CBA of S PDF

Cost Benefit Analysis to Support the Impact Assessment
accompanying the revision of Directive 1999/32/EC on the

Sulphur Content of certain Liquid Fuels

Final Report

Report to European Commission

ED45756
Issue 3
23
rd
December 2009
Cost benefit analysis to support the impact assessment AEA/ED45756/Issue 3
sulphur content of certain liquid fuels

ii

Title Cost Benefit Analysis to support the impact assessment accompanying the
revision of Directive 1999/32/EC on the sulphur content of certain liquid
fuels

Customer European Commission

Customer reference ENV.C.5/FRA/2006/0071 Service Request No. ENV C5/GK/em/Ares (08)
36661

Confidentiality,
copyright and
reproduction
This Report has been prepared in response to Service Request ENV
C5/GK/em/Ares (08) 36661 Under framework contract No
ENV.C.5/FRA/2006/0071 between the Commission and AEA Technology.

File reference AEA/ED45756/Issue 3

AEA
Gemini Building
Harwell IBC
Didcot OX110QR
UK

Telephone: +44 (0) 870 190 6604

AEA Technology plc is certificated to ISO9001 and ISO14001

Authors Name Peter Bosch, Peter Coenen (TNO), Erik Fridell, Stefan
strm (IVL), Tom Palmer (AEA Technology), Mike
Holland (EMRC)

Approved by Name Mike Woodfield (AEA)

Signature

Date 23
rd
December 2009

iii


iv

Table of contents

1 Summary 1
2 Introduction 5
3 Methodology 7
3.1 Overview of methods 7
4 Scenario Description 8
4.1 Description of the Baseline Scenario 8
4.2 Description of the Policy Scenarios 10
5 Dispersion results; concentrations and exceedance of critical loads 14
5.1 Introduction 14
5.2 SEA1 15
5.3 SEA2 (applies only to 2015) 16
5.4 SEA3 17
5.5 SEA4 (2020 only) 17
5.6 SEA5 (2020 only) 18
5.7 Land based A 18
5.8 Land based B 18
5.9 ASEA1-BSEA1-ASEA4-BSEA4 18
6 Benefits assessment 19
6.1 Background 19
6.2 Methodology 19
6.3 Results 22
7 Cost analysis of the new IMO marine fuel and engine standards and a
potential extension of emission control areas 34
7.1 Cost calculation method description 34
7.2 Abatement Technology Description 35
7.3 Unit abatement costs 37
7.4 Scenario abatement costs 42
7.5 Results 43
8 Comparing Costs and Benefits 52
8.1 Preliminary comparison of costs and benefits 52
8.2 Detailed uncertainty assessment 53

v

8.3 Summary of the CBA results 58
9 Emission abatement equipment 59
9.1 Summary 59
9.2 Introduction 60
9.3 Abatement of sulphur oxides 61
9.4 A comparison between reducing SO
2
emissions with low sulphur fuel or scrubbers
62
9.5 Emission criteria for EGC 63
9.6 Wash-water criteria 63
9.7 Environmental assessments 65
9.8 Environmental impact related to the IMO wash water criteria 66
9.9 Monitoring 69
9.10 Conclusions and suggestions 70
10 Use of distillate fuels by ships at berth 73
10.1 Introduction 73
10.2 Outline of the problem 73
10.3 Types of ships where the risks may be relevant 74
10.4 Relevant number of ships 74
10.5 The nature of the risks 75
10.6 Technical solutions and costs 77
10.7 Conclusions 78
11 Requirements with regard to maximum sulphur content of heating
oil for condensation boilers 79
11.1 Objective 79
11.2 Introduction to oil fired condensation boilers 79
11.3 Heating oil quality and availability 79
11.4 Construction of condensation boilers 80
11.5 Condensation boilers and sulphur content of heating oil 80
11.6 Conclusions 81
12 References: 82
13 Appendix 1 - Emission factors used 86
14 Appendix 2 - Emission Figures by Scenario and Area Baseline 90
15 Appendix 3 - Relevant geographical representations of the dispersion
calculations 93
16 Appendix 4 - Results per EU member states 125

vi

17 Appendix 5 - Health Benefit Diagrams 153
18 Appendix 6 - Methodology for cost calculations 160

AEA 1
1 Summary
This report documents the Cost Benefit Analysis to Support the Impact Assessment accompanying the
revision of Directive 1999/32/EC on the Sulphur Content of certain Liquid Fuels.

Maritime activities constitute a significant fraction of anthropogenic emissions of air pollutants in Europe.
International shipping in the EU is expected to increase in the coming decade and there is a risk that
this will offset the effects of the current emission reduction legislation for maritime activities and the
future emissions will surpass the target levels established by the European Commission in its Thematic
Strategy on Air Pollution for land based sources.

A number of ongoing developments will lead to reductions in emissions of sulphur oxides (SO
x
) and
nitrogen oxides (NO
x
). In this report the effect of these developments is determined in order to inform the
Commissions review of Directive 1999/32/EC relating to a reduction in the sulphur content of certain liquid
fuels.

This study has examined the effects of emission reduction measures (for the years 2015 and 2020) in
maritime activities by:
1. Analysing the benefits of: the new IMO marine fuel and engine standards (SO
2
and NO
x

reduction respectively), and

the potential extensions of emission control areas. The benefits of
different scenarios have been evaluated against a revised NECD 2008 baseline scenario. The
study did not take into account the effects of the current economic crisis (assuming this will only
be temporary). The benefits were quantified with respect to a baseline scenario for emissions,
air quality (concentrations) as well as their impact on human health.
2. Calculating the costs of emission reductions that result from the new IMO marine fuel and
engine standards and the potential extension of emission control areas.
3. Assessing the effects, costs, risks and technical requirements that are associated with the use
of emission abatement equipment and steam boilers by ships as well as the use of
condensation boilers for heating purposes.

Benefit analysis
The following scenarios were investigated and were compared to a Baseline scenario (revised NECD
2008 baseline):
Designation of the Baltic Sea, North Sea and the English Channel SECAs (2015
1
and 2020
2

3
)
Inclusion of Mediterranean as a SECA (2015)
Inclusion of the Black Sea as a SECA (2015 and 2020)
Reduction of land based sources
4
by reduction of the S-content in gasoil to 0.05 % (2015 and
2020)
Reduction of land based sources by reduction of the S-content in gasoil to 0.001 % (2015 and
2020)

For each scenario the emissions of relevant air pollutants were calculated, and the effects on
ambient air quality were quantified using state of the art modelling tools. These concentrations were
used to determine the health benefits of the emission reduction using the methods developed in the
CAFE and EC4MACS studies for EC DG-ENV.

Cost analysis
The cost calculations for NO
X
and SO
2
abatement from shipping in this report contain two parts. The
first consists of calculating unit costs (per PJ) for each abatement option under consideration, the
second the calculation of scenario specific costs based on the penetration of the low-NO
X
and low-SO
2

1 In all 2015 marine scenarios Tier I & Tier II for NOx reduction is assumed
2 In all 2020 marine scenarios Tier I - III for NOx reduction is assumed simulating the designation of the Baltic Sea ,
North Sea , the English Channel and the Mediterranean as NECA
3 For 2020 also scenarios are elaborated with a reduced S content (0,5 %) in marine fuels used outside the SECA
4 Only for heating in households and agriculture

AEA 2
technology options. The calculations allow for scenario comparison and comparison of alternative
technologies to reach the required standards.

For each scenario, costs are calculated, specifying high and low costs for NO
X
and SO
2
abatement. For
SO
2
abatement, separate cost calculations are presented for scrubber costs and low S-fuel costs

Comparison of costs and benefits
Annual costs and benefits (low and high estimates) for the different scenarios are given in the following
tables - taking the results directly from the analysis. The low bound for costs is the lower estimate for
fitting a scrubber to reduce SO
2
emissions together with the costs of NO
x
control. The upper bound for
costs is for the high estimate for using low sulphur fuels to reduce SO
2
emissions and the costs of NO
x
control. The range for benefits is based on use of either the value of a life year (VOLY, giving the lower
bound) or the value of statistical life (VSL, giving the upper bound) to value mortality.

In all cases there is no overlap between the estimates for costs and benefits as shown above, indicating
that for all scenarios benefits would exceed costs..

Further analysis was undertaken to assess the effect of further uncertainties on the balance of costs
and benefits. This found that the probability of benefit exceeding cost can be described as very likely /
virtually certain, using terminology developed by the Intergovernmental Panel on Climate Change
(IPCC). With one slight exception (the last 2020 scenario: All European seas Included in the SECA and
reduced S-content outside the SECA ) the probability of benefits exceeding costs can be described as
likely, very likely or virtually certain even for worst case assumptions on cost.

