European Commission Directorate General Environment Service Contract on Ship Emissions: Assignment, Abatement and Market-based Instruments Contract No: 070501/2004/383959/MAR/C1 Task 1 - Preliminary Assignment of Ship Emissions to European Countries Final Report August 2005 Entec UK Limited
Report for European Commission Directorate General Environment B-1049 Brussels BELGIUM Main Contributors Andriana Stavrakaki Emily De Jonge Christoph Hugi Chris Whall Will Minchin Alistair Ritchie Alun McIntyre Issued by Christoph Hugi European Commission Directorate General Environment Service Contract on Ship Emissions: Assignment, Abatement and Market-based Instruments Task 1 - Preliminary Assignment of Ship Emissions to European Countries Approved by Alistair Ritchie August 2005 Entec UK Limited Entec UK Limited 17 Angel Gate City Road London EC1V 2SH England Tel: +44 (0) 207 843 1400 Fax: +44 (0) 207 843 1410 c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc Certificate No. EMS 69090 Certificate No. FS 13881 In accordance with an environmentally responsible approach, this document is printed on recycled paper produced from 100% post-consumer waste, or on ECF (elemental chlorine free) paper
i Executive Summary Introduction This report presents the deliverables under Task 1 of the European Commission contract on Ship Emissions: Assignment, Abatement and Market-Based Instruments. The overall objective of this task is to make preliminary assignments of ship emissions to European countries. Emissions to be assigned are sulphur dioxide (SO 2 ), nitrogen oxides (NOx), volatile organic compounds (VOCs), particulate matter (PM) and carbon dioxide (CO 2 ), for the years 2000, 2010, 2015, and 2020. Seven different methods were applied to assign emissions to each EU25 Member State plus Bulgaria, Romania, Turkey and Croatia. Each of the methods was appraised against several criteria, within an overall multi-criteria analysis. The findings of this task will provide supporting information to the Commission in considering how emissions from international maritime traffic could be included in the National Emission Ceilings Directive (Directive 2001/81/EC, NECD) and will further highlight the relative contribution of ship emissions to overall EU emissions. Assignment methods The assignment methods to be investigated were a selection of top-down and bottom-up methodologies, summarised as follows: A. Assignment according to Location of Emissions ship emissions estimated in each country s inland waterways, ports, 12 and 200-mile zones; B. Assignment according to Flag of Ship ship emissions estimated for the flagged fleet of a country; C. Assignment according to Industry Fuel Sales Estimates ship emissions estimated for each country based on industry fuel sales estimates and generic emission factors for the fuel; D. Assignment according to Reported Fuel Consumption ship emissions estimated for each country based on reported fuel consumption and generic emission factors for the fuel; E. Assignment according to Freight Tonnes Loaded total ship emissions of the 29 countries estimated based on Method A (for 200 mile zones) and split among the countries based on their relative share of freight tonnes loaded; F. Assignment in proportion to National Emissions total ship emissions of the 29 countries estimated based on Method A (for 200 mile zones) and split among the countries based on their relative share of national emissions; and G. Assignment according to Country of Departure/Destination ship emissions estimated as in Method A but assigned to countries based on port of departure and destination.
ii The scope of this task is further outlined in Section 1 of the main section, with general aspects of the methodology and general assumptions given in Section 2 and specific details of each assignment method given in Sections 3 to 9, for methods A to G. Summary of results of preliminary assignment of emissions As expected, there are significant differences in the distribution of ship emissions between countries using these different assignment methods, and Sections 3 to 9 of this report should be referred to for detailed results. For each method, figures are presented in these sections on: Each country s share of total ship emissions of NOx, SO 2, VOC, CO 2 and Particulate Matter (PM) for the 29 European countries (EU25 + Bulgaria, Romania, Turkey and Croatia); Each country s ship emissions of NOx, SO 2, VOC, CO 2 and PM as a percentage of total emissions for that country (as given in the RAINS model); Preliminary NOx ship emissions for each country (for 2000, 2010, 2015 and 2020); Preliminary SO 2 ship emissions for each country (for 2000, 2010, 2015 and 2020); Preliminary VOC ship emissions for each country (for 2000, 2010, 2015 and 2020); Preliminary CO 2 ship emissions for each country (for 2000, 2010, 2015 and 2020); and Preliminary PM ship emissions for each country (for 2000, 2010, 2015 and 2020). Under the various methods, the countries with dominant assignments of emissions are briefly highlighted below. For assignment based on location of emissions (Method A), for the 12 mile zones the UK has the highest emissions assigned from ships, followed by a group including Greece, Germany, Italy, Netherlands, Spain, Denmark and France, each with broadly similar amounts of emissions. For the 200 mile zones, Italy has the highest emissions from ships, followed by Greece, Spain and the UK. For assignment based on flag of ship (Method B), Cyprus has the highest emissions from ships, followed by Malta, and for some pollutants Greece, while for others Germany. For assignment based on fuel sales (Method C) and fuel consumption (Method D), the Netherlands has the highest emissions from ships, followed by Spain, Belgium and Greece. For assignment based on freight tonnes loaded (Method E), the Netherlands has the highest emissions from ships, followed by the UK, Italy and France. For assignment based on national emissions (Method F), countries will be ranked in different orders, dependent on the particular pollutant. This is due to the fact that ship emissions are assigned based on national emission ceilings, and these ceilings show different proportions between pollutants for different countries. For assignment based on country of departure / destination (Method G), the UK, Italy and Spain have the highest emissions from ships.
