of port emissions for Port of Houston are shown in Figure 5. As shown in the
figure, NOx emissions are reduced by 5 percent, PM2.s by 81 percent and SOx by 90 percent by
switching from HFO to MGO within 24 nm of port. This amounts to 402 metric tonnes of NOx,
544 metric tonnes of PM2.s, and 5,116 metric tonnes of SOx.
Figure 5: Port of Houston Emissions Assuming a 24 nm Fuel Switching Zone
8,000
7,000
o> 6,000
"3 5>000
H 4,000
§ 3,000
£ 2,000
LU
1,000
0
HFO (3.0% S)
MGO (0.1% S)
NOx
SOx
10 3.0 percent sulfur was used for inventory calculations for the Gulf Region because SEMARNAT used 3.0% in their inventory
calculations for Mexican ports. 3.0% is assumed to be conservative for the Gulf Region based upon the two demonstration
projects. HFO had a sulfur content of 3.37% and 3.79% for the Maersk and Hamburg Sud demonstrations, respectively. In
addition, SEMARNAT indicates average HFO used in Mexico is 3.8% sulfur. Larger reductions should be expected if the
sulfur fuel levels are greater than the 3.0% assumed here.
11ICF International, Current Methodologies in Preparing Mobile Source Port Related Emission Inventories, Final Report, April
2009. Available at http://www.epa.gov/sectors/sectorinfo/sectorprofiles/ports/ports-emission-inv-april09.pdf.
ICF International 10
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Based upon the emission inventory using the methodology described in Appendix A, emissions
for all ships operating on HFO in the various modes are shown in Figure 6. The largest
emissions are during the 24 nm cruise followed by emissions generated during transit down the
Houston Ship Channel and hotelling. Figure 7 shows emissions of PM2.s and SOx by ship type.
Tankers produce the highest emissions across all modes followed by container ships. Tanker
ships made 3002 calls at the Port of Houston while container ships only made 783 calls in 2007.
Figure 6: Port of Houston Emissions by Mode
3,000
Fuel Switch
Cruise
Transit
Manuever
Hotel
NOx
PM2.5
SOx
Figure 7: Port of Houston Emissions by Ship Type
PM
Auto
2.5 ^Carrier
2% _BuIk Carrier
10%
SOx Bulk
Carrier
RoRo
2%
Other RoRo
1% 2%
Genera
I
14%
ICF International 11
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Port of Alta Mira, Mexico
Using the methodology described in Appendix A, emissions on HFO and MGO were calculated
for 200512 if all ships entering or leaving the Port of Alta Mira used that fuel within 24 nm of the
Mexican coastline. Fuel switching was assumed to occur prior to the 24 nm boundary.
Comparisons of port emissions for Port of Alta Mira are shown in Figure 8. As shown in the
figure, NOx emissions are reduced by 6 percent, PM2.s by 76 percent and SOx by 84 percent by
switching from HFO to MGO within 24 nm of port. This amounts to 51 metric tonnes of NOx, 66
metric tonnes of PM2.s, and 615 metric tonnes of SOx. These emissions reductions are lower
than those for the Port of Houston due to the fact that total annual calls at the Port of Alta Mira
were 1,138 compared to Port of Houston's 5,778 calls.
Figure 8: Port of Alta Mira Emissions Assuming a 24 nm Fuel Switching Zone
1,200
HFO(3.0%S)
MGO(0.1%S)
NOx
PM2.5
SOx
Based upon the emissions inventory prepared using the methodology in Appendix A, emissions
for all ships operating on HFO in the various modes are shown in Figure 9. The largest
emissions are during the 24 nm cruise followed by emissions generated during hotelling. Figure
10 shows emissions of PM2.s and SOx by ship type. Container ships produce the highest
emissions at the port.
12 2005 was used for the Mexican port inventories because call data at the Mexican ports was only available for 2005.
ICF International 12
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 9: Port of Alta Mira Emissions by Mode
01
Ol
I/J
E
600
500
400
300
100
Fuel Switch
• Cruise
Manuever
• Hotel
NOx
PM2.5
SOx
PM,
Figure 10: Port of Alta Mira Emissions by Ship Type
Auto
.Carrier Bulk
.Carrier
9%
RoRo
4%
Passenger
0% General
9%
SOx Auto
.Carrier
6%
Bulk Carrier
9%
Emission reductions possible by extending a low-sulfur fuel switching zone out to 200 nm
instead of 24 nm is shown in Figure 11. As shown in the figure, emission reductions can be
increased by a factor of 5 by increasing a fuel switching zone from 24 nm as specified in this
study to a 200 nm boundary, such as that established by the North American EGA.
ICF International 13
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 11: Effect of Fuel Switching Zone Distance for Port of Alta Mira
3,500
3,000
2,500
.
4-1 QJ
I §2'000
1,500
1,000
500
NOx
PM2.5
SOx
Port of Veracruz, Mexico
Using the methodology described in Appendix A, emissions on HFO and MGO were calculated
for 200512 if all ships entering or leaving the Port of Veracruz used that fuel within 24 nm of the
Mexican coastline. Fuel switching was assumed to occur prior to the 24 nm boundary.
Comparisons of port emissions for Port of Alta Mira are shown in Figure 12. As shown in the
figure, NOx emissions are reduced by 6 percent, PM2.s by 78 percent and SOx by 87 percent by
switching from HFO to MGO within 24 nm of port. This amounts to 70 metric tonnes of NOx, 94
metric tonnes of PM2.s, and 892 metric tonnes of SOx. These comparatively lower emissions
reductions are due to the fact that total annual calls at the Port of Veracruz were 1,446
compared with 5,778 calls at the Port of Houston.
Based upon the emissions inventory prepared using the methodology in Appendix A, emissions
for all ships operating on HFO in the various modes are shown in Figure 13. The largest
emissions are during the 24 nm cruise followed by emissions generated during hotelling. Figure
14 shows emissions of PM2.s and SOx by ship type. Container ships produce the largest
contribution of emissions at the port.
ICF International 14
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 12: Port of Veracruz Emissions Assuming a 24 nm Fuel Switching Zone
c
c
_u
'C
4->
o;
1,600
1,400
1,200
1,000
800
§ 600
1
£ 400
LLJ
200
0
HFO(3.0%S)
MGO(0.1%S)
NOx
PM2.5
SOx
Figure 13: Port of Veracruz Emissions by Mode
700
600
o>
c 500
•Z 400
4->
0>
g 300
o
'55
.E 200
100
Fuel Switch
Cruise
Manuever
Hotel
NOx
PM2.5
SOx
ICF International 15
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 14: Port of Veracruz Emissions by Ship Type
PM
Tanker
2.5 6%
Container
^
RoRo
3%
Tanker SOx
6%.
Emission reductions possible by extending a low-sulfur fuel switching zone out to 200 nm
instead of 24 nm is shown in Figure 15. As shown in the figure, emission reductions can be
increased by a factor of 4 by increasing a fuel switching zone from 24 nm as specified in this
study to a 200 nm boundary, such as that established by the North American EGA.
Figure 15: Effect of Fuel Switching Zone Distance for Port of Veracruz
4,500
4,000
3,500
i/>
O •£• 3,000
4-1 01
112'500
ce. o
c £ 2,000
.2 01
£ ^
•| — 1,500
LLJ
1,000
500
0
NOx
PM2.5
SOx
3.2. Dispersion Modeling
In order to perform a screening-level assessment of health and ecosystem risk associated with
fuel switching at a port in Mexico, the emissions calculations of Section 3.1 were used to
estimate the air dispersion of key pollutants and their deposition to key sensitive ecosystem
ICF International 16
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
areas - coral reefs that are within a designated National Sea Park area near the Port of
Veracruz. The process of obtaining necessary meteorological data is discussed first, followed by
an overview of the modeling methodology and then a discussion of results. Technical details are
presented in Appendix Band then a discussion of results. Technical details are presented in
Appendix B.
Methodology Overview
Meteorology
Necessary meteorological data for the AERMOD model is prepared with the AERMET
preprocessor to incorporate the needed planetary boundary layer turbulence structure. Minimum
required inputs for AERMET include hourly surface wind speed and direction, temperature, sky
cover, and morning upper air sounding. Other inputs include various measurements of pressure,
humidity, cloud coverage, surface heat/radiation flux, and afternoon sounding data.
Obtaining this information required contacting numerous sources within various US and
Mexican agencies. Appendix B documents the sources contacted and the information available
from each. Ultimately, a variety of sources were compiled together into a single meteorological
record from February 2008 through February 2009, providing one year of relevant
meteorological data.13
Figure 16 shows a wind rose for the resulting annual meteorological record. This represents the
meteorology driving the dispersion simulations. It is clear that the dominant wind direction is
from the northeast, which is an on-shore direction for Veracruz. That is, there is a tendency for
pollution emitted at and approaching the port to be blown toward land, increasing the potential
for emissions to impact air quality and human health.
13 All analysis here is performed following US EPA guidance as much as practicable. As such, Appendix W to 40 CFR Part 51,
November 9, 2005, Section 8.3.1.2 states that, "Five years of representative meteorological data should be used when
estimating concentrations with an air quality model. Consecutive years from the most recent, readily available 5-year period
are preferred. The meteorological data should be adequately representative, and may be site specific or from a nearby NWS
station. Where professional judgment indicates NWS-collected ASOS (automated surface observing stations) data are
inadequate {for cloud cover observations}, the most recent 5 years of NWS data that are observer-based may be considered
for use." However, this data was not available for the current study location, and one year of composite data was created.
It is common to have different meteorological record years than emission years, typically to average out inter-annual variability
in meteorological records. However, an extended record was not available here. While an extended record may produce more
"typical" results, that hypothesis is untestable until a longer record becomes available. Instead, all available data was
employed. That this results in different years for emissions and meteorology is immaterial, as the results are meant to show
general impacts of fuel switching in the present time-frame, not those specific for any particular year. Further discussion
appears both below and in Appendix B
ICF International 17
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 16: February 2008 through January 2009 Composite Record Wind Rose
Station *76692 - Veracruz (Hybrid)
Wind SeeM
Diiocilon (blowing troim
WEST
•
.. - • •
^| ee in
|H 5,7- 8.B
• 18- -
] Zt-*«
I I 0.5- il
Ctfmr ll.TtK
MM
Jan 1 -D«. 31
00.00 • 23:00
H.71H
AVU WMD SPEH-
l.tsnvi
ICF International
S7}4hn.
11/17/2010
Ship Emissions
The emissions inventory, prepared as discussed above, includes emissions estimated for the
Port of Veracruz and two other ports for each of the four ship operating modes (cruise,
approach, maneuvering, and hotelling) under both a business as usual case - using only HFO -
and a fuel-switching case - using a combination of HFO and MGO. Annual emissions are
considered, from all vessels calling on each of the Ports for calendar year 2005. This is
discussed in Section 3.1, under the "Port of Veracruz" heading. The difference in these two
cases estimates the annual emission reductions achievable if all vessels included in the
inventory were to switch from high- to low-sulfur fuel.
ICF International 18
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Dispersion Model, Settings, and Other Inputs
To estimate the corresponding annual reduction in air pollutant concentration and deposition,
the ship emission quantities were used in an air dispersion model. Emissions from all four
operating modes were included. The individual results allow a demonstration of the fate of these
pollutants emitted from ships calling on the Port of Veracruz, while the difference between them
indicates the corresponding reduction in pollutant concentrations.
All modeling was conducted using U.S. EPA's AERMOD14 model. This allows current state-of-
the-science characterization of dispersion at a regional scale while balancing the resolution
required with the amount of input data. A domain with a radius of 50 km from the Port area was
characterized with the model.
Geolocation of ship activity was assigned using a GIS application. The port and ship channel
were modeled as a series of area sources. This was done by mapping the four operating modes
to four operating areas. Hotelling and maneuvering were assigned to the harbor area. The
shipping lane out to 24 nm from the harbor area was assigned the cruise emissions. The
shipping lane beyond 24 nm was assigned the fuel switch operating mode emissions, plus a
portion of the HFO cruise emissions to account for the operation on that fuel.
Receptors were assigned in a radius of 50 km from the harbor area, in increments of 10
degrees angularly and from 0.5 to 2.5 km radially, with decreasing resolution further from the
harbor area. Additionally, receptors were placed at the reef and island network near the harbor
to characterize deposition to those areas. Figure 17 shows the modeling domain, including the
various source and receptor locations.
14 AERMOD is a next generation dispersion model designed as the successor to the prior ISCST3. It is formulated as a steady-
state Gaussian plume model, but with updated PBL turbulence parameterization, and was added to Appendix W to 40 CFR
Part 51 as the preferred/recommended model for most modeling applications, including single and multi source simulations of
most types of emissions, including on- and off-road mobile and stationary sources in most environments, and domains up to
50 km from a source.
ICF International 19
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 17: Dispersion Modeling Sources and Receptors
Sources:
Activity in shipping lanes
Hoteling/Maneuvering
within the harbor
•Deposition monitors in island/reef
network
•Concentration monitors throughout
domain (50 km from port in all
directions)
Local terrain effects on the dispersion calculations were also included. A commercial source15 of
data for the local topography was used and processed using the AERMAP terrain preprocessor.
All emissions were assumed to be released at 50 m above the ground with an initial vertical
dimension of 23 m16. All receptors were taken at a breathing height of 1.8 m. The former will
lead to significantly diluted concentrations at ground level as pollutants are dispersed in the air.
The latter produces virtually identical concentrations to those at ground level. Effects of both dry
and wet removal of pollutants was considered. Also, all sources were modeled as area sources.