It is to be remembered that the probabilities calculated here are quantified only against benefits for
improved health. The inclusion of impacts to materials (including cultural heritage) and ecosystems, in
particular, would add to the benefits and further increase the already high probabilities. Taking these
issues together it is concluded that more detailed assessment of the limitations of the analysis finds that
the initial conclusion, that benefits of the proposed actions will exceed costs, is indeed robust.
2015
(billions)

Baltic Sea , North
Sea and the
English Channel
as SECA
Inclusion of
Mediterranean in
the SECA
Inclusion of the
Black Sea in the
SECA
Costs low 0.6 1.8 1.9
high 3.7 10 10

Benefits low 8 11 11
high 16 24 25
2020

(billions)

Baltic Sea , North
Sea and the
English Channel
as SECA
Inclusion of
Mediterranean and
the Black Sea in
the SECA
Baltic Sea ,
North Sea and
the English
Channel as
SECA

Reduced S-
content outside
the SECA
Inclusion of
Mediterranean
and the Black
Sea in the SECA

Reduced S-
content outside
the SECA
Costs low 0.9 2.0 2.6 2.6
high 4.6 12 11 14

Benefits low 10 14 15 16
high 23 32 34 36

AEA 3
Comparison of the benefits of the land based scenarios against the maritime scenario (Baltic Sea ,
North Sea and the English Channel as SECAs) shows that the additional health benefit from lowering
the S content in fuel on land amounts to 0.4 to 0.8 billions (depending on the S content) in 2015. For
2020 the additional benefits are calculated at 0.6 billions. These figures are only small compared to the
benefits of the maritime scenario (which amount to 17-35 billions) and therefore do not constitute an
alternative to the maritime emission reduction.

Technical criteria for exhaust abatement equipment
Exhaust abatement equipment for the reduction of emissions of sulphur dioxide from ships is an
alternative to reduced S content in marine fuels. The only technology available at present is scrubbing.
Scrubber units can be of two kinds, seawater scrubbers and closed scrubbers, or a combination of two.
The use of scrubbers to clean the exhaust from marine engines using residual oil could in principle lead
to high concentrations of a number of harmful compounds in the marine environment around the ships.
The present situation is somewhat incongruous in the sense that washwater limits for scrubbers are set
by IMO while there still are no limits for the gaseous exhaust (except for SO
2
and NO
X
). It is
recommended that criteria are developed that promote the abatement of PM. At least the levels of PM
reduction should be established to allow for voluntary or market based reductions of PM emissions.
Further, it is recommended, when possible, to keep the EU regulations in line with the IMO guidelines.

The present IMO washwater criteria contain limits for PAH, pH, turbidity and nitrates. Of these all except
the nitrate criterion are intended for continuous monitoring onboard the ship. A major factor when
formulating the criteria is that they should be possible to monitor online. However, it would also be
possible to add more criteria that are to be tested during certification. This could give further assurance
that the units are environmentally safe and would not be constrained by the possibility to measure
online in a simple way.

With the present criteria for pH it seems as if dilution and buffering in the sea will rapidly give acceptable
levels. Even in confined areas it is likely that the effect is limited. However, it is recommended that
further studies are undertaken for brackish and river water and that the criteria then are reviewed. It is
suggested that the IMO guidelines are adopted for the pH criteria but at the same time allowing for
revisions as more data becomes available.

The criteria for PAH are expressed as phenanthrene equivalents. This is not a well established concept
and should be defined in the guidelines. It may be appropriate to use phenanthrene for monitoring
purposes but the criteria ought to guarantee low emissions of a range of PAHs and thus it is necessary
to establish how the measured value relates to e.g. PAH 16 or THC.

The criteria for turbidity may be used to establish that unwanted solid material is not formed in the
scrubber unit. However this criterion is not appropriate to monitor PM reduction because there is no
relation between particle concentration and turbidity.
It is recommended that the concentrations of particulate matter, metals, and PAH are regulated and
tested during commission, and, further, that the correlation to turbidity is investigated in order to have
continuous measurement available onboard.

The present criteria for nitrates could in principle lead to large additional contributions of nitrates to
sensitive areas. The best way to deal with this issue is to use abatement technology for NO
X
reduction.
It is recommended to set up new criteria regarding nitrate emissions which prevent increased nitrogen
loads to sensitive areas.

It seems appropriate to adopt the IMO guidelines regarding monitoring, i.e., allowing for the possibility of
not using continuous monitoring of the SO
2
emissions. The reason is that there will be units that can be
sufficiently monitored by other means.

There is only limited experience from trials with marine scrubbers for SO
2
. It is important to follow the
trials taking place and make sure that thorough environmental impact assessments are conducted in
order to evaluate and develop the washwater criteria.


AEA 4
Use of distillate fuels by ships at berth
From 1 January 2010 ships using ports within the EU will be required to use low-sulphur fuel, maximum
0.1 wt% sulphur, while at berth, with an exception for those making only short stops (Directive
2005/33/EC). This will most likely make it necessary to use distillate fuel while at berth. There are
concerns that there are risks associated with switching from residual oil to distillate fuel in boilers.

One can state that there are real safety concerns with switching from HFO to MGO in existing boilers.
The main safety concern is that a distillate fuel may vaporise in hot boilers with heated pipes. If the
boiler system has not been modified, there isin the case of flame failure and improper handling an
explosion risk. Although many boilers can be used with MGO, all boilers should be inspected before
using MGO at berth.

There are technical solutions to solve these issues where fuel piping systems, burners, and flame
inspection systems are modified. The control system may have to be modified and new handling
routines established. For LNG ships, modifications are required. The costs range from 5 000 to 25 000
for small auxiliary boilers, around 150 000 per ship for tankers and 70 000 to 1 400 000 for LNG
tankers.

Requirements with regard to maximum sulphur content of heating oil for condensation boilers
A typical issue concerning condensation boilers used for space heating is use of the latent heat of the
water vapour in the exhaust gases. This increases the fuel efficiency and therefore decreases the CO
2

emission by 6% compared to conventional boilers. Because the flue gases are cooled down below the
dew point, condensation will occur on the heat exchanger in the boiler and using heating oil will cause
an acidic (and thus corrosive) condensate. It is suggested that this phenomena sets limits for the
maximum sulphur content of heating oil to be used in condensation boilers. In this study its is
established that heating oils with sulphur content up to the maximum allowed concentration of 1000
mg/kg can be used in condensation boilers. However, to maintain high efficiency, heat exchangers
have to be cleaned more frequently than in conventional boilers. Using low sulphur oil will prevent this.
Furthermore are there no technical obstacles which prevent the use of low sulphur heating oil in
condensation boilers designed for standard heating oil.


AEA 5
2 Introduction
Maritime activities constitute a significant fraction of anthropogenic emissions of air pollutants in Europe.
In 2000, SO
2
and NO
x
emissions from international maritime shipping in Europe amounted to
approximately 30 percent of the land based emissions in the EU-25. While legislation is in force to
control emissions from international shipping, the expected increase in the volume of ship movements
will offset the positive environmental impacts of these measures and will lead to further growth in ship
emissions. Under business-as-usual assumptions, by 2020, emissions from maritime activities would
come close to the projected baseline emission levels from land based sources and surpass the target
levels established by the European Commission in its Thematic Strategy on Air Pollution for land based
sources.

This anticipated increase in ship emissions will counteract the envisaged benefits of the costly efforts to
control the remaining emissions from land based sources in Europe. While at present emissions from
ships are responsible for 10 to 20 percent of sulphur deposition in coastal areas, their contribution is
expected, by 2020, to have increased to more than 30 percent in large areas in Europe, and up to 50
percent in coastal areas.

Analysis by IIASA in 2007 demonstrated that limiting air pollution from shipping reduces the necessity to
further control emissions from land based sources and provides important cost savings in achieving air
quality targets in Europe.

A number of ongoing developments will lead to reductions in emissions of sulphur oxides (SO
x
) and
nitrogen oxides (NO
x
), these include:
Annex VI of MARPOL and the Technical Code on Control of Emissions of Nitrogen Oxides from
Marine Diesel Engines (the new IMO marine fuel and engine standards) from the International
Maritime Organisation (IMO) have been revised and formally adopted by the Marine Environmental
Protection Committee (MEPC) of IMO in October 2008. These revisions set criteria for a gradual
reduction in the maximum permitted sulphur content of marine fuels as well as the maximum
permitted emissions of nitrogen oxides from marine diesel engines;
Annex VI of the Marine Pollution Convention (MARPOL) defines Sulphur Emission Control Areas
(SECAs) where stricter limits on sulphur emissions apply. As of now, two areas are classified as
SECAs: the Baltic Sea and the North Sea/English Channel. Article 7 of Directive 1999/32/EC
requires the Commission to give consideration to proposals for the addition of SECAs.
The possibility that European institutions will designate emission control areas for NOx;
The agreement by the European institutions on the Climate and Energy Package. (26 March
2009)
The main purpose of this report is to determine the effects of these developments in order to inform the
Commissions review of Directive 1999/32/EC relating to a reduction in the sulphur content of certain liquid
fuels.