iii Figures 1 to 7 show the relative share of emissions between EU 1 countries for each method, where 1.0 is the index given to the country with the largest share for the particular method concerned. Assignments of ship NOx emissions in 2000 have been used for illustrative purposes. Ship emissions other than NOx will, in general, show a similar pattern, with the exception of Method F, where emissions are assigned in proportion to national emissions and therefore different pollutants will have different patterns. Actual emissions assignments in tonnes are given later in the report. As was expected the figures depict that the allocated emissions and the relative ranking among the countries significantly varies between the different methods. 1 EU25 countries plus Bulgaria, Romania, Turkey and Croatia
iv Figure 1 1.0 Normalised emissions of countries for Method A (-). 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 Relative amount of ship emissions for European countries based on assignment by location (Method A), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) Method A (12 miles) Method A (200 miles) 0.0 AUT BEL DNK FIN FRA GER GRC IRL ITA LUX NLD PRT ESP SWE GBR CYP CZE EST HUN LVA 25 EU countries and 4 candidates LTU MLT POL SVK SVN BGR ROM TUR CRO Figure 2 1.0 Relative amount of ship emissions for European countries based on assignment by flag (Method B), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) Normalised emissions of countries for Method B (-). 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 AUT BEL DNK FIN FRA GER GRC IRL ITA LUX NLD PRT ESP SWE GBR CYP CZE EST HUN LVA 25 EU countries and 4 candidates LTU MLT POL SVK SVN BGR ROM TUR CRO
v Figure 3 1.0 Relative amount of ship emissions for European countries based on assignment by fuel sales (Method C), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) Normalised emissions of countries for Method C (-). 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 AUT BEL DNK FIN FRA GER GRC IRL ITA LUX NLD PRT ESP SWE GBR CYP CZE EST HUN LVA 25 EU countries and 4 candidates LTU MLT POL SVK SVN BGR ROM TUR CRO Figure 4 1.0 Relative amount of ship emissions for European countries based on assignment by fuel consumption (Method D), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) Normalised emissions of countries for Method D (-). 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 AUT BEL DNK FIN FRA GER GRC IRL ITA LUX NLD PRT ESP SWE GBR CYP CZE EST HUN LVA 25 EU countries and 4 candidates LTU MLT POL SVK SVN BGR ROM TUR CRO
vi Figure 5 1.0 Relative amount of ship emissions for European countries based on assignment by freight tonnes (Method E), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) Normalised emissions of countries for Method E (-). 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 AUT BEL DNK FIN FRA GER GRC IRL ITA LUX NLD PRT ESP SWE GBR CYP CZE EST HUN LVA 25 EU countries and 4 candidates LTU MLT POL SVK SVN BGR ROM TUR CRO Figure 6 1.0 Relative amount of ship emissions for European countries based on assignment by national emissions (Method F), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) Normalised emissions of countries for Method F (-). 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 AUT BEL DNK FIN FRA GER GRC IRL ITA LUX NLD PRT ESP SWE GBR CYP CZE EST HUN LVA 25 EU countries and 4 candidates LTU MLT POL SVK SVN BGR ROM TUR CRO
vii Figure 7 1.0 Relative amount of ship emissions for European countries based on assignment by country of departure / destination (Method G), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) Normalised emissions of countries for Method G (-). 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 AUT BEL DNK FIN FRA GER GRC IRL ITA LUX NLD PRT ESP SWE GBR CYP CZE EST HUN LVA 25 EU countries and 4 candidates LTU MLT POL SVK SVN BGR ROM TUR CRO Based on the above figure the ranking in Table 1 can be derived for the different allocation methods. In this table, a figure 1 indicates the country with the highest allocated emissions, 2 is the country with the second highest emissions, etc. Table 1 Country Ranking based on the emissions allocated to each country under the different assignment methods i.e. 1 = most allocated emissions A Location (12 Mile Zones) A Location (200 Mile Zones) B Flag C Fuel sales D Fuel consumption E Freight tonnes F National emissions G Departure/ Destination Austria AUT 24 24 24 23 22 22 20 25 Belgium BEL 11 13 22 3 3 6 14 7 Denmark DNK 4 6 5 10 10 5 18 12 Finland FIN 12 16 11 12 11 12 15 11 France FRA 8 5 12 5 5 4 6 4 Germany GER 3 8 3 7 7 7 2 6 Greece GRC 2 2 4 4 4 11 8 8 Ireland IRL 15 12 23 17 17 16 21 16 Italy ITA 7 1 7 6 6 3 3 2 Luxembourg LUX 28 28 15 23 22 21 26 29 Netherlands NLD 6 7 6 1 1 1 11 5 Portugal PRT 13 10 21 13 12 14 12 13
viii Country A Location (12 Mile Zones) A Location (200 Mile Zones) B Flag C Fuel sales D Fuel consumption E Freight tonnes F National emissions G Departure/ Destination Spain ESP 5 3 14 2 2 8 5 3 Sweden SWE 10 11 10 9 9 9 16 10 United Kingdom GBR 1 4 8 8 8 2 1 1 Cyprus CYP 19 18 1 16 16 24 25 18 Czech Republic CZE 26 26 27 23 22 27 9 26 Estonia EST 14 14 19 15 18 17 23 19 Hungary HUN 27 27 28 23 22 26 13 27 Latvia LVA 18 17 25 21 21 15 22 17 Lithuania LTU 23 23 18 19 19 20 19 20 Malta MLT 21 19 2 18 14 25 27 14 Poland POL 17 20 17 11 15 13 4 15 Slovakia SVK 28 28 26 23 22 28 17 28 Slovenia SVN 25 25 29 23 22 23 24 22 Bulgaria BGR 22 22 13 22 20 19 10 24 Romania ROM 20 21 20 20 22 18 7 23 Turkey TUR 9 9 9 14 13 10 28 9 Croatia CRO 16 15 16 23 22 29 28 21 Depending on the assignment method the uncertainty of the presented results is estimated in the range of ±15-45%. More detail is given in the respective sections and in Appendix E. Assessment of methods Sections 3 to 9 present the detailed findings of the assessment of each of the methods against a set of specific criteria, within an overall multi-criteria analysis. The following criteria were considered: Costs to calculate assigned emissions; Simplicity and transparency of assignment method; Data sources and quality; Consistency and accuracy; Degree of influence for countries on key variables; and Fairness and appropriateness. Within the remit of this study, the relative importance of these criteria was not identified. As such, it is outside the scope of this study to recommend any particular assignment method. A comprehensive appraisal will need to take into account various political, technical, legal, environmental and economic factors.
ix Table 2 depicts the summary of the multi-criteria analysis and the ranking of the assignment methods based on the overall score of the methods, where 0 is the worst score and 5 is the best. A discussion of each method against each criteria is given in the relevant sections. Table 2 Summary table of assessment of assignment methods (0 is worst score, 5 is best score) Criteria C1. Costs to Calculate Emissions C2. Simplicity and transparency of Assignment C3. Data sources reliability and data quality C4. Consistency and accuracy C5. Degree of influence for Countries on key variables C6. Fairness and appropriateness A Location B Flag C Fuel sales Assignment Method D Fuel consumption E Freight tonnes F National emissions G Departure/ Destination 1.5 1.5 4.5 4.5 2.5 2.5 1.5 2 2 3.5 3.5 3 3 2 4.5 3.5 2.5 2.5 2.5 3.5 3.5 4 2.5 2.5 2.5 2.5 3 2.5 3 3 2.5 2.5 2 1 2.5 4 1 2.5 2.5 2 0 2 Figure 8 shows graphically the assessment of the individual criteria for each method. Figure 8 Illustration of assessment of assignment methods 5 A B C D E F G 4 Assessment (0-5) 3 2 1 0 1. Costs to Calculate Emissions 2. Simplicity and Transparency of Assignment Method 3. Data sources 4. Consistency and reliability and data accuracy quality 5. Degree of influence for Countries on key variables 5. Fairness and appropriateness
x Figure 9 shows the aggregated result assuming, purely for illustrative purposes, equal weightings of the individual criteria. As it is not considered appropriate within the scope of this study to recommend specific weightings for the various criteria, it is not possible to draw any firm conclusions from this figure, although it is clearly useful in showing how the methods might compare if each of the criteria is given an equal weighting. Figure 9 Illustration of mean scores of multi-criteria assessment (Applying equal weighting to each criteria, for illustrative purposes only) 5 4 3 2 1 0 A B C D E F G For each method, a summary of the main advantages, disadvantages and potential areas for further investigation is given in Table 3 below. Table 3 Summary of main advantages, disadvantages and areas for further investigation of alternative assignment methods Assignment Method Main advantages Main disadvantages Potential areas for further investigation A Location Regarded as most fair / appropriate of the alternative methods, in terms of consideration for location of emissions. Has potential to be a relatively accurate method. Relatively expensive compared to top-down methods. Dependent on Member States having control over emissions in specified sea areas. Choice of sea area eg 12 mile, 200 mile zone, etc. Enhancements to database (in-port times, increased coverage of vessels (<500GT), ship specific emission factors, spatial resolution etc).
