Emissions From Maritime Shipping Sector In A Freight Context

James Corbett, P.E., Ph.D.

Presentation to OECD/ECMT JTRC WG on Transport GHG Reduction Strategies 21-22 May 2007

1 EE RA Freight is an important multimodal transport function

Shipping is an integral part of global trade

2 EE RA Freight is an increasing contributor to the economy

1.20

21% 1.00

0.80

0.60

Index (2000 = = (2000 1.0) Index 0.40

12% 0.20

10% - 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 3 EE RA U.S. GDP SUM of goods and services Air pollution and Climate Impacts from Ships

 Two reasons to reduce ship emissions:  Ships contribute to problems TODAY  Growth in shipping makes problems worse TOMORROW

 Other reasons (depending on perspective)  Controls more cost-effective than other modes  Impacts mitigation may be asymmetric

4 EE RA Mode share comparison

100,000

10,000

1,000

100 Cargo Volume (Gtkm) Volume Cargo

10

1 Road Shipping Aviation Rail 5 EE U.S. Freight EU25 Freight Seaborne Trade RA Ships are more heterogeneous than onroad transport

 Tug and towboats  1-30 barges: 0.5 - 4 MW  High speed ferries  150-350 passengers: 2-4 MW  Roll-on\Roll-off  200-600 vehicles: 15-25 MW  Tankers  250,000 tons of oil: 25-35 MW  LNG fleet: 20-30 MW  Container  1750 TEU: 20-25 MW  4300TEU: 35-45 MW  6000 TEU: 55-65 MW

6 EE RA Ship emissions estimates bounded th th 1.E+03 Whiskers: 5 and 95 bounds Boxes: 25th and 75th bounds 1.E+02 Points: Best estimates of various studies

Fuel and CO2 Traditional Air Pollutants and Black Carbon HFCs

1.E+01

1.E+00 Best Estimate: ~2.7% of anthropogenic CO Tg per Year per Tg 2 1.E-01

Best Estimate: ~15.4% 1.E-02 of anthropogenic NOx Best Estimate: ~7.8% of anthropogenic SOx 1.E-03

Methane Cargo HC Total NMHC Refrigerants Black Carbon Organic Carbon EE Registered Fleet PM 7 RA Registered Fleet Fuel Use Registered FleetRegistered CO2 (as FleetRegistered C) NOx (as Fleet N) SOx (asRegistered S) Fleet Engine HC Ship traffic differs by vessel type

Containership Tanker

Bulk Carrier General Cargo Refrigerated Cargo Ro-Ro

Passenger

Trade driven by commodity demand & resource supply

8 EE RA Trade import patterns are clear … … connected to domestic freight system

9 EE RA Different top-down proxies provide different regional pictures

And these differences may produce different impacts assessments

10 EE RA Forecasting Summary

 Power-based trends used for forecasting  First-order indicator of proportional change in emissions, adjusted for control measures  Forecasts are primarily extrapolations of BAU that can be bounded and/or adjusted  North American trends validated by comparison with other modal trends and ship trade-energy models, at multiple scales

 Ship emissions growth rates are faster than GDP

 Future emissions with IMO-compliant SECA will be greater than base year emissions in 2002.

11 EE RA Fast-growing sectors can dominate forecast

18,000

16,000 Data Forecast using Persistence 14,000

12,000 container tons, approx. reefer tons 10,000 dry bulker tons chemicals tanker tons 8,000 gas carrier tons oil tanker tons total tons

Million Metric Tons Million 6,000

4,000

2,000

- 12 EE 1980 1990 2000 2010 2020 RA U.S. Containership energy use driven by strong growth in “heavy-leg” activity

45 Note the significant activity to move empty containers 40 Trade-based models can account for this … through utilization factors, etc. 35

30

25 ~6.5% ~7%

20 Million TEUs Million 15

10 ~10%

5 Linkage: Containership cargoes generate truck, rail activity 0

1980 1985 1990 1995 2000 2005 13 EE RA TEU throughput Total Cargo TEUs Export TEUs Import TEUs Shipboard power trends indicate strong growth in energy demand

3,500

3,000

2,500

2,000

1,500

1,000 Average Installed Power (kW) Power Installed Average

500

0 1965 1970 1975 1980 1985 1990 1995 2000 2005 14 Building a valid range of world forecasts … starting with trade and energy 2.5 Implication: World (ocean) freight emissions on 2 track to double before 2050 (pre-2030?)