This study has examined the effects of emission reduction measures in maritime activities by:
Analysing the benefits of the new IMO marine fuel and engine standards as well as the potential
extensions of the emission control areas. The benefits of different scenarios have been
evaluated against a revised NECD 2008 baseline scenario. All existing and foreseeable
Community and national measures were taken into account, as far as these have a direct
impact on emissions and impacts on future activity are known. The study did not take into
account the effects of the current economic crisis. It has been assumed that the crisis is only
temporary (noting apparent recovery of several major Member State economies in the second
half of 2009) and that the economy will subsequently evolve as was expected before the crisis.
The benefits were quantified with respect to the baseline scenario for emissions, air quality
(concentrations) as well as the impact on human health.
Calculating the costs of emission reductions that result from the new IMO marine fuel and
engine standards and the potential extension of emission control areas.

AEA 6
Assessing the effects, costs, risks and technical requirements that are associated with the use
of emission abatement equipment and steam boilers by ships as well as the use of
condensation boilers for heating purposes.


AEA 7
3 Methodology
3.1 Overview of methods
In general terms the cost benefits analysis consists of the different steps illustrated in figure 3.1.

Figure 3.1 Flow scheme for the cost benefit analysis

EMISSIONS
tonnes/year
DISPERSION
Increase in ambient concentrations
in all affected regions
IMPACT
Using exposure-response curves,
e.g. change in building erosion
BENEFITS
Heath benefits, Damage costs
Stock at risk
e.g. inventory building materials,
population distribution
EXPOSURE
People, crops, materials
COSTS FOR
EMISSION REDUCTION
Base line &Policy scenarios
- Prices different fuel grades
- Abatement costs
COMPARISON of COSTS & BENEFITS
EMISSIONS
tonnes/year
EMISSIONS
tonnes/year
DISPERSION
DISPERSION
IMPACT
IMPACT
BENEFITS
BENEFITS
Stock at risk
e.g. inventory building materials,
population distribution
EXPOSURE
EXPOSURE
COSTS FOR
EMISSION REDUCTION
Base line &Policy scenarios
- Prices different fuel grades
- Abatement costs
COMPARISON of COSTS & BENEFITS

In the first step the baseline scenario is developed. All other policy scenarios will be compared to this
baseline. The emissions of SO
2
, NO
x
, PM, CO and NMVOC for these scenarios are calculated and fed
into the dispersion model. The scenario descriptions and emission estimates are given in section 4. The
most relevant parameters for the dispersion modelling are given in section 5. The assessment of the
benefits is elaborated in section 6 and section 7 describes the cost calculations for the emission
reduction. In section 8 a comparison is made between costs and benefits based on the results of the
two previous sections.


AEA 8
4 Scenario Description
4.1 Description of the Baseline Scenario
The baseline for the study is traceable to the NEC6/C&E package/current policies baseline scenario
from IIASA/GAINS, which has been used as baseline for simulations of the NECD plus the proposals for
the Climate and Energy Package of the Commission of 23 January 2008. It includes the most up to
date (Spring 2008) current control policies as laid down in legislation and those presently in the decision
phase of European Institutions. It includes: 'Current legislation' in all sectors plus: revision of the IPPC
Directive for large stationary combustion sources in industry and power sector (less strict BAT case
interpretation); Euro VI on heavy-duty vehicles and Ceilings from the NEC Directive. Assumptions are
documented in detail in the NEC6 report, page 18 and 19 (Amman et al, 2008).

As the NEC6 scenarios do not fully cover international shipping (they contain national maritime fuel
consumption but not international bunkering) a separate baseline has been developed for international
shipping. The starting point for the fuel consumption of international shipping is the 2000 fuel
consumption - from Cofala et al. 2007 (page 11). The fuel consumption is given by fuel type (MD/HF),
vessel type (cargo/ferry) and regional sea. It is based on the 2000 inventory of emissions produced by
ENTEC, which is in itself based on a detailed analysis of four months of vessel movement data from the
year 2000. Cofala applies an overall 2.5 per cent growth rate for cargo vessels and 3.9 per cent rate for
passenger vessels in the baseline whereas, for this study, we adopt the baseline of the Ex-tremis model
(Chiffi et al., 2007), which is based on a detailed model of shipping in EU27 countries and is
differentiated for marine diesel and heavy fuel oil. The growth factors for bunker fuel consumption from
Ex-tremis (for MD and HF) for EU ship movements are used to project the regionally detailed fuel
consumption from international shipping in European seas in 2010, 2015 and 2020. Over the period up
to 2020 the overall growth in HF use is 2.7% per year, while use of MD decreases with 0.5% per year

We have also used the outcomes of the Ex-tremis baseline to aproximate the fuel consumption by ship
types and engine size. Ex-tremis only covers EU27 countries and misses the traffic over European
seas generated by non-EU countries in Europe (Chiffi et al., 2007). Assuming, however, that the
developments in transport and developments in the fleet in EU and outside EU will run in parallel, it is
possible to apply the distribution of ship types (bulk carrier, tanker, containership, general cargo, etc.),
and engine class (gas turbine, high speed diesel, medium speed diesel, slow speed diesel, steam
turbine) in each of the years to the regional fuel consumption for international shipping in 2010, 2015
and 2020.

Emission factors have been added by detailed ship-classification and by regional sea. The baseline is
a reference scenario and does not include any policies - it is based on fixed emissions factors and, for
instance, does not include the 1997 IMO protocol, introducing Tier I for ships built after 2000 and which
took effect in 2005. Neither does it include the Tier II regulations that apply for ships built after 2011. As
a result the emissions reductions in scenario SEA1 compared to the baseline suggest a relatively larger
shift than would occur in reality. An overview of emission factors used is provided in Annex 1. It should
be noted that the sulphur content of HF in the baseline outside SECAs is 2.70% (according to TFEIP,
2009), which is the current worldwide average. Inside SECAs it is 1.45% representing the present
regulation of 1.5%. This is used to be able to illustrate the impact of the agreed limit of 0.1% from 2015
that is used in one of the scenarios (SEA1).

The emission factors for particulate matter have been adapted from Cooper and Gustafson (2004) and
have been chosen according to the sulphur content of the fuel and the speed of the engines. Table 3 in
Annex 1 specifies the emission factors used in the emission calculations e.g. for slow speed diesel
engines running on heavy fuel oil the reduction in the PM emission factor resulting from a decrease in
sulphur content in HFO from 1.5% to 0.1% is 54%.

As it is difficult to model the introduction of scrubbers, we have not introduced a detailed calculation
differentiating the part of the sulphur limit that can be reached by using scrubbers and the part that can
be reached by using low sulphur fuels. For the calculation of the sulphur emissions the way emissions
reductions are realised does not make a difference. It does, however, create uncertainty in the

AEA 9
calculation of PM emissions. Although there are few measurements to substantiate this, scrubbers
reduce PM emissions roughly by half (see also Section 9.4). This reduction is in the same order of
magnitude as the reduction in the emission factors for PM, when going from 1.5 to 0.1% S in HFO for
the majority of the vessel; consequently the error introduced by not accounting for the share of
scrubbers in the PM calculation is limited.

The resulting baseline emissions per sea area are given in Table 4.1a.

Table 4.1a Baseline emissions international shipping this report, 2000 and 2020
2000 2020
Sea area SO
2
NO
x

NMV
OC
PMtotal SO
2
NO
x

NMVO
C
PMtotal
x 1000 tonnes
North Sea 417 664 23 22 399 1171 41 39
Black Sea 53 83 3 6 93 147 5 11
Mediterranean 1024 1593 56 118 1795 2805 98 210
Baltic Sea 177 283 10 9 169 498 17 17
NE Atlantic 472 742 26 56 829 1310 46 100
Total 2144 3365 118 212 3758 5931 207 377

The 2000 inventory of emissions produced by ENTEC was used to distribute the emissions per sea
area to generate the gridded data needed for the dispersion modelling. The national data from the
IIASA-GAINS baseline are also gridded according to sector specific distribution patterns.

The land based emissions are the same as in the C&E package current policy scenario in Amman et al.
(2008). The baseline sea-based emissions are more or less comparable to those in Cofala et al. (2007).
Differences, however, are caused by the split in ship types we have introduced by using Ex-tremis,
which brings about a split in HF and MD that differs from the split used by Cofala et al. (2007). Further
differences are brought about by the larger differentiation in emission factors for NO
x
, NMVOC, CO and
PM which we have applied for different engine types. Cofala et al. (2007) assume lower sulfur and PM
emissions in 2020 (See table 4.1a and 4.1b)

Table 4.1b Baseline emissions international shipping according to Cofala et al, 2007, 2000 and
2020
2000 2020
Sea area SO
2
NO
x

NMV
OC
PM2,5 SO
2
NO
x

NMVO
C
PM2,5
x 1000 tonnes
North Sea 443 648 24 50 406 946 nr 67
Black Sea 56 81 3 6 91 118 nr 10
Mediterranean 1069 1563 60 121 1714 2311 nr 198
Baltic Sea 187 278 10 21 171 404 nr 28
NE Atlantic 494 723 26 56 804 1048 nr 91
Total 2249 3293 123 254 3186 4827 nr 394
Note: nr = not reported.