xi Assignment Method Main advantages Main disadvantages Potential areas for further investigation B Flag No major advantages in comparison to other methods. C Fuel sales Potentially cheap. Direct link to fuel sales a key driver for ship emissions. D Fuel consumption E Freight tonnes F National emissions G Departure / destination Potentially cheap. Direct link to fuel consumption a key driver for ship emissions. Direct linkage to freight a key driver for ship movements. No major advantages in comparison to other methods. Direct linkage to movements / activity associated with each country. No consideration for location. Potential for perverse incentives by encouraging registration of ships with flags of other countries. Flags don t necessarily represent good coverage of national fleets. No consideration for location. Currently available statistics not sufficiently comprehensive or (possibly) consistent for this method. No consideration for location Currently available statistics not sufficiently comprehensive or (possibly) consistent for this method. No account for non-freight traffic (eg ferries) which can be significant for some countries. No consideration for location. Potential for double counting freight movements. No correlation between national emissions and ship emissions. No consideration for location. No consideration for location. Careful consideration in how to assign emissions along a journey to departure / destination ports. On the basis of the disadvantages of this assignment method as currently presented, this method appears one of the least appropriate of those considered. Improving the accuracy, consistency and comprehensiveness of underlying statistics. Building in consideration for location. Legal definitions to prevent circumventing. Improving the accuracy, consistency and comprehensiveness of underlying statistics. Building in consideration for location. Legal definitions to prevent circumventing. Avoiding double counting of freight movements. Building in consideration for location. Legal definitions to prevent circumventing. On the basis of the disadvantages of this assignment method as currently presented, this method appears one of the least appropriate of those considered. Approaches for assignment to departure / destination ports. Building in consideration for location. Legal definitions to prevent circumventing. In general terms, assignment by location (Method A) has key advantages due to its intrinsic consideration of location of emissions and due to a relatively good potential accuracy. For air pollutants, location clearly has an important influence on the consequent environmental impacts. Assignment by location would appear to be consistent with assignment of land based emissions under the NECD, and the Sulphur in Marine Fuel Directive already sets a precedent for sea area based emission controls with the SOx Emission Control Areas covering Member States territorial seas, exclusive economic zones and pollution control zones. As such, this method is considered clearly worthy of further investigation as a potential means of assigning ship emissions. Areas for further investigation with this method include choices over the size of zone for assigning emissions and potential enhancements to the underlying database.
xii A number of other methods are also considered worthy of further investigation, namely assignment by fuel sales (Method C), fuel consumption (Method D), freight tonnes (Method E) and departure / destination (Method G), as they also have certain positive characteristics as shown in the above table. In particular, Methods C and D could be relatively cheap methods. However, amongst other factors, any further consideration of these would need to investigate how location could be addressed within the method, and how circumventing could be prevented, eg through legal definitions. Of the specific methods that have been considered in this study, the ones considered to have most limitations in assigning ship emissions include assignment by flag (Method B), and assignment in proportion to national emissions (Method F). For the former method this is due to flags not necessarily representing good coverage of national fleets, the potential for perverse incentives in decisions on which countries ships are flagged in, and the lack of consideration for location of emissions; and for the latter method there is no correlation between national emissions and shipping emissions. Each of the methods considered in this report, however, could be subject to potential modification, which may alter the assessment findings.
xiii Contents 1. Introduction and Scope 1 1.1 This report 1 1.2 Scope of study 1 1.2.1 Assignment methods 1 1.2.2 Pollutants 2 1.2.3 Years of interest 2 1.2.4 Geographic areas 3 1.2.5 Limitations of Scope 5 2. Methods and Basic Assumptions 7 2.1 Discussion of Different Assignment Methods 7 2.2 Assessment of the Different Methods 9 2.3 Basic Assumptions 12 2.3.1 Emission Factors 12 2.3.2 In-Port Activities 14 2.4 Role of New Technologies 17 3. Method A - Assignment According to Location of Emissions 19 3.1 Introduction 19 3.2 Method 19 3.2.1 Ship Emissions at Sea 19 3.2.2 In Port Emissions 21 3.2.3 Emissions on Inland Waterways (IWW) 21 3.2.4 Total emissions for Method A 26 3.3 Assigned Emissions 26 3.3.1 Emissions for 12 Mile Zones 26 3.3.2 Emissions for 200 Mile Zones 30 3.4 Assessment of Method 34 4. Method B - Assignment According to Flag of Ship 37 4.1 Introduction 37 4.2 Method 40
xiv 4.2.1 Ship Emissions at Sea 40 4.2.2 Ship Emissions in Ports 42 4.2.3 Ship Emissions from Inland Waterways 42 4.3 Assigned Emissions 42 4.4 Assessment of Method 47 5. Method C - Assignment According to Industry Fuel Sales Estimates 49 5.1 Introduction 49 5.2 Method 49 5.3 Assigned Emissions 50 5.4 Assessment of Method 55 6. Method D - Assignment According to Reported Fuel Consumption 57 6.1 Introduction 57 6.2 Method 57 6.3 Assigned Emissions 58 6.4 Assessment of Method 62 7. Method E - Assignment According to Freight Tonnes Loaded 65 7.1 Introduction 65 7.2 Method 65 7.3 Assigned Emissions 66 7.4 Assessment of Method 71 8. Method F - Assignment in Proportion to Land Based National Emissions 73 8.1 Introduction 73 8.2 Method 73 8.3 Assigned Emissions 73 8.4 Assessment of Method 77 9. Method G - Assignment According to Country of Departure / Destination 79
xv 9.1 Introduction 79 9.2 Method 79 9.3 Assigned Emissions 80 9.4 Assessment of Method 85 10. Additional Assignment Method 87 11. Conclusions 89 Table 2-1 Summary description of the assignment methods 8 Table 2-2 Scale used to assess the cost criterion C1 9 Table 2-3 Scale used to assess the simplicity and transparency criterion C2 10 Table 2-4 Scale to assess data source reliability and data quality criterion C3 10 Table 2-5 Scale to assess the consistency and accuracy criterion C4 11 Table 2-6 Scale to assess the degree of influence on key variables criterion C5 11 Table 2-7 Scale to assess the fairness and appropriateness criterion C6 12 Table 2-8 Sulphur contents of fuels used for deriving emission factors. 14 Table 2-9 Comparison of annual emissions estimated for ports found in literature with the estimates Table 2-10 of this study 15 Assumptions for the duration (hours) of in-port activities, based on port surveys and database analysis. 16 Table 4-1 Percentage of ships tonnage under the different flags 37 Table 4-2 Number of ships under the flag of the 29 countries (2000) 39 Table 4-3 Distance table used in method B (km) 41 Table 5-1 Average emission factors for different fuel types used by ships in 2000 49 Table 6-1 Summary of literature data on worldwide fuel consumption by ships 57 Table 7-1 Total emissions calculated in Task A (200 mile zones) for the 29 countries 65 Table 11-1 Ranking based on the emissions allocated to each country under the different assignment methods i.e. 1 = most allocated emissions 94 Table 11-2 Summary table of assessments (0 is worst score, 5 is best score) 96 Table 11-3 Summary of main advantages, disadvantages and areas for further investigation of alternative assignment methods 97 Figure 1-1 Sea areas differentiated in the emission calculations and assignments 4 Figure 3-1 200 mile zones used in this study 20 Figure 3-2 A3: The countries share of reported total ship emissions for inland waterways in 2000 (TREMOVE, 2004) 23 Figure 3-3 A 3: Inland waterways ship emissions for NOx (TREMOVE, 2004) 24 Figure 3-4 A3: Inland waterways ship emissions for SO2 (TREMOVE, 2004) 24 Figure 3-5 A 3: Inland waterways ship emissions for VOC (TREMOVE, 2004) 25 Figure 3-6 A 3: Inland waterways ship emissions for CO 2 (TREMOVE, 2004) 25 Figure 3-7 A 3: Inland waterways ship emissions for PM (TREMOVE, 2004) 26 Figure 3-8 A: Each country s share of total ship emissions in the 29 countries 12 mile zones in 2000 27 Figure 3-9 A: Each country s ship emissions in 12 mile zones at sea as percentage of total RAINS emissions in 2000 (For Denmark the ratio of ship emissions to total emissions for SO 2 is 191%, thus not shown on the graph) 27 Figure 3-10 A: Assigned NOx ship emissions (at sea: 12 mile zones) 28 Figure 3-11 A: Assigned SO 2 ship emissions (at sea: 12 mile zones) 28 Figure 3-12 A: Assigned VOC ship emissions (at sea: 12 mile zones) 29 Figure 3-13 A: Assigned CO2 ship emissions (at sea: 12 mile zones) 29 Figure 3-14 A: Assigned PM ship emissions (at sea: 12 mile zones) 30 Figure 3-15 A: Each country s share of total ship emissions in the 29 countries 200 mile zones in 2000 31 Figure 3-16 A: Each country s ship emissions in 200 mile zones at sea as percentage of total RAINS emissions in 2000 (Note: a logarithmic scale was chosen to depict the wide range 1.5-425%) 31 Figure 3-17 A: Assigned NOx ship emissions (at sea: 200 mile zones) 32 Figure 3-18 A: Assigned SO 2 ship emissions (at sea: 200 mile zones) 32 Figure 3-19 A: Assigned VOC ship emissions (at sea: 200 mile zones) 33 Figure 3-20 A: Assigned CO 2 emissions based on 200 mile zones at sea 33