This method was used because the precise locations of the emission releases that occur in the
harbor area could not be determined, and the shipping lanes represent a non-steady state
emission source. Instead, area sources, with vertices of each source determined using the
digitized "footprint" of the harbor area and emissions distributed uniformly (horizontally)
throughout the areas was used. This will somewhat dilute the effects of emissions relative to
treatment of them as point sources, but is required given the uncertainty in ship location.
Emissions were characterized in terms of the official standards for air quality in Mexico,
"NORMA Oficial Mexicana NOM-Oxx-SSA1-yyyy", where xx represents the pollutant and yyyy
the year of its implementation. Each has both a chronic (i.e., long term exposure) and acute
15 www. mapmart.com
16 As determined in the California Air Resources Board's (ARB) "Diesel Participate Matter Exposure Assessment Study for the
Ports of Los Angeles and Long Beach", April 2006. This height was determined as, "...the average ship stack height is about
43 m tall. When the emissions are released from the top of a ship's exhaust stack, there is a plume rise that occurs which was
estimated to average to be about 7 meters. This results in an average release height of 50 meters."
ICF International 20
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
reference level. As for US NAAQS, the acute values are not defined in terms of peak
concentrations, but relative to a certain permissible number of exceedances per year. The
model was set to determine these "design values". Table 3 shows the standards for the
pollutants considered here, and the definition of the design value reported by the model. As only
a single year of meteorological data was available, the values represent those from this one
year.
Table 3: NORMA Oficial Mexicana for Modeled Pollutants
Valores de Concentracion Maxima
Contaminante Norma
Exposicion Aguda
Exposicion Cronica
Concentracion Frecuencia As Applied Concentracion As Applied
y tiempo maxima Here y tiempo Here
promedio aceptable promedio
Bioxido de
azufre (SO2)
Particulas
fraccion gruesa
(PMio)
NOM-
022-
SSA1-
1993
NOM-
025-
SSA1-
1993
0.13ppm 1 vez al ano 2nd Highest
(24 Horas)
High 24-Hour
Concentration <
130 ppb
120|jg/m3 2%delas 98th Percentile
m* LJ \ mediciones of Annual 24-
(24 Horas) de ^ ^Q^ ^
al ano2 Concentrations
0.03 ppm
(media
aritmetica
anual)
50 ug/m3
(media
aritmetica
anual)
Annual
Average
Concentration
< 30 ppb
Annual
Average
Concentration
< 50 ug/m3
120 g/mj
Particulas
fraccion fina
(PM2.5)
NOM-
025-
SSAI-
1993
65 ug/m
(24 Horas)
2% de las
mediciones
de 24 horas
al ano2
98th Percentile
of Annual 24-
Hour
Concentrations
< 65 Dg/m3
15 ug/m
(media
aritmetica
anual)
Annual
Average
Concentration
< 15 ug/m3
Results
Concentration of Pollutants
The emissions, meteorology, and other inputs discussed above were included in the dispersion
model to predict downwind concentrations. Figure 18 shows the resulting values of PM2.s for the
98th percentile of all 24-hour average concentrations reported at each receptor location when
operating on HFO. Figure 19 shows similar PM2.s values, but with all ships undergoing fuel
switching.
17
Note that the same background image is used in Figure 18 through Figure 25, with
concentrations shown in color overlapping the background image. The coastline generally runs
from the north-northwest to the south-southeast, facing east. The Port of Veracruz is centered in
each image, with the inner harbor shown in pink.
17 Note that although Figure 19 is labeled "MGO", it actually includes operations on both MGO within 24 nm of the port and HFO
beyond 24 nm.
ICF International 21
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 18: Estimated 24-hour Average Concentrations of PIVh.son HFO
PM2.5 24 Hour HFO Concentrations (ugi'ml) - Veracruz Port, Mexico
t nrm M»r» Arw
^HM Shipprg [.an* - Port la 20 «*
-------
Benefits of Fuel Switching
Figure 19: Estimated 24-hour Average Concentrations of PIVh.swith Fuel Switching
PM2.5 24 Houi MGO C o ri c « 111 r n s ( u
-------
Benefits of Fuel Switching
Figure 20 and Figure 21 show similar results to Figure 18 and Figure 19, but for annual average
PM2.s concentrations.
Figure 20: Estimated Annual Average Concentrations of PIVh.son HFO
PM2.5 Annual HFO C once nl ra I Ion * (ug/m3) - Vtracruz Port. Mexico!
(!rflr
^^_ Shining L»n» - Pert to 20
^^^ 3hionna Lane - 2410 SO
aiupttng Laiw - 20 to
I=MI 5 HFO u»imS
ICF International 24
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 21: Estimated Annual Average Concentrations of PIVh.swith Fuel Switching
'M2.6 Annual MGO Concentrations (uc|.'rn3) - Veracruz Port. Mexico!
As above, with a 2 km radius of the Port's center annual average concentrations are reduced
seven-fold under the fuel switching scenario, from an average concentration of 0.47 to 0.06
(ig/m3.
Figure 22 shows the resulting values for the 24-hour average concentrations of SO2 from ships
operating solely on HFO fuels for the modeled year. As described in Table 3, these represent
the 24-hour SO2 design value, which is the 2nd highest high of the series of 24-hour average
SO2 concentrations at each receptor location. Similarly, Figure 23 shows the 24-hour
concentrations of SO2 from ships operating in a fuel switching mode.
ICF International 25
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 22: Estimated 24-hour Average Concentrations of SOzon HFO
SO2 24 Hour HFO Concentrations (ppb) - Veracruz Port, Mexico
firm Haibcr *aa
Shfjpng Lwnc • EVrt hi M r*itw
• 2« » SO ITIIM
- 30 10 34 mlt*
30! HFO «*
010- OCC
0»-100
100-500
Greater ttfjn .
ICF International 26
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 23: Estimated 24-hour Average Concentrations of SCh with Fuel Switching
SO 2 24 Hour MGO Concentrations (ppb) • Veracruz Port, Mexico
Leoeod
^P »nar Harbor AJ*»
•BK Shionng Lmw • fcx-l to 20 MM
O^B Ehtpang Larw ^« » 50 tiit«
3fiiptf np Lana - 2Q U 24 milt*
902 WOO SOT
•
-------
Benefits of Fuel Switching
Figure 24: Estimated Annual Average Concentrations of SChon HFO
SO2 Annual RO Concentrations (ppb) - Veracruz Port, Mexico
Inner Harbor Area
^^^ Shipping Lane - Port to 20 mites
^^^ Shipping Lane * 24 to 50 miles
Shipping Lane - 20 to 24 miles
SO2 HFO ppb
Less than 0.05
0.05-0.10
^ 0.10-0.50
050-1.00
Greater than 1 00
ICF International 28
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
Figure 25: Estimated Annual Average Concentrations of SChwith Fuel Switching
SO2 Annual MGO Concentrations (ppb) - Veracruz Port. Mexico
»n» • Perl to 20 mil»u
3*1 W»3 iv« • JO lo ZJ miles
^^^ Shaping L*n» - 24 to GC mites
SO2 MOO ooa
Lssstftsn U D^
ooc-c ic
a 10-oso
0.90 - rod
Greater ttun 1 DD
As above, there are extensive pollutant reductions visible from moving to a fuel-switching
regime. Within a 2 km radius of the Port's center, annual average concentrations of SO2 are
reduced 25-fold under the fuel switching scenario, from an average concentration of 1.5 to 0.06
ppb.
Deposition of Pollutants
Deposition of pollutants from ship exhaust can also impact sensitive ecosystems, including
areas of natural productivity, critical habitats and areas of cultural and scientific significance.
The same dispersion modeling discussed above was also used to estimate the reduction in
deposition of sulfur (as SO2) to the local waters of the Gulf of Mexico. This deposition includes
both from dry and wet settling of SO2 from ship exhaust.
Figure 17 shows a series of receptors established to characterize the impact to the island and
reef network off the coast of Veracruz. The deposition at each of these receptors was calculated
with the AERMOD model. The total deposition was then calculated for each of the two
reef/island areas under both the HFO fuel usage case and the fuel switching case. Figure 26
ICF International 29
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
shows the total deposition to each of the areas in each case under the assumption that all
vessels in each inventory either switched from high- to low-sulfur fuel within 24 nm of shore or
operate solely on HFO. Table 4 tabulates these values.
Figure 26: Estimated Annual Deposition of S02
100,000
10,000
c
o
o
Q.
a;
Q
1,000
100 -
• Reef Area 1
.... • Reef Area 2
10
HFO MGO Difference
Table 4: Estimated Annual Total Deposition of S0218
I Percent
Units HFO MGO Difference Reduction
Reef Area 1
Area
Total Annual SO2 Flux
Total Annual Deposition
Area
m2
g/m2
kg
m2
283,474,477
0.19
53,000
0.01
1,900
0.18
52,000
96%
57,673,276
Reef Area 2
Total
Total Annual SO2 Flux
Total Annual Deposition
g/m2
kg
Total Annual SO2 Deposition kg
0.0093
540
54,000
0.00081
47
2,000
0.008
490
91%
52,000 96%
These results indicate that about 52,000 kg (or 96 percent of the baseline value) of SO2
deposition could be avoided to the reef and island network surrounding Veracruz if all vessels
calling on the Port were to move to a fuel switching regime within 24 nm of shore.
Health and Environmental Effects
In its Proposal to the I MO regarding the Designation of a North American Emission Control Area
to Reduce Emissions from Ships19, the US EPA indicated that ships "generate emissions that
18 Values may not sum correctly due to rounding.
19 Proposal to Designate an Emission Control Area for Nitrogen Oxides, Sulphur Oxides and Particulate Matter, Submitted by the
United States and Canada to the International Maritime Organization (IMO) Marine Environment Protection Committee, 2 April
2009, especially Annex 1. Available at http://www.epa.gov/oms/oceanvessels.htm.
ICF International 30
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
elevate on-land concentrations of harmful air pollutants such as PM2.5 and ozone, as well as
SOxand NOx. Human exposure to these pollutants results in serious health impacts such as
premature mortality and aggravation of heart and lung disease."
The US EPA has indicated20 that particle pollution generally, and fine particles (PM2.s)
particularly, consist of solids and liquids in such microscopic sizes that they are easily inhaled
deeply into the lungs where they can cause serious health problems. These health problems
include:
• Respiratory effects, such as irritation of the airways, coughing, or difficulty breathing
• Decreased lung function
• Aggravated asthma
• Development of chronic bronchitis
• Irregular heartbeat
• Heart attacks,
• Premature death, and
• More subtle indicators of cardiovascular disease.
People with heart or lung diseases, children and older adults are considered particularly
sensitive to particulate air pollution, although all people may experience temporary symptoms
from exposure to elevated levels of particle pollution.
In addition to direct human health effects, PM2.5 is responsible for other "welfare" effects,
including a degraded environment. Environmental effects of PM2.s include:
• Visibility reduction: Fine particles (PM2.s) are the major cause of reduced visibility (haze)
• Environmental damage. Particles can be carried long distances before settling to ground
or water surfaces where they can acidify lakes and streams, alter the aquatic nutrient
balance, deplete nutrients from the soil, damage forests and crops, and affect
ecosystem diversity.
• Aesthetic damage: Particle pollution can also stain and damage stone and other
materials, including culturally important objects such as statues and monuments.
• Climate change: Particles can influence the radiative balance and influence climate.
Although, globally, particles are thought to cool the planet through both direct and
indirect effects, some species, such as black (elemental) Carbon act as warming
agents.21
In addition to the general PM health effects, EPA and other agencies have noted that exposure
to particulate matter from diesel exhaust (DPM) has also been associated with additional
20 http://www.epa.gov/air/particlepollution/health.html
21 See, for example, http://www.ipcc.ch/ipccreports/tar/wg1/index.php?idp=160
ICF International 31
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
adverse health effects. Marine diesel engines emit DPM, a complex mixture of particulate
compounds that consists of fine particles (< 2.Sum), including a subgroup with a large number
of ultrafine particles (< 0.1 urn) that adsorb organic compounds, are easily respirable, and
consist of several organic compounds that have mutagenic and carcinogenic properties. In
EPA's 2002 Diesel Health Assessment Document (Diesel HAD), inhalation of diesel exhaust
was classified as a likely human carcinogen. Some studies also investigate the impact of ship
emissions on climate and air quality, including through characterizing emissions of black
carbon22'23.
This study has indicated that local concentrations of PM2.s pollution could be reduced as much
as 43 to 88 percent over the entire modeling domain by moving to a fuel-switching mode for
ships calling on the Port of Veracruz.
The US EPA has indicated that there is significant scientific evidence linking short-term human
exposures to concentrations of SO2 in the air to an array of adverse respiratory effects. (Note
that several of these effects are interrelated to sulfate exposure through particulate matter.)
These health effects include bronchoconstriction and increased asthma symptoms, and are
particularly important for asthmatics, especially during episodes of elevated breathing (such as
during exercise). Short-term exposures to SO2 are correlated to increased hospital admissions
for respiratory illnesses, particularly for children, the elderly, and asthmatics.
Environmental effects of increased concentrations of SO2 include acidification of lakes and
streams through deposition, accelerated corrosion of buildings and monuments, and reduced
visibility.
Note that adverse effects are also attributable to other gaseous sulfur oxides (e.g. SO3), which
are also linked to exhaust emissions. However, they tend to be at concentrations much lower
than that of SO2. Thus the primary effects can be determined by studying SO2 concentrations
alone.