AEA 10

4.2 Description of the Policy Scenarios
4.2.1 Overview

Policy scenarios have been designed to enable the critical comparison of the effects of policy measures
that restrict emissions by:
The (already decided) designation of the Baltic and the North Sea/English Channel as SECA
The designation of the Mediterranean Sea as SECA
The designation of the Black Sea as SECA
Lowering the maximum sulphur content of maritime fuels outside designated SECA areas to
0.5%
Application of the IMO Tier I and Tier II standards
Designating the Baltic, the North Sea/English Channel and the Mediterranean as NO
x
emission
control areas (in which IMO Tier III would apply to new engines from 2015)).

Additionally, two scenarios for decreasing the emissions from land based sources have been elaborated
in which the sulphur content in heating oil for use by households and agriculture has been assumed to
be reduced to 0.05 % and 0.001% respectively (see Table 4.2 and 4.3).

Table 4.2: Scenarios for 2015
Scenario Year
S-content
marine fuel
outside SECAs
(by mass)
S-content
marine fuel in
SECAs (by
mass)
NOx Tier
Standards
SECA regions

Baseline

2015 2.7 % 1.45%
SEA1 2015 2.94% 0.10% Tier I & II
Baltic Sea
North Sea / English Channel
SEA2 2015 2.94% 0.10% Tier I & II
Baltic Sea
Mediterranean Sea
SEA3 2015 2.94% 0.10% Tier I & II
Baltic Sea
Mediterranean Sea
Black Sea
The baseline and SEA1 scenario were also combined with reduced emissions for land based emissions
Land
based A
2015
0.05 % S in MD and Gasoil for use by Consumers and Agriculture
(Abase and ASEA1)
Land
based
B
2015
0.001 % S in MD and Gasoil for use by Consumers and Agriculture
(Bbase and BSEA1)


AEA 11

Table 4.3: Scenarios for 2020
Scenario Year(s)
Max. S-content
marine fuel
outside SECAs
(by mass)
Max. S-content
marine fuel in
SECAs (by
mass)
NOx Tier
Standards
SECA regions
Baseline 2020 2.7 % 1.45% Baseline scenario (Task A1)
SEA1 2020 2.94% 0.10% Tier I & II,& III
1)

Baltic Sea
SEA3 2020 2.94% 0.10% Tier I & II,& III
Baltic Sea
Mediterranean Sea
Black Sea
SEA4 2020 0.50% 0.10% Tier I, II & III
Baltic Sea
SEA5 2020 0.50% 0.10% Tier I, II & III
Baltic Sea
Mediterranean Sea
Black Sea
The SEA4 scenario was also combined with reduced emissions for land based emissions
Land
based A
2020
0.05 % S in MD and Gasoil for use in Consumers and Agriculture
(ASEA4)
Land
based
B
2020
0.001 % S in MD and Gasoil for use in Consumers and Agriculture
(BSEA4)
Note: 1) Simulation of NECA in the Baltic Sea, North Sea/English Channel and the Mediterranean Sea (Tier III standards for ships
constructed on or after 1 January 2016)

Comparing the scenarios with each other reveals the relative benefits of individual policy measures. For
example, the comparison of SEA1 with SEA2 illustrates the impact of designating the Mediterranean
Sea as SECA, the comparison of SEA1 with SEA3 illustrates the impact of designating the
Mediterranean Sea and the Black Sea as SECA, the comparison of SEA2 with SEA3 illustrates the
additional impact of designating the Black Sea as SECA, etc.

4.2.2 SEA1

The SEA 1 scenario uses the baseline for the land based emissions. For the sea based emissions the
maximum S content within SECAs is reduced from the 1.45%, which is the post 2006 standard, to a
maximum of 0.1%; the designation of SECAs remain the same as in the baseline scenario - namely the
Baltic Sea and the North Sea/English Channel. It is expected that the additional demand for low sulphur
fuels for use in the SECAs will lead to a shift towards high sulphur fuel outside SECAs; a rough
estimate, based on the worldwide fuel consumption in and outside SECAs whereby the total amount of
sulphur is kept constant, results in an increase of the current average sulphur content outside SECAs
from 2.7 to 2.9%.

The recently revised
5
IMO standards have been used for the NO
x
emissions; for 2020 we have applied
the Tier III standards for ships constructed after 1/1/2016 in the Baltic Sea, the North Sea/English
Channel and the Mediterranean Sea areas.

The age distribution of the population of ships has been estimated based on the age distribution of the
world merchant fleet by type of vessel as of 1st January 2005 (Source: Chiffi et al 2007, Technical

5
Revised Annex VI of MARPOL and the revised NOx Technical Code adopted in October 2008, these will enter into force 1 July 2010.

AEA 12
Annex, Maritime Transport Inventory, Table 2.5). In the absence of better data it has been assumed that
the age distribution will not change over the period up to 2020.

Ships constructed on or after 1 January 1990 but prior to 1 January 2000 (and with a displacement 90
litres per cylinder and rated output 5000 kW) will be required to comply with the NO
x
emission limits in
force today, subject to the availability of approved engine upgrade kit (Tier I). After contacts with marine
engine producers MAN and Wrtsila we estimated that about 1/3 of the pre-2000 engines are already in
compliance with IMO Tier 1. Thus, for 70% of the ships constructed from January 1990 to January 2000
the baseline emission factors have been used, while for the rest of this age cohort Tier 1 emission
factors have been applied
6
.

4.2.3 SEA2

The SEA2 scenario only differs from SEA1 in the designation of the Mediterranean as a SECA region; it
has been calculated for 2015 only.

4.2.4 SEA3

The SEA3 scenario adds the Black Sea as a SECA in comparison to SEA2. In the scenario run for 2020
it is additionally assumed that the Baltic Sea, the North Sea/English Channel and the Mediterranean will
be designated as NO
x
Emission Control Areas, in which Tier III standards apply.

4.2.5 SEA4

For the SEA4 scenario, which has been calculated for 2020 only, the sulphur content of fuels used
outside SECAs has been reduced to 0.5%. For the rest SEA4 is comparable to the 2020 run of SEA1,
with only the Baltic and the North Sea/English Channel designated as SECAs and the Baltic, the North
Sea/English Channel and the Mediterranean Sea designated as NO
x
Emission Control Areas.

4.2.6 SEA5

The SEA5 scenario is the most advanced scenario: The sulphur content of fuels used outside SECAs
has been reduced to 0.5%; all European seas (except the remaining part of the North Eastern Atlantic
Ocean) have been designated as SECAs allowing a maximum sulphur content of 0.1% only. The Baltic
Sea, the North Sea/English Channel and the Mediterranean Sea have been designated as NO
x

Emission Control Areas, implying IMO Tier III standards.

4.2.7 Land based A

Apart from regulating the quality of fuels used by maritime activities, Directive 1999/32/EC on the
reduction of sulphur content of certain liquid fuels, also covers medium distillates and gasoil used in
small scale combustion on land. Combustion in industry falls under the large combustion plants directive
(Directive 94/66/EC). The current limit for gasoil in the fuel quality directive is a sulphur content of 0,10
% (since 1 January 2008). Scenario Land based A assumes a further reduction of the sulphur content
used of gasoil by domestic heating and agriculture to 0.05%. As the baseline scenario includes
emissions from all European countries, it is necessary to mention that this reduction only has been
applied for the EU27 countries only.

The LandA scenario has been combined with the baseline sea emissions (ABASE) and the SEA1
scenario (ASEA1).

4.2.8 Land based B

The Land based B scenario takes the reductions a step further, limiting the sulphur content of gasoil by
domestic heating and agriculture in EU27 to 0.001%. The Land based B scenario has been combined
with the baseline sea emissions (Bbase) and the most stringent sea scenario SEA4 (BSEA4).

6
This results in a conservative estimate of the NOx emissions

AEA 13

4.2.9 SO
2
emissions in the different scenarios
The emission figures for SO
2
in the different areas and scenarios as calculated in this study are
summarised in the Tables 4.4.and 4.5. Appendix 2 gives the detailed emission figures for all scenarios
all areas and all pollutants.

Table 4.4 Sulphur dioxide emissions, 2000-2020, all scenarios
Baltic Sea Black Sea Remaining NE Atlantic ocean
x 1000
tonnes
2000 2015 2020 2000 2015 2020 2000 2015 2020
baseline 177 149 169 53 82 93 472 730 829
SEA1 10 12 58 90 102 795 903
SEA2 10 12 58 90 102 795 903
SEA3 10 12 2 3 3 795 903
SEA4 10 12 10 15 17 135 154
SEA5 10 12 2 3 3 135 154

Mediterranean Sea North Sea
X 1000
tonnes
2000 2015 2020 2000 2015 2020
baseline 1024 1582 1795 417 351 399
SEA1 1722 1955 24 28
SEA2 59 67 24 28
SEA3 59 67 24 28
SEA4 293 333 24 28
SEA5 59 67 24 28

Note: In SEA1 and SEA2 the sulphur emissions in non-SECAs are higher than in the baseline because it is assumed that the
demand for low sulphur fuels in SECAs will push up the average sulphur content in other seas.