xvi Figure 3-21 A: Assigned PM ship emissions (at sea: 200 mile zones) 34 Figure 4-1 B: Each country s share of total ship emissions in the 29 countries in 2000 (based on ship flag) 42 Figure 4-2 B: Each country s ship emissions by flag as percentage of total RAINS emissions (Note: a logarithmic scale was chosen to depict the wide range 1.5-8,010%) 43 Figure 4-3 B: Assigned NOx ship emissions based on ship flag 44 Figure 4-4 B: Assigned SO 2 ship emissions based on ship flag 45 Figure 4-5 B: Assigned VOC ship emissions based on ship flag 45 Figure 4-6 B Assigned CO 2 ship emissions based on ship flag 46 Figure 4-7 B: Assigned PM ship emissions based on ship flag 46 Figure 5-1 C: Each country s share of total ship emissions in the 29 countries in 2000 (based on fuel Figure 5-2 sales estimates) 50 C: Each country s ship emissions on the basis of fuel sales, as percentage of total RAINS emissions in 2000 (For the Netherlands the ratio of ship emissions to total emissions for SO2 is 756%, thus not shown on the graph) 51 Figure 5-3 C: Assigned NOx ship emissions based on fuel sale estimates 51 Figure 5-4 C: Assigned SO2 ship emissions based on fuel sale estimates 52 Figure 5-5 C: Assigned VOC ship emissions based on fuel sale estimates 53 Figure 5-6 C: Assigned CO 2 ship emissions based on fuel sale estimates 54 Figure 5-7 C: Assigned PM ship emissions based on fuel sale estimates 54 Figure 6-1 Figure 6-2 D: Each country s share of total ship emissions in the 29 countries in 2000 (based on fuel consumption) 59 D: Each country s ship emissions based on fuel consumption, as percentage of total RAINS emissions in 2000 (For the Netherlands the ratio of ship emissions to total emissions for SO 2 is 755%, thus not shown on the graph) 59 Figure 6-3 D: Assigned NOx ship emissions based on fuel consumption 60 Figure 6-4 D: Assigned SO 2 ship emissions based on fuel consumption 60 Figure 6-5 D: Assigned VOC ship emissions based on fuel consumption 61 Figure 6-6 D: Assigned CO 2 ship emissions based on fuel consumption 61 Figure 6-7 D: Assigned PM ship emissions based on fuel consumption 62 Figure 7-1 Figure 7-2 E: Each country s share of total ship emissions in the 29 countries in 2000 (based on freight tonnes loaded) 66 E: Each country s ship emissions on basis of freight tonnes loaded, as percentage of total RAINS emissions in 2000 (Note: a logarithmic scale was chosen to depict the wide range 1.5-560%)) 67 Figure 7-3 E: Assigned NOx ship emissions based on freight tonnes loaded 67 Figure 7-4 E: Assigned SO 2 ship emissions based on freight tonnes loaded 69 Figure 7-5 E: Assigned VOC ship emissions based on freight tonnes loaded 69 Figure 7-6 E: Assigned CO 2 ship emissions based on freight tonnes loaded 70 Figure 7-7 E Assigned PM ship emissions based on freight tonnes loaded 70 Figure 8-1 F: Each country s share of total ship emissions in the 29 countries in 2000 (based on terrestrial emissions) 74 Figure 8-2 F: Each country s ship emissions allocated in proportion to RAINS emissions in 2000 (For Latvia the ratio of ship emissions to total emissions for SO2 is 175%, thus not shown on the graph) 74 Figure 8-3 F: Assigned NOx ship emissions based on terrestrial emissions 75 Figure 8-4 F: Assigned SO2 ship emissions based on terrestrial emissions 75 Figure 8-5 F: Assigned VOC ship emissions based on terrestrial emissions 76 Figure 8-6 F: Assigned CO2 emissions based on terrestrial emissions 76 Figure 9-1 Allocation cases 80 Figure 9-2 Figure 9-3 G: Each country s share of total ship emissions in the 29 countries in 2000 (based on departure/destination of ships) 80 G: Each country s ship emissions on basis of departure / destination of ships, as percentage of total RAINS emissions in 2000 (Note: a logarithmic scale was chosen to depict the wide range 1.5-500%) 81 Figure 9-4 G: assigned NOx ship emissions based on departure/destination of ships 82 Figure 9-5 G: assigned SO2 ship emissions based on departure/destination of ships 82 Figure 9-6 G: assigned VOC emissions based on departure/destination of ships 83 Figure 9-7 G: assigned CO 2 ship emissions based on departure/destination of ships 83 Figure 9-8 G: assigned PM ship emissions based on departure/destination of ships 84 Figure 11-1 Relative amount of ship emissions for European countries based on assignment by location (Method A), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) Figure 11-2 Relative amount of ship emissions for European countries based on assignment by flag 91 (Method B), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) Figure 11-3 Relative amount of ship emissions for European countries based on assignment by fuel 91 sales (Method C), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) 92
xvii Figure 11-4 Relative amount of ship emissions for European countries based on assignment by fuel consumption (Method D), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) 92 Figure 11-5 Relative amount of ship emissions for European countries based on assignment by freight tonnes (Method E), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) 93 Figure 11-6 Relative amount of ship emissions for European countries based on assignment by national emissions (Method F), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) 93 Figure 11-7 Relative amount of ship emissions for European countries based on assignment by country of departure / destination (Method G), using NOx emissions in 2000 for illustrative purposes (1.0 is the index given to the country with the largest share using this method) 94 Figure 11-8 Assessment of the individual criteria. 96 Figure 11-9 Illustration of mean scores of multi-criteria assessment (Applying equal weighting to each criteria, for illustrative purposes only) 97 Appendix A Appendix B Appendix C Appendix D Appendix E References Underlying Assumptions Data Used for Calculating Ship Emissions Calculated Emission Data Uncertainty Analysis
xviii
xix Glossary ACC AE EEA GRT GT HC HSD IWW kwh MCR MDO ME MGO MSD PM RO S Sfc SSD ST Accession Candidate Country (Bulgaria, Romania, Turkey, and Croatia for the purpose of this study) auxiliary engine European Economic Area country (Norway & Iceland for the purposes of this study) gross registered tonnage gas turbine hydrocarbons high speed diesel Inland Waterways kilo Watt hour maximum continuous rating marine diesel oil main engine marine gas oil medium speed diesel particulate matter residual oil sulphur specific fuel consumption slow speed diesel steam turbine
xx Country Codes AUT BEL DNK FIN FRA GER GRC IRL ITA LUX NLD PRT ESP SWE GBR CYP CZE EST HUN LVA LTU MLT POL SVK SVN BGR ROM TUR CRO Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden United Kingdom Cyprus Czech Republic Estonia Hungary Latvia Lithuania Malta Poland Slovakia Slovenia Bulgaria Romania Turkey Croatia