Studies have shown that atmospheric inputs related to emissions from fossil fuel combustion
and other sources of strong acids (such as nitric (HNO3) and sulfuric (H2SO4) acids) alter
surface seawater alkalinity, pH, and inorganic carbon storage which can disrupt natural
biogeochemical cycles. This is expected to have the greatest impact in near-coastal waters,
where the ecosystem responses to ocean acidification most affect the human population.24
Sulfate emission in particular, and thus Sulfuric acid deposition, may be mitigated with switching
to lower sulfur fuels. This study has indicated that annual SO2 concentrations over the entire
modeling domain could be reduced 46 to 96 percent by moving to a fuel-switching mode for all
22 E.g.: Lack et al., Particulate emissions from commercial shipping: Chemical, physical and optical properties, J. Geophys.
Res., vol 114, 2009.
23 E.g.: Lauer et al., Global model simulations of the impact of ocean-going ships on aerosols, clouds, and the radiation budget,
Atmos. Chem. Phys., vol. 7, 2007, p5061-5079.
24 E.g.: Doney et al., Impact of Anthropogenic Atmospheric Nitrogen and Sulfur Deposition on Ocean Acidification and the
Inorganic Carbon System, Proc. Nat. Acad. Sci., September 11,2007, vol. 104, no. 37, p14580-14585.
ICF International 32
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
vessels calling on the Port of Veracruz. Similarly, deposition of SO2 to sensitive reefs and
islands off the coast could be reduced 96 percent, or by about 52,000 kg annually.
In its IMO application, the US and Canadian Governments concluded that emissions from ships
contribute to a large number of adverse human health impacts and that designation of the
proposed EGA would reduce the risk of premature mortality and contribute to the avoidance of
many morbidity-related health impacts. While acknowledging that this study has not quantified
the effects of fuel switching on overall concentrations of air pollutants, as noted on Page 23, the
reductions of PM and SOx concentrations associated with fuel switching imply that similar
results could be achieved in Mexico through reduced use of HFO fuel in shipping.
ICF International 33
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Benefits of Fuel Switching
ICF International 34
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
4. Fuel Switching Demonstrations
EPA partnered with two maritime shipping companies (Maersk Line and Hamburg Sud) to
demonstrate fuel switching on two container ships which regularly travel between the Port of
Houston and Mexico. The first fuel switch demonstration was conducted on the Maersk
Roubaix in November 2009. The second fuel switching demonstration was conducted on the
Hamburg Sud Cap San Lorenzo in April 2010 and a third was conducted on this ship at the Port
of Santos in August 2010. EPA did not estimate or measure emissions reductions at the Port of
Santos for this report.
4.1. Demonstration Design
The first component of the demonstration design was the identification of the vessel type to
include in the study. EPA identified vessels most representative of the shipping fleet calling on
the Port of Houston and Mexican Gulf ports and thus potentially contributing the most to ship
emission affecting those areas. As can be seen from Figure 7, the most common ship types
which call at Port of Houston were tankers followed by container ships. As shown in Figure 27,
32 percent of the ships that stopped at Port of Houston travelled to Mexico.
Figure 27: Ship Destinations from Port of Houston
East Europe
2%
Central America
6%
Caribbean
9%
Of the ships that went to Mexico from Port of Houston, 43 percent went to Alta Mira and 18
percent went to Veracruz as show in Figure 28. Typically container ships travel to Alta Mira and
Veracruz while tankers tend to go to Cayo Areas, which is an oil terminal in the Gulf over 90
miles from shore. As the project goal was to demonstrate the impact of ship emissions
reductions near land, EPA chose to test fuel switching on container ships which typically go into
ports on the coast and thus - all other things being equal - would tend to have greater impacts
on coastal and inland air quality.
ICF International 35
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
The general fuel switching demonstration design included switching from HFO to MGO before
reaching 24 nautical miles from the coast line and traveling into and out of the port, operating at
least the main and auxiliary engines on MGO as shown in Figure 29.
Figure 28: Mexican Port Destinations from Port of Houston
Pajaritos
Tuxpan
Progreso
3%
Coatzacoalcos
9%
Figure 29: Schematic of Fuel Switching Demonstration Design
Switch from
ROtoMGO
Cruise on MGO
24 nm
ICF International 36
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
4.2. Fuel Switching Logistics
Ship Operation on HFO
Since HFO may contain contaminants and other components that would reduce the reliability of
the fuel injectors, these materials must be separated or removed from the fuel oil prior to use.
Additionally there is a viscosity specification for the fuel oil to ensure proper operation of the fuel
pumps. Equipment is on a vessel to clean the fuel oil and maintain the proper viscosity.
Although, generally heavier fuels require more complex fuel treatment systems, all systems
prevent heavy fuel oils from solidifying in the fuel system, improve operational efficiency, and
maintain the fuel circulation, injection, and combustion systems. These systems consist of
storage and settling tanks, filters, and purifiers.
As shown in Figure 30, fuel is transported from storage tanks to the settling tank by transfer
pumps. Settling tanks hold enough fuel for approximately 2 days of travel and have coils to heat
the fuel. If heating is not maintained, the fuel will become too viscous to pump. In the settling
tanks, heavy fuel solids settle to the bottom while fuel to be burned is drawn from the top of the
tank. Fuel is then pumped from the settling tank through a pre-heater and into one or more
centrifugal separators by feed pumps. This fuel is then pumped to the service (day) tank, where
approximately one day's reserve of pre-treated and cleaned fuel is maintained at an appropriate
temperature to maintain fuel viscosity for use in the engine. The engine fuel supply system then
draws fuel beyond that necessary for combustion from the day tank to the injection system and
circulates the additional fuel back to the day tank to prevent solidification throughout the supply
system. Sets of supply and circulating pumps pressurize the system and transfer fuel from the
day tank, the final engine fuel filter, and injectors while a pre-heater and viscosity meter
maintain fuel viscosity throughout the fuel system.
ICF International 37
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Figure 30: Vessel Fuel System
HFO HFO
Supply Circulating
pump pump
Source: MAN B&W Diesel25
Most ships carry both residual and distillate fuels on board. Older ships would use distillate
fuels in their auxiliary engines while operating the main engine on HFO. Newer ships operate
both main and auxiliary engines on HFO but keep a supply of distillate fuel for fuel line cleaning
and startup. Because of the viscous nature of residual fuels (see Figure 31), they need to be
heated to a minimum of 95°C to be liquid enough to pump and be injected into the engine
cylinder. Distillate fuels are significantly less viscous and can be pumped and injected at
around 40°C. Thus when fuel switching from HFO to MGO, the fuel temperature must be
reduced from a minimum of 95°C to 40°C to make sure the fuel's viscosity does not get too low
for the fuel pumps.
25 MAN B&W, Operation on Low-Sulphur Fuels Two-Stroke Engines, available at:
http://www.manbw.com/article 005271 .html
ICF International 38
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Figure 31: Residual Fuel Unheated
Switching from HFO to MGO
When switching from HFO to MGO, the fuel temperature must be lowered slowly (~2°C per
minute) to prevent thermal shock to the fuel system. In addition, due to the solvent nature of
MGO, initial use of the fuel will tend to pick up solids and cat fines (catalyst material left in
residual fuel from the refinery process) which may plug fuel filters. Once the fuel system is
flushed, however, this should not be a problem. Also systems used to operating on HFO may
tend to leak initially when switching to MGO. This can usually also be remedied quickly.
Although all the above mentioned concerns are legitimate, it should be noted that Maersk26
illustrated that all its vessels switch both main and auxiliary engines to MDO with less than 0.2
percent sulfur within 24 nautical miles of their California destination port for main engines and
within 24 nautical miles of the California border for auxiliary engines. They have noted no
problems from this program to date after over 1700 port calls.
The Maersk study included 111 vessels and over 1000 fuel switches consuming 29.4 MT of
MDO per switch from April 2006 to January 2009. The resulting total emissions reduction has
been calculated at almost 900 tons per year, including a 95 percent SOx, 86 percent PM, and
12 percent NOx reduction (which includes low-NOx auxiliary mode). These reductions are
greater than anticipated by the program. In the Maersk study, all vessels used separate service
tanks for high- and low-sulfur fuels (DMA and DMB, with DMX for lifeboat engines and
emergency generator use) to minimize compatibility issues. Also, as all fuel switching in this
program is considered short term, they made no cylinder lube oil BN27 change. Maersk noted
that fuel switching is considered "normal engineering practice" and provides no special training
for its crews.
Maersk Line's Fuel Switch Experiences in California Waters, presentation by A.P. Moller-Maersk Group,
Regulatory Affairs, Maersk Marine Technology, April 27, 2010.
27 Base Number (BN or TBN) is a measure of the cylinder lube oil's ability to neutralize acid.
ICF International 39
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
4.3. Maersk Roubaix Demonstration
The Maersk Roubaix (IMO 9332688) is a 1,118 TElr8 container ship and is pictured in Figure
32. It was built in 2006. Specific ship details are provided in Table 5. Its regular route during
the demonstration is shown in Table 6. For this project, the Maersk Roubaix switched fuels at
two ports on this route - the Ports of Houston and Progreso.
Figure 32: Maersk Roubaix
Table 5: Maersk Roubaix Specifications
Ship Specifications
Container Ship- 1,118TEUs
Main Engine - MAN 7L58/64 - 9730 kW MSD
Auxiliary Engines - 3 x MAN 6L16/24 connected to 570 kW
440/220 V generators
1 x 1400 kW 440/220V shaft generator
Service Speed - 19.6 knots
Maximum Speed - 21.0 knots
Controllable pitch (CP) propeller
Fuel Capacity - 180 tonnes MGO - 1000 tonnes HFO
28 A TEU is a twenty foot equivalent unit used to measure container capacity. Standard containers like those shown in Figure 32
are typically 2 TEUs.
ICF International 40
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Table 6: Estimated Schedule for Maersk Roubaix
Port Name
Houston
New Orleans
Mobile
Progreso
Santo Tomas De Castilla
Belize
Puerto Cortes
Manzanillo
Cartagena
Barranquilla
Manzanillo
Puerto Limon
Puerto Cortes
Santo Tomas De Castilla
Houston
Country
USA
USA
USA
Mexico
Guatemala
Belize
Honduras
Panama
Columbia
Columbia
Panama
Costa Rica
Honduras
Guatemala
USA
Arrival Date
11/9/20097:00
11/11/200913:00
11/13/20098:00
11/15/200923:00
11/18/20096:00
11/19/20095:00
11/20/20096:00
11/23/20096:00
11/26/20092:00
11/27/20091:00
11/28/200918:00
11/29/200920:30
12/2/20097:00
12/3/20096:00
12/7/20097:00
Departure Date
11/9/200923:00
11/12/20090:01
11/13/200914:00
11/16/200914:00
11/18/200915:00
11/19/200918:30
11/20/200919:30
11/23/200921:00
11/26/200916:30
11/27/200921:00
11/29/20096:30
11/30/200911:00
12/2/200922:00
12/3/2009 14:30
12/7/200923:00
The Maersk Roubaix switched fuels from HFO to MGO before 24 nm from the entrance to the
Houston Ship Channel. The HFO had a sulfur content of 3.37% while the MGO had a sulfur
content of 0.14%29. The ship entered the Houston Ship Channel and docked at Barbour's Cut
on November 10, 2009. It left Barbour's Cut and the Houston Ship Channel on November 11,
2009. The ship then switched fuels back to HFO upon leaving the 24 nm boundary. On
November 16th, the Roubaix travelled to Progreso, Mexico, switching fuel back to MGO before
24 nm from the Port of Progreso. It left the same day and switched back to HFO once outside
the 24 nm boundary. The main and auxiliary engines and boilers were all operating on MGO
when the ship was within the 24 nm boundary at the Ports of Houston and Progreso.
No emission measurements were made, but emission reductions were calculated as discussed
in Appendix A. Total emission reductions due to fuel switching at the Port of Houston are
shown in Figure 33. Emission reductions due to fuel switching at the Port of Progreso are
shown in Figure 34. Emissions reduced at the two ports are 0.12 metric tonnes of NOx (6%),
0.24 metric tonnes of PM2.5 (87%) and 2.05 metric tonnes of SOx (94%) with a differential fuel
cost of $6,456, which represents only 2% of the total fuel costs for a round trip voyage shown in
Table 6.
Fuel Price
Fuel prices were estimated at $460/metric tonne for HFO and $740/metric tonne for MGO base
upon Bunkerworld.com prices for Houston on September 15, 2010.
29
Per documentation submitted to EPA by Maersk.
ICF International 41
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Figure 33: Estimated Emissions for Fuel Switch at Port of Houston
HFO MGO HFO MGO HFO MGO
NOx PM2.5 SOx
Figure 34: Estimated Emissions for Fuel Switch at Port of Progreso
u
'^
+•1
QJ
0.60
0.50
0.40
0.30
0.20
0.10
Boiler
• Auxiliary
•- • Main
HFO MGO
NOx
HFO MGO
PM2.5
HFO MGO
SOx
Operational Issues
Maersk indicated no operational issues when switching from one fuel to the other.
ICF International 42
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
4.4. Hamburg Slid Demonstration
The Cap San Lorenzo (IMO 9215684) is a 3,739 TEU container ship and is pictured in Figure
35. It was built in 2001. Specific ship details are provided in Table 7. Its schedule during the
demonstration period is shown in Table 8. Fuel switches were conducted while at the Ports of
Veracruz, Alta Mira and Houston from April 14-16, 2010. The fuel switch at the Port of Santos
in Brazil occurred during a port call in late August 2010.