Table 4.5 Sulphur dioxide emissions land based sources, baseline and scenarios
x 1000 tonnes
2000 2010 2015 2020

Baseline 16726 10324 9897 9734
Land based A 9851 9694
Land based B 9806 9665

Note: exclusive of emissions related to national maritime fuel consumption (e.g. fisheries), which are included in the GAINS-totals
Source: GAINS Europe, NEC6, C&E package current policy


AEA 14
5 Dispersion results; concentrations and
exceedance of critical loads
5.1 Introduction
The LOTOS-EUROS model was used to compute (from the emission data as presented in the previous
section) the atmospheric dispersion of ship emissions, and their effect on the European distribution of
sulphur- and nitrogen deposition, and on ground-level ozone and particulate matter concentrations.

The domain of the LOTOS-EUROS model (Schaap et al., 2005) runs from 10 West till 60 East and
from 35 North till 70 North; it covers the whole of Europe (Iceland excluded) with a horizontal
resolution of 0.5latitude x 0.25longitude (~30x30 km). In vertical direction it extends to 2-3 km above
the surface resolved in four layers: a planetary boundary layer with a fixed surface layer of 25 metre
height, and two reservoir layers on top of the planetary boundary layer.

The background concentrations (from other parts of the world) are provided by a global model, TM5.
Transport is assumed to be by advection and diffusion and is based on analysed meteorological fields.
The dispersion calculation uses1997 meteorology - to stay in line with two preceding studies in 2006
and 2007 concerning the trade of emissions of NO
x
and SO
2
; in addition 1997 was put forward by TNO
(and supported by a.o. EMEP) as a year which represents an average atmosphere over Europe.

Dry deposition is described via a multiple serial and parallel resistance approach (surface-to-air, laminar
and turbulent resistance) and wet deposition is described by scavenging rates and precipitation data.
The chemistry is based on a modified CBM-IV scheme in combination with a thermodynamic equilibrium
module and it describes the chemistry for ozone, acidification, eutrophication and aerosols.

Anthropogenic emissions of 7 species (NO
x
, SO
2,
NMVOC, CH
4
, PM
10
, PM
2.5
and NH
3
) are based on
Gains Europe (NEC report Nr. 6). Emissions of the remaining component CO come from TNOs own
emission database. Furthermore, the model accounts for natural emissions from trees (isoprene,
terpene), and from soil (NO
x
).

The standard output (on an hourly basis)
7
is generated as a longitude-latitude grid which was
subsequently converted into the EMEP grid. Ecosystem burden is expressed in terms of the sulphur and
nitrogen deposition and the average accumulated exceedence (AAE) for acidification and
eutrophication. The model also delivers: concentrations of primary PM
10
and PM
2.5
and secondary
species as sulphate, nitrate and ammonium; concentrations of ground-level ozone are expressed in the
four ozone metrics - AOT40c, AOT40f, SOMO35 and annual mean.

The results of this modelling exercise will be discussed in the next paragraphs. Emphasis is laid on the
comparison between the baseline results and the specific scenarios and the difference between the
scenarios and changes between 2025 and 2020.

The most relevant results in terms of geographic distribution over Europe for the exceedence of critical
loads for eutrophication and acidification, the distribution of ozone and PM
10
are presented in maps in

7 The model has been used in several model inter-comparison studies and it shows a good performance (Roemer et al.,
2003; van Loon et al., 2004). It was used in earlier studies on SO2 and NOx emission trading (Roemer et al., 2006, van
Harmelen et al 2007). During the first EU study the model output for 2010 (meteorology: 1997) was compared with the
results from the EMEP model (See also appendix 3). LOTOS-EUROS showed somewhat (~20%) lower sulphur
deposition, mainly caused by differences in the wet deposition. In a current study it appears that the difference in total
sulphur deposition over Germany between LOTOS-EUROS and EMEP is very small. The nitrogen deposition is
somewhat (+30%) higher in LOTOS-EUROS than in EMEP.
The PM10 concentrations in LOTOS-EUROS are about a factor of two lower than the measured values. Underestimation
of PM10 values is a common problem for nearly all large scale models. The problem occurs with the primary PM10
concentrations (the part of PM10 that is emitted directly). Quite a good comparison with measurements is seen for
secondary PM10 (sulphate, nitrate).


AEA 15
appendix 3. Appendix 4 includes tables of the dispersion calculations aggregated to the different EU
member states.

5.2 SEA1
The main feature of this scenario is the lowering of the sulphur content in the Baltic and the North
Sea/English Channel SECAs, and the implementation of IMO Tier I and II for 2015. These measures
result in a reduction of exceedance (level and area) of the critical load for acidification in all vulnerable
areas in Europe. In coastal areas (up to ca 100 km from the coast) of the Baltic and the North Sea the
exceedances have been reduced considerably (by up to 100%) whereas the reduction for inland areas
is10-20%. In 2020, compared with 2015, acid deposition has further decreased, due to lower NO
x

emissions, leading to a larger coastal area that is protected. In inland areas, the gains compared to the
2015 baseline are more limited.

As an effect of implementing IMO Tier I and II, the NO
x
emissions decrease in all European seas,
resulting in a similar decrease of eutrophication all over Europe (0-10%). Larger fractional reductions
are apparent in those coastal zones of the Mediterranean where shipping is a major source of NO
x
and
exceedances of critical loads are already low - the absolute reduction, however, is the same as
elsewhere in Europe. For 2020 additional reductions are expected as a result of implementing IMO Tier
III in the Baltic, the North Sea and the Mediterranean.

Both in 2015 and 2020, the concentrations of PM
10
are reduced by 20-30% (in the coastal areas around
the North Sea, English Channel and Baltic Sea).

PM
10
concentrations in the northern part of Europe decrease by up to 10 % but in the Mediterranean
region PM concentrations rise as a result of increasing, uncontrolled, sulphur emissions. The increase in
sulphur emissions (as sulphate) outweighs the simultaneous decrease in NO
x
(as nitrate).

All round the North Sea and the English Channel the concentrations of ozone show an increase, which
is the result of the decreasing NO
x
emissions due to the introduction of IMO Tier I and II in a NO
x
rich
environment. Similar trends in modelled ozone response to a limited decrease of NO
x
emissions are
seen in urban environments and other NO
x
rich conditions over land. In the Mediterranean, where NO
x

concentrations are much lower, the reduction in NO
x
emissions leads to lower ozone concentrations - as
expressed by SOMO35 and AOT40f. By 2020 the effect of introducing IMO Tier III can be clearly seen
(Annex 2) in the Mediterranean Sea and the Baltic but less so in the North Sea because of the higher
ambient NO
x
concentration.


AEA 16
5.3 SEA2 (applies only to 2015)
SEA2 differs from SEA1 in that the Mediterranean Sea too is designated as a SECA. By 2015 this
addition results in an extra reduction of the exceedance of eutrophication by 10% and acidification by
20%. In the Mediterranean area the reductions to eutrophication can be locally larger than 20 %
because maritime NO
x
emissions here are a larger determining factor of ambient concentrations of NO
x
.

The reduction of PM
10
concentrations in the Mediterranean area is also considerable; on shipping routes
in the southern part of the Mediterranean Sea reductions up to 70% are seen whereas the
concentrations over the rest of the Mediterranean Sea decrease by 3040% and in coastal areas up to
20 %.

As a result of the significant reduction of SO
2
emissions, compared to SEA 1, the ozone concentrations
in the whole Mediterranean area (land and sea) increase slightly (see difference between the following
maps) as the result of heterogeneous chemistry in which sulphur reduction causes a slight increase of
nitrogen oxides through the formation of additional N
2
O
5
.


AEA 17

5.4 SEA3
2015
SEA3, by comparison with SEA2, assumes that the Black Sea is also designated as a SECA. Although
this results in a relatively minor additional emission reduction in Europe as a whole, and a negligible
effect in Western Europe states, the reductions in exceedance of critical loads for acidification and
eutrophication in areas close to the Black Sea (Romania and Bulgaria) is significant (up to 30% in some
grid cells).

Reduction of PM
10
is only significant over the sea and apart from a small reduction in PM
10

concentrations in Bulgaria it has a very small effect over the land. The ozone effect is also small and
resembles that of SEA2 on the Mediterranean area; the Black Sea, however, is much smaller and the
rise in ozone concentration more limited.

2020
The SEA3 scenario in 2020 includes emission reductions (as compared to SEA1) in both the
Mediterranean Sea and the Black Sea. The effects are quite comparable to the effects of the two
separate reductions combined as discussed under SEA2 and SEA3 (2015).
5.5 SEA4 (2020 only)
The most important difference in SEA4 from the preceding scenarios is the lowering of the sulphur
content of fuels used outside SECAs to 0.5%. In SEA4 the non-SECA areas are the Mediterranean Sea,
the Black Sea and the Remaining North-Atlantic. The measures result in a large reduction of the sulphur
emissions in these areas as compared to SEA1 scenario. For acidification the effect of SEA4 is mainly
seen in the UK, which is heavily influenced by the emissions in remaining North-Atlantic. Emission
reduction in the Mediterranean Sea and the Black Sea has little effect in reducing the exceedance of
acidification due to the location and distance of sensitive ecosystems.