1 1. Introduction and Scope 1.1 This report This report presents the deliverables under Task 1 of the European Commission contract on Ship Emissions: Assignment, Abatement and Market-Based Instruments. The overall objective of this task is to make preliminary assignments of ship emissions to European countries. In particular, this task is intended to assign ship emissions of sulphur dioxide (SO 2 ), nitrogen oxides (NOx), volatile organic compounds (VOCs), particulate matter (PM) and carbon dioxide (CO 2 ) at a national level and to illustrate the significance of ship emissions, in particular in relation to land based emissions. Seven different methods were applied to assign preliminary ship emissions to EU Member States and candidate countries for the years 2000, 2010, 2015, and 2020. Each of the methods was appraised against several criteria, within an overall multi-criteria analysis. The findings of this task will provide supporting information to the Commission in considering how emissions from international maritime traffic could be included in the National Emission Ceilings Directive (Directive 2001/81/EC, NECD). 1.2 Scope of study 1.2.1 Assignment methods Based on the above objectives this task considers the following seven assignment methods: A. Assignment according to location of emissions (section 3). Ship emissions are estimated for ships within 12 mile and 200 mile zones of countries including ports and inland waterways. B. Assignment according to flag of ship (section 4). World wide ship emissions of ships travelling under a country s flag are estimated. C. Assignment according to industry fuel sales estimates (section 5). Based on a country s marine fuel sales estimates and generic emission factors potential ship emissions are estimated. D. Assignment according to reported fuel consumption (section 6). Based on a country s reported ship fuel consumption and generic emission factors potential ship emissions are estimated. E. Assignment according to freight tonnes loaded (section 7). Ship emissions are allocated in proportion to a country s reported freight tonnes. c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
2 F. Assignment in proportion to national emissions (section 8). Ship emissions are allocated in proportion to a country s national emissions. G. Assignment according to country of departure/destination (section 9). Ship emissions of each ship movement are allocated based on the country of departure and destination of that movement. 1.2.2 Pollutants For all above assignment methods the following emissions from ships are estimated and assigned: Sulphur Dioxide (SO 2 ) Nitrogen Oxides (NO X ) Volatile Organic Compounds (VOC) 2 Primary Particulate Matter (PM) Carbon Dioxide (CO 2 ) Where comparisons are made with total country emissions of NOx, SO 2, VOC and PM, the figures used for total emissions are based on the RAINS model 3 used for the Clean Air for Europe programme (climate policy scenario-revised version, activity path as in August 2004, emission vector as in November 2004). Where comparisons are made with total country emissions of CO 2, the figures used for total emissions are based on the Greenhouse Gas Inventory Data from the United Nations Framework Convention on Climate Change 4. CO 2 emission data are not available for Lithuania, Malta, Slovenia, Bulgaria, Romania, Turkey and Croatia. 1.2.3 Years of interest The emissions from ships are estimated for four years i.e. 2000, 2010, 2015, and 2020. For the three future scenarios (2010, 2015, 2020) a constant growth rate (+2.6%/year 5 ) of the travelled 2 These are exhaust emissions only, i.e. not including VOCs emitted during loading, unloading and gasfreeing of petro-chemical vessels. Loading and unloading emissions were quantified in a separate study for the EC available here: http://www.europa.eu.int/comm/environment/air/pdf/vocloading.pdf 3 It is noted that total emissions in RAINS include inland waterway emissions and emissions from international sea traffic (bunkers), national sea traffic within the EMEP area, and national fishing. International shipping is divided into five regions: Atlantic Ocean within the EMEP region, Baltic Sea, Mediterranean Sea and North Sea (including the English Channel). National sea traffic within the EMEP area and national fishing covers all ships operated between ports in the same country plus fishing vessels. The seagoing ships are divided in the RAINS database into two categories; medium (up to 1000 GRT) and Large (above 1000 GRT) vessels. PM emission data are not available for Bulgaria, Romania, Turkey and Croatia. 4 http://ghg.unfccc.int/default1.htf?time=02%3a49%3a16+pm 5 This growth rate is based on SCENES forecasts that correspond well with the long term growth of the world shipping performance of ton-miles (+2.5%/year) over the period 1970-99, UNCTAD (2000). The uncertainties of c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
3 distances on all routes is assumed for methods A, B and G. For the other methods this growth rate is directly applied to the derived emissions for the year 2000 to estimate future emissions. 1.2.4 Geographic areas For methods A, B and G the emissions are not just estimated on a country by country basis but also with a higher spatial resolution. The following nine sea areas as presented in Figure 1-1 are used for spatial resolution in addition to in port, 12 mile zones and 200 mile zones for each of the countries: 1. Baltic Sea, 2. Black Sea, 3. North Sea, 4. Irish Sea, 5. English Channel, 6. Mediterranean Sea, 7. North-East Atlantic Ocean (as defined in ENTEC 2002) 6, 8. Rest of EMEP area, 9. Rest of world outside EMEP area (for method B, C, D and G only). the future, the lack of data and models (economic and ship technology) do not justify the development of more sophisticated growth rates for individual vessel categories, for the purpose of this report. 6 The chosen spatial definition is to a certain extent arbitrary and comparisons with other calculations might be difficult for this specific sea area. c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
4 Figure 1-1 Sea areas differentiated in the emission calculations and assignments c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
5 1.2.5 Limitations of Scope The main objective was to estimate preliminary assignments of emissions for the 29 countries testing different potential assignment methods to support future decisions. Given this context and the limitation of resources, the estimated emissions are often based on generic assumptions and average parameter values rather than country specific inputs and should therefore be regarded as preliminary estimates to give a sense for the expected order of magnitude of emissions on a country basis under different methods, and the relative distribution of emissions between countries under different methods. It was not within the scope of work to develop new datasets, but to use and programme the emissions database developed under the Entec 2002 ship emissions quantification study for the Commission (Entec 2002). The underlying vessel movements data for that study was based on the most comprehensive databases available at the time 7 (the Lloyds Maritime Intelligence Unit (LMIU) ship movements database, in combination with the Lloyd s Maritime Information System (LMIS) vessel characteristic database), which includes movements of commercial ships > 500 gross tonnes (GT) (approx. 31,000 ships worldwide). As such, for the bottom up calculations in method A, B and G commercial ships > 500 GT are included in the estimates presented in this study. There are estimated to be approximately 47,000 ships in the range 100-500 GT (based on the LMIS vessel characteristic database) and additionally all smaller ships < 100 GT will not be accounted for in the bottom-up approaches used in this study. This range is likely to include a small ferries and fishing vessels. However the fuel consumption for the range 100-500 GT is estimated to be small i.e. <8% of total estimated consumption for >100GT (Endresen et al., 2003). Therefore it can be assumed that the total amounts of emissions for the range <500 GT are an order of magnitude smaller than the ones calculated for the range >500 GT. The emissions of smaller vessels are, however, more likely to be released closer to land. For the bottom-up approaches, emissions from main and auxiliary engines are taken into account but additional emissions from production processes on ships or operation of boilers are not included. For the overall fleet, such processes are not expected to contribute a significant proportion of emissions, although it has not been possible to quantify this within the resources of this study. 7 Also referred to as the Lloyd s Register Fairplay databases. c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
6 c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