Figure 35: Hamburg Siid Cap San Lorenzo
Table 7: Cap San Lorenzo Specifications
Ship Specifications
Container Ship - 3,739 TEUs
Main Engine - Sulzer 7RTA84 - 28,760 kW SSD
Auxiliary Engines -4 x MAN 9L28/32 connected to 1890 kW
440/220 V generators
Service Speed - 22.5 knots
Maximum Speed - 23.9 knots
Fixed pitch (FP) propeller
Fuel Capacity - 349 tonnes MGO - 6,062 tonnes HFO
ICF International 43
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Table 8: Estimated Cap San Lorenzo Schedule
| Port
Veracruz
Altamira
Houston
Cartagena
Suape
Santos
Rio Grande
Navegantes
Paranagua
Santos
Rio de Janeiro
Salvador
Puerto Cabello
Cartagena
Veracruz
| Country
Mexico
Mexico
USA
Columbia
Brazil
Brazil
Brazil
Brazil
Brazil
Brazil
Brazil
Brazil
Venezuela
Columbia
Mexico
| Arrive
4/16/20100:00
4/17/20105:00
4/19/20106:00
4/24/201022:00
5/4/20100:00
5/7/20107:00
5/9/201019:00
5/11/201015:00
5/11/201017:00
5/13/201020:00
5/15/201010:00
5/17/201017:00
5/24/20105:00
5/30/20101:00
6/4/20100:00
| Depart
4/16/201012:00
4/17/201023:00
4/20/201018:00
4/26/201022:00
5/4/201012:00
5/7/201022:00
5/10/20109:00
5/12/20107:00
5/13/20107:00
5/14/201020:00
5/15/201022:00
5/18/20105:00
5/28/201015:00
5/31/201013:00
6/4/201012:00
The Cap San Lorenzo switched fuels from HFO to MGO upon entering the Port of Veracruz.
The HFO had a sulfur content of 3.79% while the MGO had a sulfur content of 0.01 % . The
ship left Veracruz on April 17, 2010. It then entered and left the Port of Alta Mira on April 18,
2010. It entered the Houston Ship Channel and docked at Barbour's Cut on April 20, 2010.
The Cap San Lorenzo also demonstrated fuel switching while calling at the Port of Santos,
Brazil from August 27 to 29, 2010.
Emission Sampling Methodology
During the Cap San Lorenzo voyage from Veracruz to Houston, stack emissions were
measured from both the main and auxiliary engines while operating on MGO and HFO at
several engine loads and speeds. In addition, emissions were measured while the vessel
switched fuels. The methods for sampling and analysis of the gases and particulate matter
(PM) conformed to the requirements of ISO 8178-1.31
Testing Fuels
Testing was performed on both the residual and distillate marine fuels. The heavy fuel oil (HFO)
met ISO 8217:2005 specifications32 and this is the fuel that the engine typically operates on the
30 Per documentation submitted to EPA by Hamburg Sud - see Appendix C for Fuel Certificates of Analysis.
31 International Standards Organization. ISO 8178-1, Reciprocating internal combustion engines: Exhaust emission
measurement. Part 1: Test-bed measurement of gaseous particulate exhaust emissions, First edition 1996-08-15
32 International Standards Organization. ISO 8217. Petroleum products: Fuels (class F), Specifications of marine fuels. ISO 8217:
2005(E), 2005
ICF International 44
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
open seas. The Certificate of Analysis (CoA) for the HFO and distillate fuels was obtained and a
one-liter sample was drawn from the main engine and auxiliary engine final filter drains,
immediately upstream of the injector rail, for subsequent analysis. Appendix C shows these fuel
documents.
Engine Testing
Engine sampling involved the use of a partial flow dilution system with single venturi (see Figure
36). The concentrations of CO2 or NOx were measured in the raw exhaust gas and diluted gas
streams using an exhaust gas analyzer. The dilution ratio was determined from both the CO2
and NOx raw and dilute concentrations, and the dilution ratios agreed within 5%, as specified in
the reference method.31 PM was measured both continuously using a Nephelometer (TSI
DustTrak 8520) and during specified speed and load points with discrete filter samples. Filter
samples were later analyzed for elemental carbon (EC) and organic carbon (OC).33 SO2 is
calculated from fuel measurements. Figure 37 shows the engine room sampling system.
Additional details are below.
Figure 36: Schematic of the Emission Sampling System
.•—Secondary Dilution
DAF : Dilution Air Filter
VN : Venturi
SP : Sampling Probe
TT : Transfer Tube
EGA : Emission Gas Analyzer
CFO : Critical Flow Orifice
DNPH : 2,4Dinitrophenylhydrazine
MBC : Multi Bed Carbotrap tubes
PTFE : Polytetrafluoroethylene
(Teflon)
PUF : Poly Urethene Foam
To Vacuum Pump
Exhaust
33 EC (elemental carbon) represents dry participates. When measuring ship emissions, OC is typically made up of unburned fuel
or oil that surrounds the EC. (BC - black carbon - is measured by a specific absorption method and is typically used
synonymously with EC.)
ICF International 45
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Figure 37: Emission Sampling of Main Engine Exhaust
OPERATING CONDITIONS OF THE AUXILIARY ENGINE WHILE MEASURING EMISSIONS
Auxiliary engine testing was done at port with the engine operating according to the 5-modes of
the ISO-8178 D2 cycle. Since the operating system of the vessel only allows operation up to
75% before a second generator set will turn on, the resulting emissions factors are weighted for
each mode achieved. Setting up the lab and measuring the modal emissions at 3 modes in
triplicate for 2 fuels was done within a 12 hour period.
OPERATING CONDITIONS OF THE MAIN ENGINE WHILE MEASURING EMISSIONS
Since the testing was carried out at sea, it was difficult to match "in-use" engine operating
conditions with the operating conditions specified for the four modes in the ISO 8178 E-3 (Table
9) marine certification test. For example, the vessel was not operated at 100% power and data
for that point were instead collected for a limited time at 85% power. All other test modes were
incorporated into the vessel operation schedule.
Table 9: Engine Operating Conditions for the ISO 8178 E-3 Cycle
Rated speed Intermediate speed
Speed, %
Power, %
Weighting factor
100
100
0.2
91
75
0.5
80
50
0.15
63
25
0.15
Due to constraints such as voyage time, sea current, wave pattern, wind speed/direction, and
cargo load, the ISO 8178 E-3 load points were approximated as shown in Table 10.
ICF International 46
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Table 10: Operating Engine Load
Engine RPM Load (%) Engine RPM Load (%)
27
52
66
92
1.4
11.7
23.7
59.3
27
52
66
92
1.5
11.8
23.7
62.0
This project has a focus on the emissions benefits of switching to a cleaner burning fuel when
entering a port. Thus the project was designed to measure emission benefits at operating
modes used in approaching or departing a harbor, not necessarily only at the ISO load points.
The four operating modes sampled included measurement of both HFO and distillate fuel at:
1) Cruise mode measurements during the 24 nautical miles prior to the entrance of the
Houston Ship Channel
2) Operation within any reduced speed zones and within the Houston Ship Channel
3) Maneuvering operations from the port entrance to the dock
4) Operations at dock at the Port of Houston
ENGINE PERFORMANCE MEASUREMENTS DURING TESTING
On-board engine performance testing was done in accordance with Chapter 6 of the NTC.34
This included measurements of the variables listed in Table 11 as well as other engine settings
necessary to define engine-operating conditions, such as waste-gate, charge air bypass, and
turbocharger status.
34,
International Maritime Organization, Marine Environment Protection Committee: Prevention Of Air Pollution From Ships;
Report of the Working Group on Annex VI and the NOx Technical Code (MEPC 57/Wp.7/Add.2 3) April 2008.
ICF International 47
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Table 11: Engine Parameters Measured during Testing
Svmbol
HA
PC
P
J BUS
TK
•Zcaclin
T caclout
?Sea
y;.';f
Parameter
Engine speed
Charge air pressure at receiver
Brake power (as specified below )
Auxiliary power (if relevant)
Charge air temperature at receiver (if applicable)
Charge air cooler coolant inlet temperature
(if applicable)
Charge air cooler coolant outlet temperature
(if applicable)
Seawater temperature (if applicable)
Fuel oil flow (as specified below)
Dimension
min
kPa
kW
kW
K
K
K
K
kg/h
GASEOUS EMISSIONS FOLLOWING THE SIMPLIFIED MEASUREMENT METHOD (SMM)
Onboard measurements followed the general requirements of Chapter 5 of the NIC in order to
perform tests safely and with minimal interference to the engine. This includes arrangements for
the sampling of the exhaust gas and transfer of the sample through properly heated lines to the
analyzers. The concentrations of gases in the raw exhaust and the dilution tunnel were
measured with a Horiba PG-250 portable multi-gas analyzer, which can continuously and
simultaneously measure up to five separate gas components. Table 12 details the gases and
ranges sampled with the Horiba instrument. Additionally, a JUM THC Analyzer HFID 3-200 was
used to measure total hydrocarbons. Both instruments meet all the specifications of the NTC.
For quality control, Appendix 4 Calibration of the Analytical and Measurement Instruments of the
NTC was followed and analyzer checks were carried out periodically with calibration gases.
Table 12: Detector Method and Concentration Ranges for Gaseous Monitoring
Component
Nitrogen Oxides (NOX)
Detector
Heated Chemiluminescence Detector
Ranges
0-25, 50, 100, 250, 500, 1000, & 2500
Carbon Monoxide (CO)
Carbon Dioxide (CO2)
Sulfur Dioxide (SO2)
Oxygen
Total Hydrocarbons
(HCLD)
Non dispersive Infrared Absorption (NDIR)
Non dispersive Infrared Absorption (NDIR)
Non dispersive Infrared Absorption (NDIR)
Zirconium oxide sensor
Heated Flame lonization Detector (HFID)
ppmv
0-200, 500, 1000, 2000, & 5000 ppmv
0-5, 10, &20vol%
0-200, 500, 1000, & 3000 ppmv
0-10, &25vol%
0-10; 100; 1,000; 10,000ppm
GASEOUS MEASUREMENTS USING A PORTABLE EMISSION MEASUREMENT SYSTEMS (PEMS)
In addition to SMM sampling, gaseous emission concentrations were also monitored with a
Portable Emission Measurement System (PEMS). This system was set up in compliance with
40 CFR Part 1065 and sampled raw gases from the same location in the exhaust conduit as
used for the SMM using a transfer line meeting the NTC.
ICF International 48
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
PARTICULATE MATTER (PM) MASS EMISSIONS MEASUREMENTS
A raw particulate sampling probe was fitted close to and upstream of the raw gaseous sample
probe in the exhaust. In order to measure PM, a sampling probe was inserted into the end of the
dilution tunnel (>10 diameters downstream) and directed to a PM sample splitter that allowed up
to three samples to be collected. A cyclone separator, sized to remove particles >2.5um was fit
into the stream. From the separator, two lines were added with 47 Gelman filter holders, one for
collecting PM on a Teflon filter and the other for collecting PM on a quartz filter. Thus the flow in
the dilution tunnel was split into two fractions, a smaller flow for measuring PM mass and PM
properties and a much larger flow that was vented outside the vessel. With the partial dilution
approach for measuring gases and PM, it is critical for the dilution ratio be determined
accurately.
Simultaneous Teflon and quartz filters were collect at each operating mode and analyzed
according to standard procedures. The simultaneous collection of quartz and Teflon filters
allows an internal quality check of the PM mass. Teflon filters used to acquire PM mass are
weighted following the procedure of the Code of Federal Regulations (CFR) (40 CFR Part 86).
Total PM was collected on Pall Gelman (Ann Arbor, Ml) 47 mm Teflo filters and weighed using a
Cahn (Madison, Wl) C-35 microbalance. Before and after collection, the filters were conditioned
for 24 hours in an environmentally controlled room (RH = 40%, T= 25 ° C) and weighed daily
until two consecutive weight measurements were within 3 ug. PM samples were also collected
in parallel on 2500 QAT-UP Tissuquartz Pall (Ann Arbor, Ml) 47 mm filters that are
preconditioned at 600°C for 5 h. A 1.5 cm2 punch is cut out from the quartz filter and analyzed
with a Sunset Laboratory (Forest Grove, OR) Thermal/Optical Carbon Aerosol Analyzer
according to the NIOSH 5040 reference method (NIOSH 1996). All PM filters were sealed in
containers immediately after sampling, and kept chilled until analyzed.
MEASURING THE REAL-TIME PARTICULATE MATTER (PM) EMISSIONS
In addition to the PM mass measurements, UCR used a Nephelometer (TSI DustTrak 8520) to
monitor transient PM emissions. The DustTrak measures the light scattered by the aerosols
and provided real-time data on the stability of the PM stream while the filter mass is
accumulating. Because this method is strongly dependent on particle size and refractive index,
it is not considered a Federal Reference Method (FRM). In this project the DustTrak was
calibrated against the discrete mass samples on the Teflon filters to provide real-time
measurements.
Emission Sampling Results
The main (propulsion) engine was tested at four different speeds, typically dead slow (1.4%
load, 5.8 knots), half speed (11.7% load, 11.7 knots), full speed (23.7% load, 14.7 knots) and
top speed (59.3% load, 20.1 knots). SO2 emissions at the various load points show a 99.7%
reduction in SO2 emissions as a result of fuel switching (Figure 38). NOx emissions at the
various load points is show a 1 to 6 percent reduction depending on speed as a result of fuel
switching although the reductions are not statistically significant (Figure 39). PM2.s emissions at
the various load points show a 47% to 71% reduction due to fuel switching (Figure 40).