For eutrophication the benefits of SEA4 (as compared to SEA1) are relatively small since NO
x

emissions stay the same.

PM
10
concentrations are lowered as a result of the PM
10
emission reductions following the reduction in
sulphur emissions. The largest changes are seen in the coastal areas of the three seas involved.

Due to the marked reduction in sulphur emissions there is a slight increase in the ozone (NOTE: this
phenomenon was described under SEA2).

AEA 18
5.6 SEA5 (2020 only)
In SEA5, apart from the lower sulphur content outside SECAs, the Baltic, the North Sea/English
Channel, the Mediterranean and the Black Sea have been assumed to be SECAs, resulting in a further
reduction of sulphur and PM emissions.

Of all the sea-based scenarios SEA5 yields the least exceedance of acidification and eutrophication and
the lowest PM
10
concentrations. The ozone related parameters are slightly increased but still less than
with the baseline ozone.

5.7 Land based A
2015
The emission reductions of the land-based measures are relatively small compared to the reductions
achieved by most of the marine-based measures. The land-based measures apply only to one of the
source categories of sulphur (domestic heating and heating in agriculture). In quite a few countries this
source contributes very little but in others it is larger.

Scenario A results in ~2-3% less sulphur deposition and a small decrease of the exceedance of
acidification, the effect on levels of eutrophication is really very small, less than 0.5%; the PM
10

concentrations drop about 1% (as a result of less sulphate formation and less direct PM emissions), and
the effects are negligible for ozone related parameters.
5.8 Land based B
2015
The land based measures taken under Scenario B are relatively stringent; the reduction in the sulphur
deposition is about twice that envisaged under Scenario A.

Scenario B leads to ~4-6% less sulphur deposition, which results in a rather small decrease of the
exceedance of acidification. For eutrophication the effects are less than 0.5%; PM
10
concentrations drop
by about 2% (less sulphate and less direct PM emissions), and the effects are negligible for ozone
related parameters.

5.9 ASEA1-BSEA1-ASEA4-BSEA4
The land based scenarios A and B are combined with the SEA1 scenarios (2015 and 2020) and with the
SEA4 scenario (2020). The effects of these scenarios compared to the SEA1 and SEA4 scenarios
(without additional land based measures) are as small as the effects of the A and B scenarios in relation
to the baseline (as discussed in the previous section).


AEA 19
6 Benefits assessment
The objective of this stage of the study is to quantify the health benefits of the scenarios described
above in terms of reduced impact, and then to monetise these benefits in order that they may be
compared directly with costs. The focus here is on health impacts because these dominate the
quantified (and monetised) benefits assessment. In the event that benefits do not exceed costs it would
be appropriate to expand the analysis to bring in other effects such as those on building materials and
crops. It would also be appropriate to consider in a qualitative way how un-monetisable impacts, such
as those on ecosystems and cultural heritage, would be likely to influence the balance of costs and
benefits.

6.1 Background
The benefits assessment model used in this study is ALPHA2 the Atmospheric Long-range Pollution
Health and environment Assessment model, version 2 (Holland et al, 2008). The original version of
ALPHA was developed at AEA Technology in the 1990s, drawing extensively on the ExternE research
programme, and was used to inform development of the ECs Acidification Strategy, the Ozone
Directive, the National Emission Ceilings Directive and the Gothenburg Protocol to the UN/ECE
Convention on Long Range Transboundary Air Pollution. The input data for the ALPHA2 programme
are the outcomes of the dispersion modelling reported in Chapter 5.

6.2 Methodology
The CAFE CBA methodology was described in three volumes (Holland et al, 2005a, 2005b; Hurley et
al, 2005)
8
and can be traced back to the beginning of the EC DG Research ExternE Programme that
started in 1991 and has continued more or less continually since. Further to this, the methodology used
here was the subject of intense consultation in 2003 and 2004 with stakeholders from the European
Union Member States, academic institutes, environment agencies, industry and non-governmental
organisations. It was also subject to formal peer review by senior experts in the USA and Europe (the
peer review report is available at the above website). Further work to refine the model is being carried
out under the LIFE+ Programmes EC4MACS Project
9
and the model as used here includes these
developments.

The quantification of impacts and their subsequent monetisation follows the impact pathway approach
as shown in Figure 6.1. It is in essence a logical progression from emission, through dispersion and
exposure to quantification of impacts and their valuation.

8
http://cafe-cba.aeat.com/html/reports.htm
9
http://www.ec4macs.eu/home/reports/Interim%20Methodology%20Reports/4_BenefitsAssessment_task_8.pdf

AEA 20

Figure 6.1. Impact Pathway Approach, illustrated with the example of the effects of emissions of NOx and
VOCs on ozone concentrations and crop yield

Impacts are generally calculated as:

Impact = Pollution x Stock at risk x Response function

Pollution may be expressed in terms of concentration or deposition. The term stock at risk relates to
the amount of sensitive material (people, ecosystems, materials, etc.) present in the modelled domain.
For the health impact assessment account is taken of the distribution of population and of effects on
demographics within the population, such as children, the elderly, or those of working age. Incidence
rates considered representative of the rate of occurrence of different health conditions across Europe
are used to modify the stock at risk for each type of impact quantified. These incidence rates ensure
that impacts are referenced to actual health data and hence should not be of an unrealistic magnitude.
The health impacts quantified are listed in Table 6.1 below. For the CAFE-CBA two sets of response
functions were identified, those for which evidence was considered most robust which were grouped as
the core set and a group for which quantification was considered less robust, referred to as the
sensitivity set. In this study we start by applying only the core set of functions, reserving the
sensitivity set for the situation in which it could be found that core benefits do not exceed costs.

Valuation is performed as follows:

Economic damage = Impact x Unit value of impact

Unit values seek to describe the full economic effect of the impacts that they are linked with. For health
impacts, for example, which dominate the analysis, this will include elements associated with the costs
of health care, lost productivity amongst workers and aversion to premature death or ill health.

A more complete account of the methods used for health impact quantification is provided in Volume 2
of the CAFE methodology (Hurley et al, 2005), though the valuation data used here have been revised
for the current study drawing on information from the EC4MACS work (Table 6.1). The methodological
issue that has received most debate over the years is the approach that should be taken to valuation of
mortality, whether it should be performed relative to the loss of life expectancy using the value of a lie
year (VOLY) or the number of deaths that can be linked to pollution exposure using the value of

AEA 21
statistical life (VSL). Whilst the study team has a strong preference for use of the VOLY, we have used
both VOLY and VSL estimates to demonstrate sensitivity on this issue.

There are two main differences between the values shown and those used in the CAFE-CBA work.
First, values have been increased in line with prices for the period 2000 to 2005, reflecting the different
price years used in this work compared to CAFE. Second, the value of a life year (VOLY) has been
reduced from 52,000 to 40,000 reflecting the results of the recent NEEDS study for EC DG Research.
The result of the NEEDS work may be preferred for several reasons, including:

Use of a larger total sample
Consideration of a larger number of countries (9) compared to the 3 used for the earlier
estimate
Inclusion of countries from the new member states (Poland, Czech Republic and Hungary) as
well as from the original EU15

A further factor is that the mean estimate from the new work was found to be reasonably stable (not
influenced by a small number of very large individual estimates of willingness to pay) and so could be
adopted as the best estimate of VOLY. In the earlier NewExt study (the basis of the previous estimate of
52,000 per VOLY), in contrast, the mean was not stable, and so two estimates of the VOLY,
representing the mean and the median of the study population, were adopted. Use of a single estimate
clearly simplifies the analysis. A problem in basing the VOLY purely on the results of the NEEDS project
is that it pays no attention at all to the results of the NewExt work. A final decision on the VOLY to be
recommended under EC4MACS has yet to be made, though it is currently likely that it will seek to
account for both the NewExt and NEEDS results. Basing the present analysis only on the NEEDS result
leads to a lower estimate of benefits, making it less likely that benefits would exceed costs than if a
higher VOLY estimate were applied. For the purposes of the current analysis we consider this
conservative approach to be appropriate, partly as it makes very little difference to the comparison of
costs and benefits in Chapter 7: a prolonged discussion here of whether or not a higher value should be
adopted for the VOLY would therefore add nothing to the present study. It is not recommended that the
40k estimate is adopted more widely without further discussion.

The CAFE-CBA also adopted two estimates of the VSL, again reflecting the median and the mean of
the study population. However, on reflection, this was considered unnecessary: whilst the NewExt mean
for VSL was considered unstable, reference to the wider European literature on characterisation of the
VSL suggests that a best estimate in the region of 1 million is appropriate.