North America doubles between 2015-2020 1.5 China supplies NA and EU – faster growth?

1

Extrapolating trends since ~1980-85 depending on data source 0.5

0 1950 1960 1970 1980 1990 2000 2010 2020 2030

Seaborne Trade (tons) Seaborne Trade (ton-miles) OECD HFO Int'l Sales Seaborne Trade (trend since 1985) University Mellon CarnegieMorgan, GrangerM. with disucssionstocredited illustration Concept 15 World Marine Fuel (Eyring, 2005) Installed Power-This work Tanker Container Ship Next Steps

 Recognize different growth rates and re- forecast spatially  Consider policy and technology Bulk Carrier Reefer interventions  Couple better with economics trends  Go multimodal  Expand globally

Ro-Ro Passenger Vessel General Cargo

16 EE RA Approaches to setting ship targets

1. Reduce emissions to improve performance, irrespective of growth. DO SOMETHING SOON

2. Reduce emissions to hold current exposure (impacts?) constant at some base year, offsetting trade-driven growth in emissions. HOLD THE LINE

3. Reduce emissions by X amount, maintaining emissions reductions (impacts?) from some base year, despite trade growth. MITIGATE IMPACTS

17 EE RA Menu of options to be matched with strategies and fleet

 Environmental control technologies  Pre-combustion: e.g., water emulsions  In-engine: e.g., humidification  Post-combustion: e.g., SCR, scrubbers, PM controls Only technology (and cost) combos get multiple pollutants Nearly all carry CO2 penalties of 1-3% for retrofits

 Alternative marine fuels and energy systems Could double fuel price (freight rate ↑), and may require phase in

 Operational (behavior) changes Possible in short term, possible multimodal logistics effects EE RA Achieves reductions in CO2 and all pollutants (win-win) 18 (Marine) Freight Transport insights

 Technology will involve fleet retrofits and new-builds  Economics determines role of alternative fuels  0.5% SECA or lower may be justified in large regions  Health effects work ongoing, but SOx control benefits appear greater than control costs  Reducing SOx and NOx will modify climate assessments  Most abatements increases CO2; reduced emissions change ozone and indirect aerosol forcing  Market incentives promising at several scales  Operational logistics changes may involve all modes  Decades required to completely achieve change

19 EE RA GIFT Network Model (under development)

20 Total fuel cycle comparisons

Winebrake, J.J., J.J. Corbett, and P.E. Meyer, A Total Fuel Life-Cycle

Analysis Of Energy And Emissions From Marine Vessels, Paper No. 07- 21 0817, in Transportation Research Board 86th Annual Meeting, Transportation Research Board, Washington, DC, 2007 A modern fleet of ships does not so much make use of the sea as

exploit a highway. -- Joseph Conrad, The Mirror of the Sea, Ch. 22, 1906 Courtesy: Courtesy: National Maritime Museum,Britain

Discussion welcome

Contact: James J. Corbett, P.E. University of Delaware 22 EE [email protected] RA Telephone: 302-831-0768 Best practices for CMV inventories

 Step 1: Identify the vessel(s) to be modeled, and engines in service Uncertainty remains, but bounding is improving

 Step 2: Estimate the engine service 100% hours for the voyage or voyage 90% Cargo Fleet 203 Mtons Registered Fleet segment 80% 248 Mtons World Fleet 70%  Step 3: Determine the engine load 289 Mtons 60%

profiles, including power and duty 50%

cycle 40% 175 Mtons Percentile 248 Mtons  Step 4: Apply emissions or fuel 30% 209 Mtons consumption rates for specific 20% engine/fuel combinations 10% 0%  Step 5: Estimate emissions or fuel 100,000,000 150,000,000 200,000,000 250,000,000 300,000,000 350,000,000 Metric Tonnes Fuel consumption for the voyage or voyage International Bunker Statistics (with statistical error bars) Estimated Cumulative Distribution of Cargo Fleet Fuel Consumption segment Estimated Cumulative Distribution of Registered Fleet (Cargo + Non-Cargo Ships) Fuel Consumption Estimated Cumulative Distribution World Fleet (Cargo + Non-Cargo + Military Ships) Fuel Consumption Best estimates (Corbett and Koehler, JGR, 2003) Steps 6+: Assign emissions spatially and Lower specific fuel consumption rates Fewer at-sea and in-port days, assuming only 1% of vessels are laid up, lower specific fuel consumption temporally both in and out of port Fewer at-sea and in-port days, more days laid up, lower specific fuel consumption regions