ICF International 49
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Speciated PM emission results35 show a significant decrease in sulfates (H2SO4) while OC
tends to increase particularly at half speed (Figure 41). Hamburg Sud engineers believe this
increase was due to increased amounts of unburned fuel being picked up by the MGO as this
was the first time the vessel had switched fuels for any length of time. In comparison, half
speed emission sampling for two other ships are compared in Figure 42. These ships have
been undergoing fuel switching regularly when entering California waters and show that typically
OC should stay the same or show a reduction when switching from HFO to MGO. Thus once
the Cap San Lorenzo starts regular fuel switching as part of the North American EGA, it is likely
that the OC on MGO should be similar to that on HFO.
25
20
I
.52 15
o
ts
£
§ 10
Figure 38: Propulsion Engine S02 Emissions
HFO(3.79%S)
MGO (0.01% S)
Dead Slow
Half
Full
Top Speed
35 Speciated PM emissions include elemental carbon (EC), organic carbon (OC) and sulfates (H2S04). EC is typically soot or dry
carbon particles also considered black carbon. OC is typically unburned oil or fuel that attaches to the EC. Sulfates are
formed from the fuel sulfur.
ICF International 50
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Figure 39: Propulsion Engine NOx Emissions
30
25
I 20
t>o
o
tS 15
c
o
8 10
E
4.0
3.5
2" 3.0
1
Is2-5
o
8 2'°
§ 1.5
1
il i.o
0.5
0.0
HfO{3.79%S)
MGO(0.01%S)
Dead Slow
Half
Full
Top Speed
Figure 40: Propulsion Engine PM2.s Emissions
,HFQ..{3,79%..S).
MGO(0.01%S)
Dead Slow
Half
Full
Top Speed
ICF International 51
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Figure 41: Propulsion Engine Speciated PIVh.s Emissions
EC OC H2S04
HFO MGO HFO MGO HFO MGO HFO MGO
Dead Slow Half Full Top Speed
Figure 42: Speciated PIVh.s Emissions Comparisons with Other Ships
0.0
HFO MGO
Cap San Lorenzo
HFO MGO HFO
West Coast Ships
MGO
ICF International 52
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Auxiliary engine emission measurements were made at 25%, 50% and 75% load. SO2
emissions at the various load points show a 99.7% reduction as a result of fuel switching (Figure
43). NOx emissions at the various load points is show an 11% reduction at 25% load but a 19%
increase at 50% load (Figure 44). In addition, the NOx emission factors are considerably lower
than the expected 13.7 g/kWh expected for Category 2 medium speed diesel engines. This
might be indicative of an older engine not running at peak performance. PM2.s emissions at the
various load points show a 52% to 67% reduction due to fuel switching (Figure 45). Speciated
PM emission results (Figure 46) show a significant decrease in sulfates (H2SO4) while OC tends
to increase particularly at 75% load. Again this is likely due to the MGO picking up unburned
fuel due to the fact that this was the first time the vessel had switched fuels for any length of
time.
Figure 43: Auxiliary Engine S02 Emissions
c
o
30
25
20
15
10
HFO(3.79%S)
MGO (0.01% S)
25%
50%
75%
ICF International 53
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Figure 44: Auxiliary Engine NOx Emissions
ao
14
12
10
8
o
ro
"- 6
c
o
'i/!
I 4
2
0
HFO(3.79%S)
MGO(0.01%SJ
25%
50%
75%
Figure 45: Auxiliary Engine PIVh.s Emissions
2.5
2.0 *
1.5
u
ro
.0
'J/5
in
E
1.0
0.5
0.0
25%
50%
75%
ICF International 54
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
0.0
Figure 46: Auxiliary Engine Speciated PIVh.s Emissions
HFO MGO HFO MGO HFO MGO
25% 50% 75%
Operational Issues
For the work performed during this study and along the test demonstration vessel route, the
0.01% MGO used was available only at the Port of Houston. According to the superintendent in
charge of the Cap San Lorenzo, the fuel switching process is simple and is part of normal
engine operational procedures. No special training is needed. He reported that the Cap San
Lorenzo's engines and fuel supply systems ran normally during the fuel switching exercise and
no difficulties were encountered. Tank capacity was sufficient to carry the needed lower sulfur
fuel for the demonstration (300 metric tonne capacity, 40 metric tonnes needed for the
demonstration).
The average sulfur level of the HFO available in Mexico was 3.8%36 which is higher than and
the global average which is estimated to be 2.7%.
Estimated Fuel Switching Emission Reductions
Using the emission factors determined from emission sampling, emission reductions at the three
ports were calculated using the methodology in Appendix A. Emission reductions due to fuel
switching at the Port of Veracruz are shown in Figure 47. Emission reductions due to fuel
SEMARNAT, Experiencias en Mexico relacionadas con las emisiones de grandes buques, presented in Veracruz, Mexico on
April 16, 2010. Available at http://www.epa.gov/international/air/workshopreport/Annexlll.pdf.
ICF International 55
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
switching at the Port of Alta Mira are shown in Figure 48. Emission reductions due to fuel
switching at the Port of Houston are shown in Figure 49. Emissions changes due to fuel
switching at the three ports are an increase of 0.01 metric tonnes of NOx (0%), 0.17 metric
tonnes of PM2.5 (54%) and 3.13 metric tonnes of SO2 (99.7%) with a differential fuel cost of
$10,171. This is approximately 2% of the voyage fuel costs for the round trip shown in Table 8.
Figure 47: Estimated Emissions for Fuel Switch at Port of Veracruz
0.90
HFO MGO
NOx
HFO MGO
PM2.5
HFO MGO
S02
Figure 48: Estimated Emissions for Fuel Switch at Port of Alta Mira
U)
01
.2
u
Auxiliary ""
Main
HFO MGO
NOx
HFO MGO
PM2.5
HFO MGO
S02
ICF International 56
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
Figure 49: Estimated Emissions for Fuel Switch at Port of Houston
HFO MGO
NOx
HFO MGO
PM2.5
HFO MGO
S02
ICF International 57
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Fuel Switching Demonstrations
ICF International 58
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
5. Summary of Key Findings
This project demonstrated that fuel switching in the Gulf of Mexico can lead to large emission
reductions of both PM and SOx emissions, as observed through on-board emission sampling
and corroborated by calculated emission reductions. Dispersion modeling showed a large
reduction in impacts on port area air quality and sensitive reefs due to fuel switching. While
acknowledging that this study has not quantified the effects of fuel switching on overall
concentrations or deposition of air pollutants, the reductions of NOx, PM and SOx
concentrations associated with fuel switching imply that similar results could be achieved in
Mexico through reduced use of HFO fuel in shipping.
Emissions monitoring showed that switching to low-sulfur marine fuel achieved significant
reductions in emissions of SOx and PM (2.5 micron in size) and small reductions of NOx - 89,
80 and 5 percent respectively -at a 2 percent increase in vessel operating costs, due to the
higher cost of lower-sulfur fuel.
Ship emission inventories were developed for the Ports of Houston, Veracruz and Alta Mira
using vessel port call data together with Lloyd's Register of Ships data. Annual emissions by
ship type, ship operating mode (e.g., maneuvering, hotelling, etc.), fuel type and fuel switching
zone boundary were calculated for each port. Tankers contributed most to annual emissions in
Houston, whereas containers were the largest sources of annual emissions for Veracruz and
Alta Mira. At all ports, the "cruise" operating mode contributed the most to total annual ship
emissions. At all ports annual emissions reductions of NOx, PM and SOx achieved through fuel
switching within a 24 nm fuel switching zone were calculated. For the Port of Houston NOx, PM
and SOx reductions were 5, 81 and 90 percent respectively; for Alta Mira NOx, PM, and SOx
reductions were 5, 76, and 84 percent respectively; and for Veracruz NOx, PM, and SOx
reductions were 5, 78, and 87 percent respectively. For the Port of Veracruz there was over a
four-fold increase in annual emissions reductions using a 200 nm fuel switching zone boundary
versus a 24 nm boundary. And for the Port of Alta Mira this increase in fuel switching boundary
resulted in a 5-fold increase in annual emissions reductions.
Dispersion modeling was conducted for the Port of Veracruz using the calculated emission
inventory. The modeling showed a large reduction in impacts of ship emissions on port area air
quality and sensitive reefs due to fuel switching within 24 nm of the port. Only emissions from
ships were modeled. The study did not include the impact of other sources on air quality, such
as those from all other activities at the port as well as all other regional sources. Air quality
modeling showed a seven-fold reduction in 24-hour average and annual average PM2.s
concentrations and a 24- to 25-fold reduction in 24-hour average and annual average SO2
concentrations. This study has indicated that local concentrations of PM2.s pollution could be
reduced as much as 43 to 88 percent over the entire modeling domain by moving to a fuel-
switching mode for ships calling on the Port of Veracruz. Deposition modeling showed a 99 per
cent reduction of SO2 deposition to sensitive reef areas off the coast of Veracruz.
ICF International 59
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Summary of Key Findings
Some key lessons learned from this project are listed below:
• Operational issues: During the demonstrations, the test vessels encountered no operational
issues of concern due to fuel switching.
• Fuel Storage: The fuel tank capacity used for MGO was sufficient to accommodate the fuel
needed for the fuel switching demonstration.
• Meteorological data availability for impact assessment: One of the most challenging
aspects for analysis of air quality impacts was obtaining the requisite meteorological data.
ICF International 60
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix A - Port Inventory Methodology
Estimating emission inventories generally involves applying emission factors37 to measures of
port activity across a range of activity sectors. The U.S. Environmental Protection Agency
(EPA) has developed a guidance document regarding the development of port emission
inventories.38 This appendix first summarizes the general methodology used for all ports then
discusses the specific methodology for each port.
General Methodology
The current practice to calculate emissions from OGVs is to use energy-based emission factors
together with activity profiles for each vessel. The emission factor is in terms of emissions per
unit of energy from the engine; emissions are the product of the emission factor and the power
needed to move the ship in a particular activity. The bulk of the work involves determining
representative engine power ratings for each vessel and the development of activity profiles for
each ship call. Using this information, emissions per ship call and mode can be determined
using the equation below.
E = PxLFxAxEF
Where E = Emissions (grams [g])
P = Maximum Continuous Rating Power (kilowatts [kW])
LF = Load Factor (percent of vessel's total power)
A = Activity (hours [h])
EF = Emission Factor (grams per kilowatt-hour [g/kWh])
Various data sources are available to those preparing port emission inventories. These include
Marine Exchange/Port Authority (MEPA) data, U.S. Army Corps of Engineers (USAGE)
entrances and clearances data, Lloyd's Register of Ships (Lloyd's Data), and Pilot data. The
importance and use of each are discussed below and shown in Figure 50.
OGVs vary greatly in speed and engine sizes based on ship type as described in Table 13.
Various studies break out vessel types differently, but it makes most sense to group vessel
types by the cargo they carry. Other characteristics that should be determined from Lloyd's
Data are the propulsion engine power and engine speed, maximum vessel speed, and engine
speed. Generally auxiliary engine power is not readily available but can be estimated from the
Port of Houston Port Air Emissions Inventory.39 Calculated auxiliary to propulsion power ratios
by ship type can be found in Table 14.
37 An emission factor is a representative value that attempts to relate the quantity of a pollutant released to the atmosphere with
an activity associated with the release of that pollutant. Marine emission factors are usually expressed as the weight
(commonly measured in grams) of pollutant divided by the energy (commonly measured in kilowatt-hours (kWh)) of the engine
used to produce that emission.
38ICF International, Current Methodologies in Preparing Mobile Source Port Related Emission Inventories, Final Report, April
2009. Available at http://www.epa.gov/sectors/sectorinfo/sectorprofiles/ports/ports-emission-inv-april09.pdf.
39 Starcrest Consulting Group, 2007 Goocfs Movement Air Emissions Inventory at the Port of Houston. Final Draft, January 2009.
Available at http://www.portofhouston.com/pdf/environmental/PHA-GM-AirEmissions-07.pdf.
ICF International 61
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix A - Port Inventory Methodology
Ship Type
Auto Carrier
Bulk Carrier
Figure 50: Data Sources and their Uses
Marine Exchange/
Port Authority
Ship-name
IMO number
Vesseltype
DWT
Flag of Registry
Vessel Speed
Date of Arrival
Time of Arrival
Date of Departure
Time of Departure
Lloyd's Register
of Ships
Shipname
IMONumber
Ship-type
DWT
Flag of Registry
Vessel Speed
Engine Type
Engine Power
Engine Speed
Build Date
Calculated Average
Vessel Movements
• Calls
• Shifts
• Time-in-Mode
• Fuel Switch
• Cruise
• Transit
• Maneuvering
• Hotelling
Container Ship
General Cargo
Miscellaneous
Average Vessel
Characteristics
Ship-type
Engine Type
DWT
Engine Power
Vessel Speed
Engine Speed
Build Date
J
Table 13: Oceangoing Vessel Ship Types
Description
Self-propelled dry-cargo vessels that carry containerized automobiles.
Self-propelled dry-cargo ship that carries loose cargo. Heavy load (HL)
carriers and self-unloaders (SU) are further defined.
Self-propelled dry-cargo vessel that carries containerized cargo.
Self-propelled cargo vessel that carries a variety of dry cargo.
Category for those vessels that do not fit into one of the other categories or
are unidentified.