Table 6.1. List of health impacts quantified with unit values applied in this analysis.
Impact / population group Pollutant Unit value ()
Acute Mortality (All ages) O
3
40,000
Respiratory Hospital Admissions (65yr +) O
3
2,364
Minor Restricted Activity Days (MRADs 15-64yr) O
3
42
Respiratory medication use (adults 20yr +) O
3
1
Chronic Mortality (30yr +) - as life years lost PM
2.5
40,000
Chronic Mortality (30yr +) as premature deaths PM
2.5
1,130,000
Infant Mortality (0-1yr) PM
2.5
1,695,000
Chronic Bronchitis (27yr +) PM
2.5
208,000
Respiratory Hospital Admissions (All ages) PM
2.5
2,364
Cardiac Hospital Admissions (All ages) PM
2.5
2,364
Restricted Activity Days (RADs 15-64yr) PM
2.5
97
Respiratory medication use (children 5-14yr) PM
2.5
1
Respiratory medication use (adults 20yr +) PM
2.5
1
Lower Respiratory Symptom days (children 5-14yr) PM
2.5
42
Lower Respiratory Symptom among adults (15yr +) with chronic symptoms PM
2.5
42

Uncertainties are of course present in all inputs to the analysis. Particularly important are those relating
to the response functions used and the valuation data. These are accounted for later in the analysis

AEA 22
when benefits of the scenarios are compared against the costs: uncertainties are only important to the
extent that they would change the conclusions drawn from the comparison of costs and benefits.
6.3 Results
Scenarios are defined in Table 6.2 drawing on information in preceding chapters. Impacts in terms of
the change in incidence of ill health, loss of longevity, etc. are shown Table 6.3 for 2015 and in Table
6.4 for 2020.

Estimates of monetised morbidity and mortality impacts for the EU27 are presented in Tables 6.5 and
6.6. Total health impacts are presented as a range, using the four different estimates of total annual
damage. The health impacts of air pollution are dominated by PM mortality, although PM related
morbidity is also significant. The most important effects (in monetised terms) for PM related morbidity
are restricted activity days and cases of chronic bronchitis.

Table 6.2. Scenario-run codes
Run Name Description Year
BASE_2015 Base 2015
SEA1_2015 Sea 1 Scenario 2015
ABASE1_2015 Land A Scenarios Base 2015
BBASE1_2015 Land B Scenarios Base 2015
AS1_2015 Land A Scenario and Sea 1 2015
BS1_2015 Land B Scenario and Sea 1 2015

BASE_2020 Base 2020

AEA 23
Table 6.3. Estimated annual health impacts in 2015 due to air pollution in the EU27
End point
Population
at risk
Impact Pollutant base_2015 Sea1_2015 Sea2_2015 Sea3_2015 abase1_2015 as1_2015 bbase1_2015 bs1_2015
Acute Mortality (All ages) All Premature deaths O3 22,368 22,047 22,153 22,153 22,366 22,051 22,363 22,054
Respiratory Hospital Admissions (65yr +) Elderly Cases O3 19,078 18,801 18,891 18,891 19,075 18,803 19,073 18,806
Minor Restricted Activity Days (MRADs 15-
64yr)
Adults Days O3 48,756,183 48,054,241 48,283,256 48,283,081 48,749,473 48,061,760 48,742,322 48,069,976
Respiratory medication use (adults 20yr +) Adults Days O3 8,920,808 8,791,789 8,834,138 8,834,103 8,919,611 8,793,172 8,918,283 8,794,675
Chronic Mortality (All ages) All Life years lost PM2.5 2,153,682 2,032,021 1,979,544 1,977,795 2,147,341 2,025,435 2,141,099 2,018,649
Chronic Mortality (30yr +) Infants Premature deaths PM2.5 221,810 209,851 204,017 203,810 221,164 209,178 220,525 208,491
Infant Mortality (0-1yr) Adults Premature deaths PM2.5 362 344 337 336 361 343 360 342
Chronic Bronchitis (27yr +) All Cases PM2.5 105,552 99,769 97,096 97,007 105,244 99,449 104,941 99,122
Respiratory Hospital Admissions (All ages) All Cases PM2.5 35,800 33,777 32,905 32,876 35,694 33,668 35,590 33,555
Cardiac Hospital Admissions (All ages) Adults Cases PM2.5 22,079 20,832 20,294 20,276 22,014 20,764 21,950 20,695
Restricted Activity Days (RADs 15-64yr) Children Days PM2.5 195,942,769 184,937,515 180,227,493 180,058,414 195,373,880 184,346,764 194,814,191 183,738,495
Respiratory medication use (children 5-14yr) Adults Days PM2.5 1,813,697 1,702,016 1,660,289 1,659,437 1,808,029 1,696,144 1,802,545 1,690,067
Respiratory medication use (adults 20yr +) Children Days PM2.5 17,371,218 16,406,775 15,976,148 15,961,648 17,320,399 16,353,956 17,270,283 16,299,718
LRS symptom days (children 5-14yr) Adults Days PM2.5 81,377,920 76,544,732 74,661,618 74,603,414 81,132,661 76,290,976 80,893,458 76,026,855
LRS among adults (15yr +) with chronic
symptoms
All Days PM2.5 173,956,597 164,208,588 159,928,445 159,785,506 173,446,398 163,678,293 172,943,386 163,132,904

Note: Results are rounded to the nearest 100.


AEA 24

Table 6.4. Estimated annual health impacts in 2020 due to air pollution in the EU27
End point
Population at
risk
Impact Pollutant base_2020 Sea1_2020 Sea3_2020 Sea4_2020 Sea5_2020 as1_2020 as4_2020 bs1_2020 bs4_2020
Acute Mortality (All ages) All
Premature
deaths
O3 22,244 21,417 21,539 21,567 21,598 21,418 21,569 21,420 21,571
Respiratory Hospital Admissions (65yr +) Elderly Cases O3 21,059 20,272 20,387 20,413 20,442 20,273 20,415 20,275 20,416
Minor Restricted Activity Days (MRADs 15-
64yr)
Adults Days O3 45,487,097 43,798,565 44,045,892 44,109,754 44,173,452 43,800,727 44,112,882 43,803,714 44,116,542
Respiratory medication use (adults 20yr +) Adults Days O3 8,722,895 8,398,086 8,445,792 8,457,617 8,469,909 8,398,511 8,458,240 8,399,083 8,458,951
Chronic Mortality (All ages) All Life years lost PM2.5 2,006,114 1,844,775 1,784,037 1,766,272 1,756,678 1,839,089 1,760,318 1,835,106 1,756,075
Chronic Mortality (30yr +) Infants
Premature
deaths
PM2.5 213,937 197,499 190,441 188,774 187,660 196,898 188,143 196,476 187,697
Infant Mortality (0-1yr) Adults
Premature
deaths
PM2.5 316 293 285 282 281 292 282 292 281
Chronic Bronchitis (27yr +) All Cases PM2.5 101,238 93,329 90,158 89,302 88,801 93,046 89,006 92,848 88,795
Respiratory Hospital Admissions (All ages) All Cases PM2.5 33,347 30,665 29,655 29,360 29,200 30,570 29,261 30,504 29,190
Cardiac Hospital Admissions (All ages) Adults Cases PM2.5 20,566 18,912 18,289 18,107 18,009 18,854 18,046 18,813 18,003
Restricted Activity Days (RADs 15-64yr) Children Days PM2.5 176,608,196 162,476,303 157,154,649 155,558,842 154,718,078 161,982,356 155,042,361 161,636,335 154,673,511
Respiratory medication use (children 5-
14yr)
Adults Days PM2.5 1,558,598 1,422,846 1,379,585 1,362,086 1,355,244 1,418,167 1,357,160 1,414,908 1,353,645
Respiratory medication use (adults 20yr +) Children Days PM2.5 16,466,061 15,161,445 14,654,192 14,512,750 14,432,597 15,115,158 14,464,269 15,082,675 14,429,751
LRS symptom days (children 5-14yr) Adults Days PM2.5 70,418,108 64,527,840 62,543,893 61,835,829 61,523,051 64,323,547 61,621,863 64,181,827 61,468,607
symptoms
All Days PM2.5 163,596,876 150,535,695 145,515,226 144,094,510 143,301,342 150,073,885 143,610,879 149,749,932 143,266,489

Note: Results are rounded to the nearest 100.