24 [Corbett and Koehler, 2003; Corbett and Koehler, 2004] Fuel consumption over past 50 years

300 280

250 240 , , 2005

213 JGR 200

150 164 124 135.9 127 100 77 106.4 108.2

Million Metric Tons MetricFuel Tons Million 65 From From Eyring et al., Part 1, 50 Containerized shipping and globalization explains diversion between usage and IEA energy allocation inventories - 1940 1950 1960 1970 1980 1990 2000 2010 25 Fuel Use (Eyring et al) International Marine Fuel Statistics Looking for preferred technology frontiers among uncertain, variable alternatives

10.00

EMD 9.00

8.00

7.00

Cost (Thousands ofDollars) Optimal Choice Line 6.00 Option 1: SCR DPF Option 2: Water in Air 5.00 Option 3: Water Fuel Emulsion SCR 4.00 Option 4: Repower

Weighted Score Weighted Option 5: Alternate Fuel 3.00

2.00 Assumes ~25% fuel economy benefits for repowering 1.00 LNC Percent PM Reduction EGR 0.00 DOC IAF 0.00 0.20 0.40 0.60 0.80 1.00

FIE Weight Assigned to Environmental Performance (NOx reduction) Percent NOx Reduction Corbett, J.J., and D.S. Chapman, Decision Framework for Emission Control Technology Selection, Decision Analysis, Manuscript number AW-05-00171, Version 1, under revision, 2006.

40% 100%

DPF Increase NOx Increase both $43,275 20% ($17,941 - $134,171) Decrease CO2 NOx and CO2 80%

0% Operational Measures Traditional 60% Control -20% Technologies

Hydrogen fuel and 40% -40% Advanced propulsion DOC $2318 EMD ECTPerformance PM SCR Other fuels and prime movers 20% ($1062 - $7243) $266,187 -60% $71,525 ($42,572 - $787,803)

($27,103 - $220,327) Change in NOx Emissions (%) Emissions Change NOx in IAF LNC -80% Decrease NOx 0% $8526 $37,225 Decrease both ($3907 - $26,645) ($7596 - $111,014) Increase CO2 FIE EGR -100% $20,870 $30,336 26 NOx and CO2 ($4061 - 61,342) ($6332 - $90,549) -60% -50% -40% -30% -20% -10% 0% 10% 20% -20% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% Change in CO2 Emissions (%) ECT NOx Performance Source: Corbett, Marine Transportation and Energy Use, Encyclopedia of Energy, 2004. Benefit-cost ratio of fuel switching … and scrubbing $40,000 Not yet published. Do not copy or quote without permission$40,000 Wang, C., and J.J. Corbett, A Preliminary Estimation of Costs and Benefits $35,000 for Reducing Sulfur Emissions from Cargo Ships in the U.S. West Coastal $35,000 Waters, Paper 07-3402, in Transportation Research Board 86th Annual Meeting, Transportation Research Board, Washington, DC, 2007 $30,000$30,000 44 -5-5 3 -4 3 -4 $25,000$25,000 22 -3-3 $20,000 11 -2-2 $20,000 00 -1-1 $15,000$15,000

$10,000$10,000

Benefits of Reduction ($/ton SO2) ($/ton Reduction of Benefits BenefitsReduction of ($/ton SO2)

$5,000$5,000

$0$0 $530 $490 $450$450 $410$410 $370$370 $330$330 $290$290 $250$250 $210$210 $170$170 $130$130 $90$90 $50$50 $10$10 FuelFuel Price Price Premium Premium ($/ton) ($/ton) 27 Summary of IPCC forecast trends