Passenger
Reefer
Roll-on/Roll-off
(RORO)
Self-propelled passenger ships
Self-propelled dry-cargo vessels that often carry perishable items.
Self-propelled vessel that handles cargo that is rolled on and off the ship,
including ferries.
Tanker
Tugs
Self-propelled liquid-cargo vessels including chemical tankers, petroleum
product tankers, liquid food product tankers, etc.
Self propelled ocean going tugs.
ICF International 62
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix A - Port Inventory Methodology
Table 14: Auxiliary Engine Power Ratios
Ship Type
Auto Carrier
Bulk Carrier
Bulk Carrier, HL
Bulk Carrier, SU
Container 0-1000 TEUs
Container 1000-2000 TEUs
Container 2000-3000 TEUs
Container 3000-4000 TEUs
Container 4000-5000 TEUs
Container 5000-6000 TEUs
Container 6000-7000 TEUs
General Cargo
Miscellaneous
Passenger Ship
RORO
Reefer
Tanker 0-30K DWT
Tanker 30-60K DWT
Tanker 60-90K DWT
Tanker 90-120K DWT
Tanker > 120K DWT
Tugs
Auxiliary to
Propulsion
Power Ratio
0.472
0.248
0.390
0.401
0.220
0.220
0.224
0.127
0.253
0.154
0.152
0.236
0.094
0.278
0.706
0.281
0.356
0.200
0.212
0.207
0.169
0.018
Fuel type also is instrumental in determining emission factors and should be determined for
each port. It is assumed as a baseline that all OGVs operate their main propulsion and auxiliary
engines on heavy fuel oil (HFO). It is also assumed that both the main and auxiliary engines
are switched from HFO to marine gas oil (MGO) before they reach 24 nautical miles (nm) from
the coastline. Fuel switching times are estimated based upon total ship propulsion power as
shown in Table 15.
Table 15: Fuel Switching Times
Total Propulsion
Power
0-10, 000 kW
1 0,000-30,000 kW
> 30,000 kW
Fuel Switching
Time
0.50 hr
0.75 hr
LOOhr
ICF International 63
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix A - Port Inventory Methodology
Ship activity during a typical port call is best accomplished by breaking down the call into
sections that have similar speed characteristics. Vessel movements for each call are described
by using five distinct time-in-mode calculations. A call combines all modes. Each time-in-mode
is associated with a speed and, therefore, an engine load that has unique emission
characteristics. While there will be variability in each vessel's movements within a call, these
time-in-modes allow an average description of vessel movements at each port. Time-in-modes
should be calculated for each vessel call occurring in the analysis year over the waterway area
near the port. The time-in-modes are described in Table 16.
Activity
Fuel Switch
Cruise
Transit
Maneuver
Hotelling
Table 16: Vessel Movements and Time-ln-Mode Descriptions
Description
This is time during which the ship is switching fuels. This mode is modeled as operating
half the time on distillate fuel and half the time on residual fuel. Fuel switching is
estimated to occur at service speed. Fuel switching times are determined by total vessel
propulsion power.
Time at service speed considered to be 94 percent of maximum speed and 83 percent of
MCR. Calculated for 24 nm from the finish of fuel switching to the coastline.
For the Port of Houston, transit time is calculated for movements within the Houston Ship
Channel. There is no transit for the two Mexican ports.
Time in the port area between the breakwater and the dock. Maneuvering within a port is
assumed to occur at 3 knots on average.
Hotelling is the time at the dock when the vessel is operating auxiliary engines only.
Auxiliary engines are operating at some load conditions the entire time the vessel is
manned, but peak loads will occur after the propulsion engines are shut down. The
auxiliary engines are then responsible for all onboard power or are used to power off-
loading equipment, or both.
Load factors are expressed as a percent of the vessel's total propulsion or auxiliary power. At
service or cruise speed, the propulsion load factor is assumed to be 83 percent. At lower
speeds, the Propeller Law should be used to estimate ship propulsion loads, based on the
theory that propulsion power varies by the cube of speed as shown in the equation below.40
LF = (AS/MS)3
Where LF = Load Factor (percent)
AS = Actual Speed (knots)
MS = Maximum Speed (knots)
Load factors for auxiliary engines vary by ship type and time-in-mode. Auxiliary engines are on
all of the time, with the largest loads occurring during hotelling. Auxiliary engine load factors for
OGVs are given in Table 17.
40 When ships move against significant river currents, the actual speed in the above equation should be calculated based upon
the following: for vessels traveling with the river current, the actual speed should be the vessel speed minus the river speed;
for vessels traveling against the river current, the actual speed should be the vessel speed plus the river speed.
ICF International 64
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix A - Port Inventory Methodology
Table 17: Auxiliary Engine Load Factor Assumptions
Ship Type
Auto Carrier
Bulk Carrier
Bulk Carrier, HL
Bulk Carrier, SU
Container 0-1 000 TEUs
Container 1000-2000 TEUs
Container 2000-3000 TEUs
Container 3000-4000 TEUs
Container 4000-5000 TEUs
Container 5000-6000 TEUs
Container 6000-7000 TEUs
General Cargo
Miscellaneous
Passenger Ship
RORO
Reefer
Tanker 0-30K DWT
Tanker 30-60K DWT
Tanker 60-90K DWT
Tanker 90-1 20K DWT
Tanker > 120KDWT
Tugs
Cruise
0.15
0.17
0.17
0.17
0.16
0.16
0.16
0.16
0.16
0.16
0.16
0.17
0.17
0.80
0.15
0.15
0.13
0.13
0.13
0.13
0.13
0.17
Transit
0.30
0.27
0.27
0.27
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.27
0.27
0.80
0.30
0.30
0.24
0.24
0.24
0.24
0.24
0.27
Maneuver
0.45
0.45
0.45
0.45
0.55
0.55
0.55
0.55
0.55
0.55
0.55
0.45
0.45
0.80
0.45
0.45
0.33
0.33
0.33
0.33
0.33
0.45
Hotel
0.26
0.10
0.10
0.10
0.16
0.16
0.16
0.16
0.16
0.16
0.16
0.22
0.22
0.64
0.26
0.32
0.26
0.26
0.26
0.26
0.26
0.22
Emission factors for OGV propulsion are given in Table 18 for slow speed diesel engines (SSD),
medium speed diesel engines (MSD), gas turbines (GT) and steam turbines (ST) and are the
most generally accepted.41
The International Maritime Organization (IMO) adopted NOx limits in Annex VI to the
International Convention for Prevention of Pollution from Ships in 1997. These NOx limits apply
for all marine engines over 130 kilowatts (kW) for engines built on or after January 1, 2000,
including those that underwent a major rebuild after January 1, 2000. The required number of
countries ratified Annex VI in May 2004 and it went into force for those countries in May 2005.
Most manufacturers build engines to emit well below the standard. EPA determined the effect of
the IMO standard to be a reduction in NOx emissions of 11 percent below engines built before
2000.42 Therefore for engines built in 2000 and later, a NOx factor of 0.89 should be applied to
41 Entec UK Limited, Quantification of Emissions from Ships Associated with Ship Movements between Ports in the European
Community, prepared for the European Commission, July 2002. Available at
42 Conversation with Michael Samulski of EPA, May 2007.
ICF International 65
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix A - Port Inventory Methodology
the calculation of NOx emissions for both propulsion and auxiliary engines. Since this standard
only applies to diesel engines, the factor is not applied to either steam turbines or gas turbines.
Table 18: Emission Factors for OGV Main Engines, g/kWh
Fuel
HFO
MGO
During
Fuel
Switch
Stroke
MSD
SSD
STM
GT
MSD
SSD
STM
GT
MSD
SSD
STM
GT
Sulfur NOx
14.0
18.1
3.00%
2.1
6.1
13.2
17.0
0.10%
2.0
5.7
13.6
17.6
Varies
2.1
5.9
PM10
1.53
1.52
1.61
1.61
0.19
0.19
0.17
0.17
0.86
0.85
0.89
0.89
PM2.5
1.41
1.39
1.48
1.48
0.17
0.17
0.15
0.15
0.79
0.78
0.82
0.82
SOx
12.49
11.44
17.89
17.89
0.40
0.36
0.57
0.57
6.44
5.90
9.23
9.23
BSFC
213
195
305
305
203
185
290
290
208
190
298
298
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 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.
Low load adjustment factors should be multiplied by emission factors when the propulsion load
factor is less than 20 percent. Low load adjustment factors are given in Table 19.
No low load adjustment factor should be applied to diesel electric or gas turbine electric engines
for loads below 20% MCR because several engines are used to generate power, and some can
be shut down to allow others to operate at a more efficient setting.
As with propulsion engines, the most current set of auxiliary engine emission factors comes
from Entec. Table 20 provides these auxiliary engine emission factors. There is no need for a
low load adjustment factor for auxiliary engines, because they are generally operated in banks.
When only low loads are needed, one or more engines are shut off, allowing the remaining
engines to operate at a more efficient level.
Fuel prices were estimated at $460/metric tonne for HFO and $740/metric tonne for MGO43.
Inventory development specific to each port is discussed below.
43
Bunkerworld.com prices for Houston as of September 15, 2010.
ICF International 66
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix A - Port Inventory Methodology
Table 19: 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
PM SO2
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
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
Fuel
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
Table 20: Auxiliary Engine Emission Factors (g/kWh)
• Fuel
HFO
MGO
Sulfur NOx
3.00% 14.7
0.10% 13.9
During Fuel Switch 14.3
PMio PM2.5
1.54 1.42
0.18 0.17
0.86 0.79
SOx
13.31
0.42
6.87
BSFC I
227
217
222
Port of Houston
The specific methodology for the Port of Houston inventory development is discussed below.
Call Data
U.S. Army Corps of Engineers Entrances and Clearances data44 for 2007 was used to develop
the port emissions inventory for the Port of Houston. The Maritime Administration (MARAD)
maintains the Foreign Traffic Vessel Entrances and Clearances database, which contains
statistics on U.S. foreign maritime trade. Data are compiled during the regular processing of
statistics on foreign imports and exports. The database contains information on the type of
vessel, commodities, weight, customs districts and ports, and origins and destinations of goods.
I MO number was used to link the Call Data to Lloyd's data to determine ship characteristics.
44 http://wvyw.ndc.iwr.usace.army.mii//data/dataclen.htm
ICF International 67
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix A - Port Inventory Methodology
Transit Times and Distances
All ships except containers, passenger ships and tankers were assumed to dock at the Port of
Houston turning basin, 51.2 nm down the Houston Ship Channel. Bulk carriers, general cargo
and RoRos travelled that distance at an average speed of 8.2 knots. Auto carriers and reefers
travelled the distance at an average speed of 10.1 knots. Passenger ships were assumed to
stop at the Bayport facility which is 35 nm down the Houston Ship Channel at an average speed
of 11.5 knots. All container ships except those operated by CMA CGM and MSC were assumed
to stop at Barbour's Cut, 32 nm down the Houston Ship Channel at an average speed of 12
knots. Container ships operated by CMA CGM and MSC were assumed to stop at Bayport with
an average speed down the Houston Ship Channel of 11.5 knots. Tankers were assumed to
stop at the Jacintoport Terminal, 41.5 nm down the Houston Ship Channel at an average speed
of 8.9 knots. All distances and speeds were derived from the Port of Houston 2007 emissions
inventory.45
Maneuvering Times
Maneuvering times were estimated from the Port of Houston 2007 emissions inventory45. The
maneuvering times per call used in the inventory calculations in this report are shown in Table
21. All maneuvering was assumed to occur at an average of 3 knots.
Table 21: Port of Houston Maneuvering Times per Call
Ship Type Mrs per Call
Auto Carrier 1.5
Bulk Carrier 2.0
Container Ship - Barbour's Cut 1.0
Container Ship - Bayport 0.5
General Cargo 2.0
Passenger Ship 0.5
Reefer 1.5
RoRo 2.0
Tanker 1.0
Hotel ling Times
Average hotellir
The hotelling times per call used in this report are shown in Table 22.
Average hotelling times were also taken from the Port of Houston 2007 emissions inventory45.
45 Starcrest Consulting Group, 2007 Goocfs Movement Air Emissions Inventory at the Port of Houston. Final Draft, January 2009.
Available at http://www.portofhouston.com/pdf/environmental/PHA-GM-AirEmissions-07.pdf
ICF International 68
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix A - Port Inventory Methodology
Table 22: Port of Houston Hotelling Times per Call
Ship Type Hours per Call
Auto Carrier 21.0
Bulk Carrier 71.3
Bulk Carrier, HL 63.3
BulkCarrier.SU 71.3
Container 0-1000 TEUs 36.5
Container 1000-2000 TEUs 36.5
Container 2000-3000 TEUs 38.4
Container 3000-4000 TEUs 41.6
Container 4000-5000 TEUs 44.2
Container 5000-6000 TEUs 73.7
Container 6000-7000 TEUs 66.1
General Cargo 49.2
Miscellaneous 46.5
Passenger Ship 12.1
Reefer 28.9
RoRo 31.1
Tanker 0-30KDWT 28.3
Tanker 30-60KDWT 34.2
Tanker 60-90KDWT 45.9
Tanker 90-120KDWT 29.9
Tanker > 120KDWT 34.6
Tugs 21.8
Mexican Ports
The specific methodology for the Ports of Alta Mira and Veracruz inventory development is
discussed below.
Call Data
Port call data for 200546 was obtained from SEMARNAT and used to develop the port emissions
inventory for the Ports of Alta Mira and Veracruz. Compared with Houston, port calls at Alta
Mira and Veracruz are relatively low — 1,138 and 1,446 compared to 5,778 at Houston. Call
data included ship name, general ship type, DWT, and date and time of arrival and departure.