7 2. Methods and Basic Assumptions In this section the different assignment methods are briefly introduced (section 2.1), the assessment approach is described (section 2.2) and the basic assumptions are briefly discussed (section 2.3). 2.1 Discussion of Different Assignment Methods The following seven assignment methods are applied in this study: A. Assignment According to Location of Emissions (section 3) B. Assignment According to Flag of Ship (section 4) C. Assignment According to Industry Fuel Sales Estimates (section 5) D. Assignment According to Reported Fuel Consumption (section 6) E. Assignment According to Freight Tonnes Loaded (section 7) F. Assignment in Proportion of National Emissions (section 8) G. Assignment According to Country of Departure/Destination (section 9) In addition, a further potential assignment method is briefly explored in section 10. Table 2-1 summarises the different assignment methods and their scope. Fundamentally, methods A, B, and G are bottom-up approaches and C, D, E, and F are topdown approaches. Methods E and F only provide the ratios to split total emissions among the different countries and rely therefore on emission inputs from another method. For each assignment method (i = A-G) there will be an underlying function (f i ) of variables in the form: Emissions allocated to country k ( kt / a) = f i ( Var1... Varn ) The methodical discussion of each assignment method will be based on the identified key variables that define the allocated emissions for a country. Obviously, the identified variables will often be functions of variables and parameters themselves. Based on the derived function and the estimated uncertainties of the input parameters the uncertainty of the preliminary emission results is estimated. This is summarised in the assessments of each method, with more detail on the quantified uncertainties of the methods given in Appendix E. c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
8 Table 2-1 Summary description of the assignment methods Parameter Assignment method A. Location B. Flag C. Fuel sales D. Fuel consumpt ion Pollutants to be included: SO 2, NOx, VOC, CO 2, PM E. Freight tonnes F. National emissions G. Departure /destinati on Years to be included: 2000, 2010, 2015, 2020 (Note 2) Countries to be included EU25+4 (Note 4) Link to results of other assignment method - - - - Method A (Note 1) Method A (Note 1) - Geographic region to be included English Channel Y Y Y Y Y Y Y North Sea Y Y Y Y Y Y Y Irish Sea Y Y Y Y Y Y Y Baltic Y Y Y Y Y Y Y NE Atlantic Y Y Y Y Y Y Y Mediterranean Y Y Y Y Y Y Y Black Sea Y Y Y Y Y Y Y Rest of EMEP Y Y Y Y Y Y Y Rest of World (outside EMEP) N Y Y Y N N Y Disaggregate by sea area? Y Y N N N N Y (Note 3) Disaggregate by vessel type? Y (Note 3) Y (Note 3) N N N N Y (Note 3) Disaggregate by movement type? (domestic/ intra EU/ international; inland/sea) Y Y N N N N N Water bodies to be included Inland waterways Y Y Y Y Y Y Y In port Y Y Y Y Y Y Y At sea Y Y Y Y Y Y Y Comparison with land based Y Y Y Y Y Y Y emissions for 2000 Notes 1. Method A has been selected as the basis for the purposes of this report, although alternative methods could be used as a basis. 2. The database model develops year 2000 emissions only. Projected emissions will be developed outside the database in a separate spreadsheet. 3. This disaggregation is undertaken in the detailed database but is not presented separately in the results. 4. Including Bulgaria, Romania, Turkey and Croatia c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
9 2.2 Assessment of the Different Methods The formal assessment consists of a multi-criteria analysis. Each assignment method is qualitatively assessed based on the following criteria that are discussed in more detail in the following subsections: Costs to calculate assigned emissions; Simplicity and transparency of assignment method; Data sources and quality; Consistency and accuracy; Degree of influence for countries on key variables; and Fairness and appropriateness. Each of these criteria is assessed and graded on a numerical scale 0 to 5 Costs to Calculate Emissions (C1) A comparison of the expected costs for the different assignment methods has been used to define the score for criterion C1. The assessment has been based on the project team s experience and time spent on assigning the emissions to the individual countries. Table 2-2 depicts the scale and relevant descriptions. Table 2-2 Scale used to assess the cost criterion C1 Description Scale C1 Very expensive 0 Expensive 1 Moderately expensive 2 Moderately inexpensive 3 Inexpensive 4 Very inexpensive 5 Simplicity and Transparency of Assignment Method (C2) The simplicity of an assignment method is given by the applied underlying model to calculate the emissions for a certain country and the availability of the input data to run the model. A high degree of transparency of the involved method/process is crucial to seek to gain the confidence of stakeholders. Table 2-3 depicts the scale used to assess the simplicity and transparency of the assignment methods. c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
10 Table 2-3 Scale used to assess the simplicity and transparency criterion C2 Description Scale C2 Most complex method, bottom-up method with high information needs on an individual ship basis, low transparency, almost no assumptions, everything based on primary data and measurements Bottom-up method, low transparency, a few weak assumptions, i.e. the result is not sensitive to these assumptions, the values that are used and generalisations Bottom-up method, transparent, a few strong assumptions, i.e. the result depends strongly on the assumptions and changes significantly for changing input values for the assumptions Combination of two approaches, top-down approach that relies on data gained from bottom-up studies, transparent, a few strong/weak assumptions 0 1 2 3 Top-down approach, transparent, a few strong assumptions, data easily available 4 Simplest method, top-down approach, highly transparent, input data easily available, closed formulas, i.e. numbers can be simply applied to gain a result y=a*b+c, many strong assumptions 5 Data Sources Reliability and Data Quality (C3) The reliability and quality of the underlying data used in an assignment method is crucial for the integrity of the outcome. The reliability covers the aspects of availability for all counties under consideration currently and in the future. The quality refers to the accuracy, level of aggregation involved and closeness to primary source of measurement. Table 2-4 Scale to assess data source reliability and data quality criterion C3 Description Scale C3 Highly aggregated data from different unidentified sources. One-off paper etc. 0 Aggregated data from many sources and no data statistics 1 Combination of different data sources, low degree of data aggregation, no statistics on data 2 Many respected data sources, but little information on data statistics 3 A few highly respected data sources from regular programs 4 Highly respected data source from regular program, data from direct measurements including data statistics 5 Consistency and Accuracy (C4) Consistency in the approach and a reasonable accuracy of the assigned amounts is important. Table 2-5 depicts the used scale and the associated descriptions. c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
11 Table 2-5 Scale to assess the consistency and accuracy criterion C4 Description Scale C4 Method with lots of strong assumptions and uncertainties and poor statistics on results 0 Method with a few weak assumptions and uncertainties and poor statistics on results 1 Assignment method relies on strong assumptions, expected statistics are not necessarily satisfactory 2 Assignment method relies on a few strong assumptions and statistics on results are satisfactory 3 Assignment method relies on a few weak assumptions and statistics on results are good 4 Assignment method has no strong assumptions and statistics on results are excellent 5 It should be noted that accuracy, as defined above, does not necessarily relate to fairness, which is defined separately below. Degree of Influence for Member States on Key Variables (C5) A method is more likely to be accepted if a clear cause-effect relationship underlies the method. Countries want to see how potential measures will change the estimated emissions. A method is therefore positively assessed the more the involved variables can be controlled on a country level. Table 2-6 depicts the scale used and the associated descriptions. Table 2-6 Scale to assess the degree of influence on key variables criterion C5 Description Scale C5 All variables used in the method are outside the influence of an individual country 0 A few variables can be influenced but the result is mainly defined by variables out of scope of an individual country 1 Variables that define a minor part of the result can be influenced by countries 2 A few key variables that define a major part of the result can be influenced by countries 3 All key variables can be controlled and influenced by an individual county 4 All variables can be directly controlled and influenced by an individual country 5 Fairness and Appropriateness (C6) Fairness and appropriateness are biased concepts and different stakeholders will have different conceptions. This is part of the natural situation due to incomplete knowledge, drivers, beliefs and doctrines and general shortcomings in thought and judgement. The authors of this study are in no way exempt from this natural situation. The fairness and appropriateness of a method is assessed based on the likelihood that someone without a country bias would agree to this assignment method. The fundamental question will therefore be: Does the assigned amount represent a perceived proper share in relative and absolute terms of the overall contribution to the ship emission problem? c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