IPCC SRES Emissions Scenarios Summary

600

500

400 A1 AIM GNP/GDP A1 AIM Oil Use Petroleum use represents all sectors A2 AIM GNP/GDP (dominated by passenger vehicles). 300 A2 AIM Oil Use Not a good trend line for marine cargo B2 AIM GNP/GDP transportation sector in isolation B2 AIM Oil Use 200 B1 AIM GNP/GDP B1 AIM Oil Use

100 ExoJoulesPower; for Trillion (EJ) GNP/GDP US$ for Cargo growth should follow GNP if US consumption is trade dominated 0 28 1990 2010 2030 2050 2070 2090 Bounding insights to transform policy debate, focus dialogue

14,000,000 Is further debate on forecast rates more useful 12,000,000 than consideration of reduction targets that offset growth in trade under range of trends? 10,000,000

8,000,000

6,000,000

4,000,000 Metric Tons SO2 (global) (global) SO2 Tons Metric 2,000,000

0 2000 2005 2010 2015 2020 2025 2030 2035

4.1% Growth (This work) 1.5% Fuel-sulfur at 4.1% Growth 1.0% Fuel-sulfur at 4.1% Growth 0.5% Fuel-sulfur at 4.1% Growth IMO GHG-study growth (3%) 1.5% Fuel-sulfur at 3% Growth 29 2002 Baseline North American Results: Hypothetical IMO-compliant SECA (1.5% S) reduces future emissions from BAU … but not compared to base year

 Reduces 700,000 metric  Increases by ~2 million

tons from 2020 no-SECA Mtons over 2002 base-year30 Freight transport and environment: multi-scale, multimodal challenge bigger than ships

More sustainable freight logistics, inventory, production, and consumption

Evaluation of economic drivers/barriers to innovation and diffusion of sustainable concepts, regulatory jurisdiction and standards

Design of transportation strategies to achieve economic, energy, and environmental goals that improve stewardship faster than growth

Forecasting trends and alternative mitigation pathways Intergrated measures of sustainable transportation beyond air pollution

Impacts analysis: Environmental and health effects Multi-scale characterization: Emissions inventories, fate-transport modeling

Emissions and discharges: Air pollution formation and control technologies

System attributes: Vessel or vehicle, engine, and propulsion design 31 Jurisdictional constraints

Can treaty consensus achieve International Treaties often are weaker these target ranges at all? Treaty than national laws

Enforced by Flag Only Enforceable Enforced by Port State by Treaty Nations State

National National National Some nations don't Sovereignty Sovereignty Sovereignty participate, even though their ships sail globally

Federal Laws

JurisdictionalEconomic instruments conflict occurs can whenwork Regional Concerns: at these levels faster than treaty State/Province Security national policy doesn't address Authority Pollution localor multinational or regional problemsor federal action Safe Navigation - if compatible at larger scales State Laws Ports try to address policy Port problems AND attract cargoes 32 Authority for regional economy Intermodal Comparisons:

Infrastructure Factors Source: Farrell, A.E., D.W. Keith, and J.J. Corbett, A strategy for introducing hydrogen introducing for strategyA J.J. Corbett, and Keith, A.E.,D.W. Farrell, Source: Emissions Carbon Fraction Size of No. of

2 3 intotransportation, (g/kg fuel) intensity of CO2 fueling fueling stations stations NOx CO ($/tC) (%) (power) 4 Marine 71 16 950 6 175 MW 28-40 Policy Energy Autos1 14 130 2300 56 2.7 MW 180,000 5

Aircraft 3 17 2100 8.7 240 MW 72 , 31

Heavy trucks 30 17 2800 16 20 MW 5,500 1357 (13),

Rail 76 9 3500 2.3

- 1367, 2003 1367, All figures for the United States. All figures rounded to two significant digits. (1) Includes both automobiles and light trucks. (2) Computed using estimated actual emissions and fuel use. (3) End user expenditures divided by carbon emissions. (4) Total of companies in the large U.S. ports providing international marine fuels (@ 4-10 per port). (5) Large hub airports Ships may be preferred niche market for new technology innovation 33 Building Empirical Network

~9000 segments & ~1700 ports ~170,000 ship trips/yr in North America

35 Spatial Distribution in Multimodal Context Spatial Distribution in Multimodal Context

36