IMO number was assigned based upon the ship name and DWTs. IMO number was used to
link the Call Data to Lloyd's data to determine ship characteristics.
Maneuvering Times
Maneuvering times per call for both ports were assumed to be 1 hour.
46
Only 2005 call data was available for Mexican Ports.
ICF International 69
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix A - Port Inventory Methodology
Hotel ling Times
Hotelling times for each ship call were calculated from the arrival and departure times. These
were used in the calculation of hotelling emissions. Average hotelling times for the two ports
are shown in Table 23 by ship type.
Table 23: Average Hotelling Times for Alta Mira and Veracruz
Average Hotelling Time per Call (hrs)
Ship Type =-, -
Alta Mira Veracruz
Auto Carrier 14.7 30.3
Bulk Carrier 68.0 105.5
Container 13.2 16.0
General Cargo 39.2 73.2
Passenger 15.5
Reefer - 95.6
RoRo 13.1 17.3
Tanker 30.3 34.9
ICF International 70
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix B - Dispersion Modeling Methodology Details
Sources of Meteorological Data
One of the most challenging tasks for analysis of air quality impacts was obtaining the requisite
meteorological data. Over the course of two months, we contacted numerous individuals in
several US and Mexican government agencies to obtain adequate data. The key individuals
who assisted us in our research include:
• Norma Angelica Tepoz Ortega (CONAGUA) - 3-hour data of Precipitation, Winds, and
Cloud Cover for years 2001-2010, somewhere in Veracruz
• Emmanuel Alvarez Ramirez (SMN) - 10 minute EMA47 data of Winds, Precipitation,
Temperature, Humidity, Pressure, Solar Radiation for 2008 and 2009 in C. Prevision and
Alvarado
• Antonio Luna Diaz Peon - April 2008 detailed hourly data from Veracruz. All needed
fields but only one month
• Alfredo Ruiz-Barradas (U. Md.) - Provided contacts for SMN and Sr. Luna
• Jorge Zavala (UNAM) - 10 minute tide gauge data of Winds, Pressure, Humidity,
Temperature48
• Dan Thompson, Caroline Corvington, Dan Beardsley (NOAA) - provided contacts within
CONAGUA
• Martin Medina (NOAA) - provided contacts within CONAGUA
• Juana Maria Tavarez Nieto (CONAGUA) - January 2009 detailed hourly data from
Veracruz, All needed fields but only one month.
Throughout this, we were not able to identify an existing "perfect" data set for modeling. The
closest dataset to what was needed were the observations taken in Veracruz and provided by
Sra. Tavarez and Sr. Luna. These records had all required fields, and we understood they were
available for three years, from 2007 to 2009. However, when we requested the remaining files
we learned they were damaged - most likely by Hurricane Karl, which struck Veracruz during
the Autumn of 2010 - and it was unclear if they would be available. Table 24 lists the various
data sources collected
47 Estacion Meteorologies Automatica. Data available at:
http://smn.cna.gob.mx/index.php7optiorFCom content&view=article&id=106:estaciones-automaticas&catid=6:slider
48 Servicio Mareografico Nacional, Universidad Nacional Autonoma de Mexico, Instituto de Geofisica. More information is
available at www.mareografico.unam.mx
ICF International 71
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix B - Dispersion Modeling Methodology Details
Table 24: Meteorological Data Record Sets
| | Record
Required
Time span all
Time step ~0, 12Z
Source
Location Merida and
Ciudad
Victoria
Surface wind speed
Surface wind
direction
Wind observation
height
Temperature
Temperature
observation height
Total sky and/or
cloud cover
Humidity
Precipitation
Morning upper air D
sounding
Other useful data
Pressure
Cloud height
Heat/radiation flux
Afternoon sounding D
data
Additional provided
I^K^
Apr-08
hourly
Uspd_m/s
Udir
Td
eld coverage,
type (at levels),
tot_sky_covera
ge
Tw, RH
Pcpjnm
StPres, SLP
CltHt
insol_hrs, min
vis
Ugst_m/s
Uavg_m/s
IHH
Feb, Mar 09
10 min
tide gauge
station in
Veracruz
(http://www.rn
areografico.un
am.mx)
19°11'31.42"
N 96° 7'24.79"
W
vto_vel
vto_dir
8 m
temp_atm
hum_rel
pres_atm
^•^KH
8/07-12/09 1/01-12/09
10 min 3-hour
EMAviaSMN at
Alvarado and C.
Prevision
(http://smn.cna. g
ob.mx/)
Prev.: 96°06'41", EST
19°08'34";
Alv.: 95°37'57",
18°42'54"
WSK kph, VTOVEL
WSMK_kph
Dir deg, VTODIR
WSMDir_deg
AvgTemp_C
(NA. estimated NUBOCT
from AvciSR)
AvqRh pet (but
corrected)
Rain mm PRE24,
PRETPO,
PRELAM
AvgBP_mbar
NUBBAJ
AvgSR_Wm2
^^m
Jan-09
hourly
Uspd_m/s
Udir
Td
eld
coverage,
type (at
levels),
tot_sky_cov
erage
Tw, RH
Pcpjnm
StPres, SLP
CltHt
insol_hrs,
min
vis
Ugst_m/s
Uavg_m/s
ICF International 72
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix B - Dispersion Modeling Methodology Details
0 NOAA GSD Sounding database (http://esrl.noaa.gov/raobs/)
1 Antonio Luna Diaz Peon - April 2008 - detailed hourly data from Veracruz, all
needed fields but only one month
2 Jorge Zavala (UNAM)- Feb, Mar 2009 - 10 minute tide gauge data of Winds,
Pressure, Humidity, Tempera
3 Emmanuel Alvarez Ramirez (SMN) - 1/2008-12/2009 (Alvarado), 8/2007-
12/2009 (C. Prevision) - 10 minute EMA data ofWinds, Precipitation,
Temperature, Humidity, Pressure, Solar Radiation
4 Norma Angelica Tepoz Ortega (CONAGUA) - 2001-2010 - 3-hour data of
Precipitation, Winds, and Cloud Cover for 10 years , somewhere in Veracruz
5 Juana Maria Tavarez Nieto - Jan 2009 - detailed hourly data from Veracruz,
same as #1
Meteorological Data Record Details
Since a single data set was not available that both covered a sufficient amount of time and
contained all required fields, we created one from the various pieces available. This included the
following:
• April 2008 and January 2009 were taken directly from the provided CONAGUA records.
These were provided in Excel™ spreadsheets with a single tab for each day of the
month. These data were aggregated to a single file for each month, translated to text
files, translated into a SAMSON-type data record which the AERMET model can read,
and then processed with AERMET for the AERMOD model.
• Other fields were taken from the 10-minute EMA station at Alvarado. These were
processed from monthly records into a single annual file, then processed from 10-minute
to hourly average values in the SAMSON49 format. This record provided all other
required fields except cloud cover and relative humidity.50 For those, the following was
done:
O Relative humidity was instead taken from the EMA observations at C. Prevision. In
the few cases where RH was invalid at both stations, it was estimated from the range
of valid measurements at other times with a random component
O Cloud cover was taken from the 3-hour records provided by Sra. Tepoz. These
records were interpolated into hourly values.
A complete set of these variables for 2008 from these sources was aggregated,
processed, quality assured, translated to the SAMSON format, and processed through
AERMET.
The full 2008, mixed source, AERMOD-ready, surface and upper air files were then combined
with the independently processed January 2009 and April 2008 data files. This has the
49 httg7Awww1jicdcjToaa.gov/Bub/soft/ra^
50 RH actually is included, but clearly incorrect for the Alvarado station for this period.
ICF International 73
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix B - Dispersion Modeling Methodology Details
advantage of having each processed separately with the AERMET tool, which allows different
surface characterization in each case.
Surface characteristics were determined by sectors within a 1 km radius of each meteorological
location. As detailed observations of land use were not available, it was estimated using Google
Earth. From images of the sites, sectors were apportioned and typical land use within each
sector approximated. Corresponding values of Bowen ratio ((30), albedo (a), and roughness (z0)
were then taken from the AERMET User's Guide.51 Although somewhat less precise than the
analyses from AERSURFACE, this approach is reasonable given the amount of data available
and the relatively homogenous and distinct land uses in the vicinity of each station (typically
farmland, wetlands, or water). Note that all stations lie inland by at least 1 km.
2008 and 2009 upper air data was taken from nearby Merida station. Observations are provided
by NOAA Global Systems Division (GSD).
Figure 51 shows the location of the observations within Veracruz for January 2009 and April
2008. These are data sets 1 and 5 from Table 24. Figure 52 shows the location of the two EMA
stations used for the remaining data other than cloud cover and upper air observations.
Figure 51: Location of Hourly Meteorological Observations in Veracruz
si USER'S GUIDE FOR THE AERMOD METEOROLOGICAL PREPROCESSOR (AERMET), U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards, Emissions, Monitoring, and Analysis Division, Research Triangle Park,
North Carolina 27711, November 2004, EPA-454/B-03-002.
ICF International 74
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix B - Dispersion Modeling Methodology Details
Figure 52: Location of 10-Minute EMA Meteorological Observations near Veracruz
C. Prevision
'% / Alvarado
• •
Veracruz . '*
For the modeling, the January 2009 observations were relabeled as 2008. This is evident in
Figure 16, which shows a wind rose from the entire modeling period, but is labeled January to
December, 2008. This is to avoid technical issues in the modeling with input meteorological data
spanning multiple years. The January 2009 records were processed as if they actually
represented January 2008, then combined with the records from February through December
2008. This is functionally equivalent to having records from a single year for all averaging
periods.
The discussion in Section 3.2 focuses primarily on the wind field. Other meteorological fields
also influence the dispersion of pollutants once emitted, but have a less direct influence than
winds. These are discussed above, although precipitation is worth noting separately as it may
significantly influence deposition. Table 25 shows the total precipitation from the composite
record used in the dispersion modeling compared to the climatological average. The composite,
annual record shows nearly twice the annual rainfall as the climatological average for Veracruz,
although it is unknown if this is due solely to inter-annual variation and accurate for Veracruz for
that year or is an artifact of the quality of the meteorological data collected or the disparate
locations included. For example, reliance on stations not in the immediate vicinity of the city
could have greater overall rainfall than Veracruz due to terrain or other local effects.
ICF International 75
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix B - Dispersion Modeling Methodology Details
Table 25: February 2008 through January 2009 Composite Precipitation
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Mean Total _ . . _ . . „ .
Rainfall52 Total Rainfall from
(mm Modeling Record (mm)
24.1
16.0
14.5
17.4
48.2
298.3
419.9
33.5
89.1
54.6
22.9
62.7
332.5
316.7
Month
Aug
Sep
Oct
Nov
Dec
Total
Mean Total Total Rainfall from
Rainfall Modeling Record
(mm) (mm)
323.1
358.6
152.9
59.6
24.4
1757.0
178.8
1214.6
682.4
429.5
74.1
3491.5
Other Model Inputs
The AERMOD model was run for each pollutant separately to accommodate the varying
emissions strengths when the ship was operating on high and low sulfur fuels. Emissions
occurring within the harbor were considered urban sources, with a corresponding regional
population of 702,394, while emissions in all other locations approaching and departing the port
in open water were considered rural. Polygon area sources were used to represent all
emissions, and were considered steady in time, with hourly emission densities showing no
seasonal or diurnal variation and fixed by the total annual emissions strength.
Both dry and wet deposition for gases and particles was employed. Particle deposition was
considered via AERMOD's "METHOD_2" option. For PM10 the fine fraction was estimated at 92
percent and the mass mean diameter as 0.77 jam. For PM2.s the fine fraction is, by definition, 1
and the mass mean diameter was estimated at 0.51 j^m. Gas dry and wet deposition
parameters were determined for SO2. In that case, gas diffusivity in air and water were
determined following Weseley53, with values of Da = 0.1246 cm2/sand Dw = 5.74E-6 cm2/s. The
Henry's Law constant for SO2 is 121.59 Pa m3/mol.54 Cuticular resistance for SO2 is taken as 30
s/cm.55 Sectors for gas deposition properties were estimated by looking out 50 km radially from
the center of the port and establishing predominate land use from Google Earth™ , while the
season for gas deposition was assumed to be lush tropical summer all year. Terrain data for the
region was determined by processing SRTM DEM 1-type commercial data through the AERMAP
preprocessor to determine elevation and hill height scales. This was obtained from
wwwjriapmart.com.
52 Source: WMO. Record available athttp://^^
53 M. L. Wesely, P. V. Doskey, and J. D. Shannon, Deposition Parameterizations for the Industrial Source Complex (ISC3)
Model, ANL/ER/TR-01/003, DOE/xx-nnnn, Argonne National Laboratory, June 2002.
54 R. Sander, Compilation of Henry's Law Constants for Inorganic and Organic Species of Potential Importance in Environmental
Chemistry, Available at: http://www.mpch-mainz.mpg.de/~sander/res/henry.html
55 Baldocchi, A Multi-Layer Model For Estimating Sulfur Dioxide Deposition To A Deciduous Oak Forest Canopy, Atmospheric
Environment Vol. 22, No. 5, pp. 869-884,1988.
ICF International 76
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix B - Dispersion Modeling Methodology Details
Model Execution
Each of these inputs was assembled into appropriate AERMOD input files following current
implementation guidelines56 and processed with the most recent version of the AERMOD model
(version 09292). Appropriate design values were determined within the model using the model's
PLOTFILE option and exported to a GIS application for plotting.