12 Table 2-7 depicts the scale used and the associated descriptions. Table 2-7 Scale to assess the fairness and appropriateness criterion C6 Description Scale C6 The assigned amount bears no relation to the country s contribution to the ship emission problem and it is very unlikely that anyone unbiased would agree to this type of assignment method. The assigned amount has a weak relation to the country s contribution to the ship emission problem and takes only partly into account the distribution of the benefits from the ship movements. The assigned amount is either an acceptable proxy of the country s contribution to the problem or reflects well the distribution of the benefits from the ship movements. The assigned amount is an acceptable proxy of the country s contribution to the problem and takes into account the distribution of the benefits from the ship movements The assigned amount is a good proxy of the country s contribution to the problem that can be accepted by a majority and reflects the distribution of benefits from the ship movements The assigned amount represents the broadly accepted share of the problem and the distribution of the benefits. 0 1 2 3 4 5 Results of Multi Criteria Analysis Once each of the assignment methods is assessed against the above criteria the different methods can be compared. Additionally a weight can be given to each criterion to allow calculation of an aggregate score for an assignment method. However, the derivation of weighting factors for the criteria would need to take into account policy objectives and other factors which are beyond the scope of this study to assess. As such, it is not considered appropriate within the scope of this study to develop specific weighting factors and therefore aggregate scores for each method have not been derived. 2.3 Basic Assumptions As highlighted in Section 1.2.5, in relation to the specific objectives of this study, it was not within the scope of work to develop new datasets, but to use and programme the GIS emissions database developed under the Entec 2002 ship emissions quantification study for the Commission (Entec 2002). However, to account for the specific requirements of this overall project and to take advantage of more recent data, the main data set was updated and amended in the following areas. 1. Emission factors (section 2.3.1) 2. In-port activities (section 2.3.2) 2.3.1 Emission Factors The emission factors for the year 2000 from the Entec (2002) study have been reviewed and supplemented by the emission factors for particulate matter (PM) at sea. The emission factors c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
13 decided upon for 2000 have been extrapolated to estimate likely future emission factors in the years 2010, 2015 and 2020 under a business as usual scenario. Existing legislation which will impact upon future emissions include: the IMO Technical NOx code under Annex VI of MARPOL 73/78 caps on the sulphur content of fuels in Annex VI of MARPOL 73/78 and the Directive on the Sulphur Content of Marine Fuels Emission factors for NOx, SO 2 8 and PM 8 have been adjusted to account for the above legislation as discussed in the following sections. The other modelled emission factors are in general kept constant over the investigated time period 2000-2020. The detailed emission factors used in this study are summarised in Annex B in Table B.3 to Table B. 22. The presented emission factors are average emission factors for different ship types based on average ME and AE composition and expected load factors of these engines. Steam turbines are being gradually phased out and only a few turbine ships are in operation in the EU fleet. 9 Therefore emission factors were kept constant for turbines. NOx Code Legislation for marine engine emissions includes the IMO Technical NOx code under Annex VI of MARPOL 73/78. Diesel engines with a power output greater than 130 kw installed on a ship constructed after January 2000 must meet this code. The IMO NOx code is implemented under MARPOL 73/78 from 2005 and the legislation will be retrospective and interim certificates have been issued confirming conformity with the code. New engines will in general meet the NOx code requirements. To meet the requirements of the NOx code, existing engines might need a variety of measures depending upon the engine speed and age. Engine tuning can reduce NOx emissions by 6% (Cooper, 2004), and this will be enough for some engines to comply with the code. However, a significant proportion of existing engines will need additional measures to ensure compliance. Based on a review of NOx abatement technologies, an average NOx reduction for new engines of about 17% compared to average emission factors used in 2000 is assumed for future emission factors. A conservative implementation rate of abatement technologies of 4%/year is assumed, similar to the ship renewal rate of 4% stated in Cooper (2004). Sulphur Content and Affected Emission Factors The assumed sulphur contents of fuels for different locations and times are depicted in Table 2-8. 8 As the SO 2 and PM emission factors (eg corresponding to movements in SOx ECAs and movements of ferries) are only affected for specific vessels and in certain sea areas these changes in emission factors have to be modelled separately. The emission factors presented in Annex B do not account for these reductions and show the regular emission factors. The only exception is the SO 2 emission factors at berth from 2010 onwards. 9 There are 48 steam turbine ships and 7 gas turbine ships in the EU flagged fleet (< 1%). c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
14 Table 2-8 Sulphur contents of fuels used for deriving emission factors. Type of vessel / location Marine Distillates (MD) (Sulphur content % ) Residual Oil (RO) (Sulphur content % ) Year 2000 All ships in all sea areas 0.2% 2.7% Year 2010-2020 All ships in emission control areas (Baltic 0.2% 1.5% Sea, North Sea & Channel) 10 Passenger ships operating on regular services to or from any Community port at sea 10 0.2% 1.5% All ships at berth in Community ports 0.1% 0.1% and on inland waterways 11 All others 10 0.2% 2.7% SO 2 emission factors are directly related to the sulphur content of the fuel. Reducing the sulphur content of fuels may have an impact upon other emissions produced, such as NOx, VOC and particulates. For the switch from 2.7% sulphur to 1.5% sulphur only the expected reductions in SO 2 (-44%) and PM (-18%) emissions have been modelled as the impact upon the other emissions is deemed not to be significant. 2.3.2 In-Port Activities The previous estimates for in port times (Entec 2002) were derived from a questionnaire survey of ports, due to limited available information from other sources. This current study reconciled the port times as depicted in Table 2-10, querying the LMIU database (2000) and using additional literature data and port surveys 12. Based on this review, the manoeuvring times and the ratio between loading/unloading to hotelling were left unchanged. However the average port times for European ports for different vessel categories were increased by about 20%. The emissions in European ports are calculated based on these average port times. It has to be stated that there are large variations between individual ports but the objective of ship owners and port authorities is clearly to minimise port times 13. Therefore actual emissions for individual ports 10 Article 4b Paragraph 3 of Directive 2005/33/EC requires Member States to ensure that marine gas oils are not placed on the market in their territory if the sulphur content exceeds 0.1% by mass. However except for at berth and for inland waterways the use is not limited to 0.1% by mass. So it could be expected that due to availability the average of the sulphur content in MD will tend to decrease towards 0.1% close to EU territory however it is conservatively assumed to stay at 0.2% until 2020 in the current model. The impact of a change to 0.1% MD on the presented results would also be insignificant. 11 In the emission calculations that is incorporated by changing the emission factors for AEs. Emission factors for MEs are not changed as they are expected to run only a short time and will therefore in general not have to switch. 12 Interviews with ports of Rotterdam, Hamburg, and Piraeus 13 The latest Lloyds database allows a much higher time resolution of port times (15 minutes instead of 1day) and could be used in future studies to get more reliable information for individual ports. c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