56 Addendum to the User's Guide for the AMS/EPA Regulatory Model - AERMOD (EPA-454/B-03-001, September 2004), U.S.
EPA, Office of Air Quality Planning and Standards, Research Triangle Park, North Carolina 27711, October 2009
ICF International 77
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix B - Dispersion Modeling Methodology Details
ICF International 78
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix C - Monitoring Methodology
Testing was performed on both the residual and distillate marine fuels. Figure 53 shows the
certificate of analysis (CoA) for the HFO sample. Figure 54 shows the certificate of analysis
(CoA) for the MGO sample. Figure 55 shows the fuel audit results.
Figure 53: HFO Fuel Certificate of Analysis
FROM
VI SWA LAB
TO
COLUMBUS SHIPMANAGEMENT GmbH
ATTN: MR. CHRISTOPH GESSNER
Fuel Sample
VLC Log No.
Bunk. Port and Date
Place and Date Sent
Supplier CHEMOIL
Date Received at VLC 03/05/10
Sample Type per Customer IFO 380
Grade RMG 380
Tamper Proof 0198357
SPECIFIED PARAMETERS FOR RMG 380
Density @ 15 degC
API Grade
Viscosity @ 50 degC
Viscosity @ 100 degC
Upper Pour Point
Carbon Residue
Ash
Water
Sulfur
Sediment
Vanadium
Al + Si
Flash Point
ADDITIONAL PARAMETERS
Minimum Transfer Temperature 42 degC
Injection Temperature (For 13 cSt Viscosity) 133 degC
Engine Friendliness Number (EFN: 1-100) 54
ICF International 79
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix C - Monitoring Methodology
COMMENTS
SUGGESTIONS & RECOMMENDATIONS TO SHIP OWNERS/OPERATORS/TECHNICAL STAFF
CARBON RESIDUE
Observation: Though within limit, carbon is high.
Improve purification efficiency through reduction of throughput and
maintaining temperature around 98=BOC in purifier heater. Higher air charge
temperature (within-limits) can reduce deposit formation. Inspect exhaust
passages and turbo charger and wash down if necessary.
Heat and store this fuel at 10=BOC above the measured pour point temperatur=
e.
Observation: Ignition delay is indicated by CCAI greater than 840 for
medium-speed engines and greater than 870 for low-speed engines.
OVERALL QUALITY:
Engine Friendliness Number (EFN) is a unique bench-mark of fuel quality
evaluated by VISWA LAB from the point of view of engine wear and tear
resulting from the use of this fuel. Based on EFN, which is calculated from
the analysis results listed in this report, the quality of this fuel is
above average.
WITH EFFECT FROM 20TH FEE 2010, KINDLY FORWARD SAMPLES BUNKERED IN ROTTERDA=
M
AND ANTWERP TO OUR LAB IN SINGAPORE USING THE APPROPRIATE AIR WAYBILLS
INCLUDED IN THE SAMPLE KIT BOX. THIS INSTRUCTION SUPERSEDES THAT GIVEN IN
THE WORLD MAP ENCLOSED IN THE KIT. IF THE FUEL IS PART OF A MACHINERY
PROBLEM INVESTIGATION, KINDLY FORWARD TO VISWA LAB HOUSTON. IF YOU HAVE ANY
QUESTIONS, PLEASE CONTACT US.
You can view this and earlier reports online at
http://www.viswalab.com/vlclogin.htm
Questions? Call Dr. R.Vis,
Tel(713)-842-1985 Fax(713)-842-1981
REPORT PREPARED Ms K Vis
Figure 54: MGO Fuel Certificate of Analysis
FROM
VISWA LAB
ICF International 80
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix C - Monitoring Methodology
TO COLUMBUS SHIPMANAGEMENT GmbH
ATTN: MR. CHRISTOPH GESSNER
Fuel Sample CAP SAN LORENZO
VLCLogNo. F100308631 Date 03/05/10
Bunk. Port and Date HOUSTON, USA - 03/03/10
Place and Date Sent
Supplier ADA
Date Received at VLC 03/05/10
Sample Type per Customer MGO
Grade DMA LS
Tamper Proof 0198351 : Sealed
Customer furnished data:
Density 860.1kg/m3
Viscosity 3 cSt
Quantity 40 M.Tons
SPECIFIED PARAMETERS FOR DMA LS
Density® 15 degC 853.0 kg/m3 (890.0 Max)
API Grade 34.30 ( 27.40 Min)
Viscosity @ 50 degC cSt ( Max)
Viscosity® 100 degC cSt ( Max)
Viscosity @ 40 degC 2.63 cSt
Upper Pour Point -9 degC ( 0 Max)
Carbon Residue 0.01 %wt. (0.30 Max)
Ash 0.003 %wt. (0.010 Max)
Water <0.01 %vol. (0.00 Max)
Sulfur 0.02 %wt. (0.10 Max)
Sediment 0.00 %wt. (0.00 Max)
Vanadium 4 wt.ppm ( 0 Max)
Al + Si < 1 ppm ( 0 Max)
FlashPoint > 70 degC ( 60 Min)
ADDITIONAL PARAMETERS
SI < 1 ppm
AL < 1 ppm
Na < 1 ppm
Ca 1 ppm
Fe 3 ppm
Pb < 1 ppm
Ni < 1 ppm
P < 1 ppm
Zn < 1 ppm
Mg < 1 ppm
CCAI
Calorific value 42.99 MJ/kg
Minimum Transfer Temperature degC
Injection Temperature (For 13 cSt Viscosity) degC
Engine Friendliness Number (EFN: 1-100)
GRADE CONFORMANCE
The fuel sample tested conforms to grade DMA LS.
COMMENTS
DENSITY WAS CONFIRMED BY REPEATED ANALYSIS. PLEASE NOTE THAT BUNKER
DENSITY IS HIGHER THAN THE LAB MEASURED DENSITY.
VISUAL APPEARANCE - CLEAR AND BRIGHT
WITH EFFECT FROM 20TH FEE 2010, KINDLY FORWARD SAMPLES BUNKERED IN
ROTTERDAM AND ANTWERP TO OUR LAB IN SINGAPORE USING THE APPROPRIATE
AIR WAYBILLS INCLUDED IN THE SAMPLE KIT BOX. THIS INSTRUCTION
ICF International 81
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix C - Monitoring Methodology
SUPERSEDES THAT GIVEN IN THE WORLD MAP ENCLOSED IN THE KIT. IF THE
FUEL IS PART OF A MACHINERY PROBLEM INVESTIGATION, KINDLY FORWARD TO
VISWA LAB HOUSTON. IF YOU HAVE ANY QUESTIONS, PLEASE CONTACT US.
You can view this and earlier reports online at
http://www.viswalab.com/vlclogin.htm
Questions? Call Dr. R.Vis,
Tel(713)-842-1985Fax(713)-842-1981
REPORT PREPARED Ms K Vis
Figure 55: Fuel Audit Results
SOUTHWEST RESEARCH INSTITUTE8
6»0 CULEBRA HD, 78238-5)66 • P.O. DRAWER 36S10 76228-0610 » SAM ANTONIO. TEXAS. USA • (210)684-5111 * WWW.SWRI.ORG
July 22, 2010
Charles Bufalino
University of California, Riverside
1 084 Columbia Avenue
Riverside, CA 92507
Phone No.:(951) 781 -5784
Fax: (951) 781-5790
Re: Fuel Analyses - PO# RT1 0297667
1.08,0511831.01.001
SXVRI WO # 53432
Dear Mr. Bufalino:
Analyses of your samples for Density and Sulfur by various methods have been completed, as
requested. The two samples were received in I L containers on July 14, 2010 and were in good
condition. Sample Identifications and test results are shown in the attached table.
Test aliquors were taken in accordance with ASTM test procedures. Analyses were performed in
accordance with the test procedure with no deviations or modifications. The analyses pertain only to
the sample(s) received by Southwest Research Institute and represent only a sampling of this batch.
This report shall not be reproduced except in flili without the express written permission of Southwest
Research Institute. If you have any questions please call me at (210)-522-6920.
Melissa T. Lcgg
Senior Research Scientist
Petroleum Products Research Department
Offiee of /\ittofflQtive Engineering — Div, 08
QMTL>/O
HOUSTON, TEXAS !713) 977-1377 • WASHINGTON, DC (301) 881-0226
ICF International 82
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix C - Monitoring Methodology
DATA SUMMARY FOR U.C. Riverside
July 22, 2010
SWRI WORKORDER #53432
D 4052 Density (API by Meter) at 60°F
Sample ID
API @ 60 F
(15.5C)
Specific Gravity
@60F
Density® 15.5C
MDO
34.2
0.8541
0.8536
HFO
11.7
0.9879
0.9872
D 2622 Sulfur - Wayelength pispersiye X-Ra
Sample ID
Sulfur, Weight %
Sulfur, ppm
MDO
0.01092
109.2
HFO
3.78968
37896.8
Engine testing
Since emission factors have to be determined after the emissions are stable for a few minutes
and then sampling continues for up to 5 minutes or longer as required by the ISO- 8178
protocol, the University of California at Riverside (UCR) was not able to achieve meaningful
emission factors at some of the operating modes, such as the maneuvering modes. Further the
vessel does not normally change fuels during maneuvering or within the channel due to safety
concerns so emission factors with statistical significance would be difficult to achieve. Instead,
trends were obtained based on calculations, such as by comparing the fuels emission factors on
different fuels at some of the modes of the ISO protocol and certain operating modes, although
not necessarily in the same locations to allow a rough comparison of emission benefits.
ICF International 83
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix C - Monitoring Methodology
GASEOUS EMISSIONS FOLLOWING THE SIMPLIFIED MEASUREMENT METHOD (SMM)
The concentrations of gases in the raw exhaust and the dilution tunnel were measured with a
Horiba PG-250 portable multi-gas analyzer, which can continuously and simultaneously
measure up to five separate gas components. Major features of the PG-250 include a built-in
sample conditioning system with sample pump, filters, and a thermoelectric cooler. The
performance of the PG-250 was tested and verified under the U.S. EPA ETV program.
For quality control, UCR followed Appendix 4 Calibration of the analytical and measurement
instruments of the NTC and carried out analyzer checks with calibration gases before and after
each set of tests. Instrument drift was held to less than 2%. Because the instrument measures
the concentration of five gases, the calibration gases will be a blend of several gases (super-
blend) made to within 1% specifications by Praxair (Los Angeles, CA). Specifications of the
Horiba instrument are provided in Table 26; JUM is similar.
Table 26: Quality Specifications for the Horiba PG-250
Repeatability ±0.5% F.S.{ INOx : £ 10Oppm CO : £ 1 QOQppm range)
±1.0%F.S.
Linearity
±2.0% F.S.
±1.0% F.
Driffl
PEMS monitoring was done in parallel with the SMM. In cases where test durations for a given
mode were less than 60 minutes, the auto zero function on the PEMS was turned off.
CALCULATION OF EMISSION FACTOR
The emission factor at each mode was calculated from the measured gaseous concentration,
the reported engine load in kilowatts (kW) and the calculated mass flow in the exhaust. An
overall single emission factor representing the engine is determined by weighting the modal
data according to the ISO 8178 -E-3 and ISO 8178- D2 requirements and summing them. The
equation used for the overall emission factor is as follows:
ITOl
Where:
AWM = Weighted mass emission level (HC, CO, CO2, PM2.s, or NOX) in g/kW-hr
gi = Mass flow in grams per hour at the ith mode,
PI = Power measured during each mode, including auxiliary loads, and
j = Effective weighing factor.
ICF International 84
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix C - Monitoring Methodology
The calculated emission factor is strongly dependent on the mass flow of the exhaust. Exhaust
flow rate was calculated as described below.
CALCULATION OF THE EXHAUST FLOW RATE ASSUMING THE ENGINE AS AN AIR PUMP
This method has been widely used for calculating exhaust flow rate in diesel engines, especially
stationary diesel engines. This method assumes the engine is an air pump, and the flow rate is
determined from displacement of the cylinder, recorded rpm, with corrections for the
temperature and pressure of the inlet air. This method assumes the combustion air flow equals
the total exhaust flow. However, for low-speed, two stroke engines, there could be scavenger air
flow while the piston is expanding and the exhaust valve is still open. This scavenger air would
not be included in the air pump calculation leading to under predicting the total exhaust flow and
the emission factors. The method works best for four stroke engines or for two-stroke engines
where there the scavenger air flow is much smaller than the combustion air.
ICF International 85
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix C - Monitoring Methodology
ICF International 86
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix D - Related Information
U.S.EPA Fuel Switching Project Site: www.epa.gov/international/fuelswitch.html
U.S.EPA Ocean Vessels and Large Ships Site: www.epa.gov/oms/oceanvessels.htm
U.S.EPA National Clean Diesel Campaign Site: http://www.epa.gov/cleandiesel/
California Air Resources Board Commercial Marine Vessel Site:
http://www.arb.ca.gov/ports/marinevess/marinevess.htm
American Petroleum Institute: API Fuel Switching Practices:
http://www.klgates.com/FCWSite/ballast water/air emissions/API Fuel Switching.pdf
ICF International 87
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
-------
Appendix D - Related Information
ICF International 88
This technical report does not necessarily represent final EPA decisions or positions. It is intended to present technical analysis of issues using
data that are currently available and were collected through this project. The purpose in the release of sue
------- |