15 might be quite different from the calculated emissions in this study but it has to be stated that the spatial resolution of this model is sea areas rather than individual ports. A comparison of calculated emissions with data found in literature is presented in Table 2-9 and shows, given the coarse approach based on estimated European averages, a reasonable correspondence. Table 2-9 Comparison of annual emissions estimated for ports found in literature with the estimates of this study NOx (t/a) Literature This study (2000) Copenhagen 2001 14 555 323 Koge 14 35 26 Hamburg 2000 15 2,400 4,300 16 Piraeus 17 ; 3,800 2,400 16 14 Saxe et al. (2003) 15 Umweltbehörde Hamburg (1997) 16 The average in-port durations used in this study are longer than durations reported for Hamburg and shorter than durations reported for Piraeus. 17 Kondopoulos et al. c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
16 Table 2-10 Assumptions for the duration (hours) of in-port activities, based on port surveys and database analysis. Vessel Description Vessel Category Manoeuvring Time Loading & Unloading Time Hotelling Time Total Time (hours) (hours) (hours) (hours) Liquefied Gas A11 1 19.5 19.5 40 Chemical A12 0.8 16.5 21.0 38 Oil A13 1.5 22.5 20.5 45 Other Liquids A14 1 20.0 22.0 43 Bulk Dry A21 1 58.5 43.0 103 Bulk Dry/Oil A22 1 71.0 5.5 77 Self-Discharging Bulk Dry A23 1 17.5 19.0 37 Other Bulk Dry A24 1 24.0 19.0 44 General Cargo A31 1 23.0 29.0 53 Passenger/General Cargo A32 0.8 9.6 4.0 14 Container A33 1 16.5 9.5 27 Refrigerated Cargo A34 1 32.0 22.0 55 Ro-Ro Cargo A35 1 11.0 13.5 26 Passenger/Ro-Ro Cargo A36 1 6.0 4.0 11 Passenger A37 0.8 15.5 6.5 23 Other Dry Cargo A38 1.1 47.0 5.0 53 Fish Catching B11 0.7 21.0 83.5 105 Other Fishing B12 0.7 16.5 66.5 84 Offshore Supply B21 2 25.0 27.5 55 Other Offshore B22 0.5 33.5 36.5 71 Research B31 0.9 0.0 82.0 83 Towing/Pushing B32 1.7 57.0 18.0 77 Dredging B33 3 24.5 5.5 33 Other Activities B34 1.1 28.0 29.5 59 Other Activities W11 0.5 0.6 26.3 27 Other Activities W12 0.5 0.6 26.3 27 Other Activities W13 0.5 0.6 26.3 27 c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
17 2.4 Role of New Technologies The future availability of continuous location and emission monitoring due to on-board global positioning system (GPS) technology, continuous emission monitoring systems (CEMS) and the development of universal ship-borne Automatic Identification System (AIS) transponders will further improve the accuracy of the emissions monitoring and positioning. All assignment methods that are based on direct ship emissions/movement related input data, e.g. A, B etc., would potentially become more accurate and gain legitimacy from these new technologies. These technologies will also support the introduction of economic instruments as discussed under the separate report for Task 3 of this contract, which rely on accurate monitoring and locating of the emissions. GPS systems are standard on the commercial fleet and in combination with a data-logger would allow historic time-location reconstructions of ship movements with a high spatial resolution. A model would therefore not rely on a generic route modelling between two ports anymore but on actual routes and could model time series rather than yearly totals. This would improve the assignment of emissions to actual locations, and also allow the time to be taken into account, which may be a factor in assessing pollution damage caused by certain types of emission. AIS transponders will allow shore-based systems to inexpensively identify and track AISequipped vessels within VHF radio distance i.e. about 40 nautical miles. The foreseen AIS long-range reporting mode would assure that up to 200 nautical miles could be covered. This allows an independent tracking and verification of ship movements as long as the system is switched on. The Safety of Life at Sea Conventions (SOLAS) Chapter V, Regulation 19 states: All ships of 300 gross tonnage and upwards engaged on international voyages and cargo ships of 500 gross tonnage and upwards not engaged on international voyages and passenger ships irrespective of size shall be fitted with an automatic identification system (AIS),. According to the convention this should happen by July 2008. This would cover all ships included in the emissions quantifications in this study. CEMS could provide highly accurate data on emissions from individual ships and is fundamentally feasible, although investment and operating costs can be high. The role of CEMS for ship emissions measurement is discussed in more detail in the Task 2 General Report, as part of this overall study. c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
18 c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
19 3. Method A - Assignment According to Location of Emissions 3.1 Introduction This assignment method uses reported individual ship movements between ports to estimate the emissions in ports and along the routes travelled. Task 1A is split into the following three subtasks: 1. Ship emissions at sea (section 3.2.1) 2. Ship emissions in ports (section 3.2.2) 3. Ship emissions from inland waterways (section 3.2.3) The emissions from sea, ports and waterways can then be summed up to give each country s total ship emissions. In the following sections for each sub-task the approaches and results are defined. 3.2 Method 3.2.1 Ship Emissions at Sea Two different areas have been considered when calculating ship emissions at sea. The 12 mile zone area that equates to territorial waters and the 200 mile zone area modelled after the exclusive economic zones 18. Note that where there is less than 400 or 24 miles between coastal states, the zone is not 200 or 12 miles wide; rather it is at the mid-point between the two coastal states. For the purpose of this study 200 mile zones as shown in Figure 3-1 for all countries with shorelines were established despite the fact that many countries have not claimed their EEZs (that is to say Finland, Greece, Ireland, Italy, Malta and Slovenia 19 ). The presented emissions for the 200 mile zone include the emissions from the 12 mile zone in order to give a 18 The United Nations Convention on the Law of the Sea (UNCLOS) is an international agreement that sets conditions and limits on the use and exploitation of the oceans. This Convention also sets the rules for the maritime jurisdictional boundaries of the different member states. Under UNCLOS, coastal States can claim sovereign rights in a 200-nautical mile exclusive economic zone (EEZ). The EEZ starts at the outer boundary of the Territorial Sea i.e., 12 nautical miles from the low-water line along the coast. It has to be noted that many countries have not yet claimed their EEZ. Additional information can be found at http://www.un.org/depts/los/index.htm. 19 http://www.un.org/depts/los/legislationandtreaties/claims.htm c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005
20 comprehensive estimate. To derive emissions for the actual EEZs of a country i.e. the zone 12-200miles the emissions of the 12 mile zone would have to be deducted from the stated emissions of the 200 mile zone as the EEZs do not include the 12 mile zone. Figure 3-1 200 mile zones used in this study For each ship movement in the database the emissions in a certain zone are calculated as follows: Formula 1 D (km) E A : At sea g) = [ME(kW) LFME (%) + AE(kW) LFAE (%)] EF v (km/h) With: ( at sea (g/kwh) D: Distance a ship travels in a certain zone of a country, e.g. 12 mile zone of country X, estimated based on port of departure, port of arrival, and assumed route. v: Average speed of a ship depending on ship category (see Annex B Table B.1). ME: Installed main engine power. This information is gained from the Lloyd s Maritime Information System (LMIS) database. LF ME : Average load factor of main engine at sea (see Annex B Table B. 2). c:\documents and settings\woodt\desktop\task 1 final report 05241_v1.doc August 2005