AEA 25
Table 6.5. Monetised annual health impacts under each scenario in 2015 in the EU27 (billion)
End Point Poll BASE_2015 SEA1_2015 SEA2_2015 SEA 3_2015 AS1_2015 BS1_2015
Acute Mortality (All ages) O3 0.89 0.88 0.89 0.89 0.88 0.88
Respiratory Hospital Admissions (65yr +) O3 0.05 0.04 0.04 0.04 0.04 0.04
Minor Restricted Activity Days (MRADs 15-64yr) O3 2.05 2.02 2.03 2.03 2.02 2.02
Respiratory medication use (adults 20yr +) O3 0.01 0.01 0.01 0.01 0.01 0.01
Total O3 Mortality 3.00 2.95 2.97 2.97 2.95 2.95
Chronic Mortality (30 yr+) VOLY valuation PM2.5 86.15 81.28 79.18 79.11 81.02 80.75
Chronic Mortality (30yr +) VSL valuation PM2.5 250.65 237.13 230.54 230.31 236.37 235.6
Infant Mortality (0-1yr) PM2.5 0.61 0.58 0.57 0.57 0.58 0.58
Chronic Bronchitis (27yr +) PM2.5 21.95 20.75 20.2 20.18 20.69 20.62
Respiratory Hospital Admissions (All ages) PM2.5 0.08 0.08 0.08 0.08 0.08 0.08
Cardiac Hospital Admissions (All ages) PM2.5 0.05 0.05 0.05 0.05 0.05 0.05
Restricted Activity Days (RADs 15-64yr) PM2.5 19.01 17.94 17.48 17.47 17.88 17.82
Respiratory medication use (children 5-14yr) PM2.5 0.002 0.002 0.002 0.002 0.002 0.002
Respiratory medication use (adults 20yr +) PM2.5 0.02 0.02 0.02 0.02 0.02 0.02
LRS symptom days (children 5-14yr) PM2.5 3.42 3.21 3.14 3.13 3.2 3.19
LRS among adults (15yr +) with chronic symptoms PM2.5 7.31 6.9 6.72 6.71 6.87 6.85

Total PM damage (low, VOLY valuation) 139 131 127 127 130 130
Total PM damage high, VSL valuation) 303 287 279 279 286 285

Total PM+O3 damage high, VSL valuation) 142 134 130 130 133 133
Total PM+O3 damage (low, VOLY valuation) 306 290 282 281 289 288

AEA 26
Table 6. 6. Monetised annual health impacts under each scenario in 2020 in the EU27 (billion)
End Point Poll base_2020 SEA1_2020 SEA 3_2020
SEA
4_2020
SEA 5_2020 AS1_2020 AS4_2020 BS1_2020 BS4_2020
Acute Mortality (All ages) O3 0.89 0.86 0.86 0.86 0.86 0.86 0.86 0.86 0.86
Respiratory Hospital Admissions (65yr +) O3 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05
Minor Restricted Activity Days (MRADs
15-64yr)
O3 1.91 1.84 1.85 1.85 1.86 1.84 1.85 1.84 1.85
Respiratory medication use (adults 20yr
+)
O3 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01
Total O3 Mortality 2.86 2.75 2.77 2.77 2.78 2.75 2.77 2.75 2.77
Chronic Mortality (30 yr+) VOLY
valuation
PM2.5 80.24 73.79 71.36 70.65 70.27 73.56 70.41 73.40 70.24
Chronic Mortality (30yr +) VSL valuation PM2.5 241.75 223.17 215.20 213.31 212.06 222.50 212.60 222.02 212.10
Infant Mortality (0-1yr) PM2.5 0.54 0.50 0.48 0.48 0.48 0.50 0.48 0.49 0.48
Chronic Bronchitis (27yr +) PM2.5 21.06 19.41 18.75 18.57 18.47 19.35 18.51 19.31 18.47
Respiratory Hospital Admissions (All
ages)
PM2.5 0.08 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07
Cardiac Hospital Admissions (All ages) PM2.5 0.05 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04
Restricted Activity Days (RADs 15-64yr) PM2.5 17.13 15.76 15.24 15.09 15.01 15.71 15.04 15.68 15.00
Respiratory medication use (children 5-
14yr)
PM2.5 0.002 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
Respiratory medication use (adults 20yr
+)
PM2.5 0.02 0.02 0.01 0.01 0.01 0.02 0.01 0.02 0.01
LRS symptom days (children 5-14yr) PM2.5 2.96 2.71 2.63 2.60 2.58 2.70 2.59 2.70 2.58
symptoms
PM2.5 6.87 6.32 6.11 6.05 6.02 6.30 6.03 6.29 6.02

Total PM damage (low, VOLY
valuation)
129 119 115 114 113 118 113 118 113
Total PM damage high, VSL valuation) 290 268 259 256 255 267 255 267 255
Total PM+O3 damage high, VSL
valuation)
132 121 117 116 116 121 116 121 116
Total PM+O3 damage (low, VOLY
valuation)
293 271 261 259 258 270 258 269 258

AEA 27
Table 6.7. Monetised annual health benefits compared to the relevant baseline under each scenario in 2015 in the EU27 (billion)
End point Pollutant SEA1_2015 SEA2_2015 SEA3_2015 AS1_2015 BS1_2015
Acute Mortality (All ages) O3 0.01 0.01 0.01 0.01 0.01
Respiratory Hospital Admissions (65yr +) O3 0.00 0.00 0.00 0.00 0.00
Minor Restricted Activity Days (MRADs 15-64yr) O3 0.03 0.02 0.02 0.03 0.03
Respiratory medication use (adults 20yr +) O3 0.00 0.00 0.00 0.00 0.00
Total ozone benefit 0.04 0.03 0.03 0.04 0.04

Chronic Mortality (30 yr+) VOLY valuation* PM2.5 4.87 6.97 7.04 0.25 5.13
Chronic Mortality (30yr +) VSL valuation* PM2.5 13.51 20.11 20.34 0.73 14.27
Infant Mortality (0-1yr) PM2.5 0.03 0.04 0.04 0.00 0.03
Chronic Bronchitis (27yr +) PM2.5 1.20 1.76 1.78 0.06 1.27
Respiratory Hospital Admissions (All ages) PM2.5 0.00 0.01 0.01 0.00 0.01
Cardiac Hospital Admissions (All ages) PM2.5 0.00 0.00 0.00 0.00 0.00
Restricted Activity Days (RADs 15-64yr) PM2.5 1.07 1.52 1.54 0.06 1.12
Respiratory medication use (children 5-14yr) PM2.5 0.00 0.00 0.00 0.00 0.00
Respiratory medication use (adults 20yr +) PM2.5 0.00 0.00 0.00 0.00 0.00
LRS symptom days (children 5-14yr) PM2.5 0.20 0.28 0.28 0.01 0.21
LRS among adults (15yr +) with chronic symptoms PM2.5 0.41 0.59 0.60 0.02 0.43
Total PM benefit (low, VOLY valuation) 7.79 11.18 11.29 8.21 8.65
Total PM benefit (high, VSL valuation) 16.44 24.32 24.59 17.36 18.29
Total PM+O3 benefit (low, VOLY valuation) 7.83 11.20 11.32 8.25 8.69
Total PM+O3 benefit (high, VSL valuation) 16.48 24.35 24.62 17.40 18.34


AEA 28
Table 6.8. Monetised annual health benefits compared to the relevant baseline under each scenario in 2020 in the EU27 (billion)
End point Pollutant Sea1_2020 Sea3_2020 Sea4_2020 Sea5_2020 as1_2020 as4_2020 bs1_2020 bs4_2020
Acute Mortality (All ages) O3 0.03 0.03 0.03 0.03 0.03 0.03 0.03 0.03
Respiratory Hospital Admissions (65yr +) O3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Minor Restricted Activity Days (MRADs 15-64yr) O3 0.07 0.06 0.06 0.06 0.07 0.06 0.07 0.06
Respiratory medication use (adults 20yr +) O3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Total ozone benefit 0.11 0.09 0.09 0.08 0.11 0.09 0.11 0.09

Chronic Mortality (30 yr+) VOLY valuation* PM2.5 6.45 8.88 9.59 9.98 6.68 9.83 6.84 10.00
Chronic Mortality (30yr +) VSL valuation* PM2.5 18.58 26.55 28.43 29.69 19.25 29.15 19.73 29.65
Infant Mortality (0-1yr) PM2.5 0.04 0.05 0.06 0.06 0.04 0.06 0.04 0.06
Chronic Bronchitis (27yr +) PM2.5 1.65 2.30 2.48 2.59 1.70 2.54 1.75 2.59
Respiratory Hospital Admissions (All ages) PM2.5 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01
Cardiac Hospital Admissions (All ages) PM2.5 0.00 0.01 0.01 0.01 0.00 0.01 0.00 0.01
Restricted Activity Days (RADs 15-64yr) PM2.5 1.37 1.89 2.04 2.12 1.42 2.09 1.45 2.13
Respiratory medication use (children 5-14yr) PM2.5 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Respiratory medication use (adults 20yr +) PM2.5 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
LRS symptom days (children 5-14yr) PM2.5 0.25 0.33 0.36 0.37 0.26 0.37 0.26 0.38
LRS among adults (15yr +) with chronic symptoms PM2.5 0.55 0.76 0.82 0.85 0.57 0.84 0.58 0.85
Total PM benefit (low, VOLY valuation) 10.32 14.23 15.37 15.99 10.68 15.75 10.93 16.02
Total PM benefit (high, VSL valuation) 22.44 31.90 34.21 35.71 23.25 35.07 23.83 35.68
Total PM+O3 benefit (low, VOLY valuation) 10.42 14.32 15.46 16.07 10.79 15.84 11.04 16.11
Total PM+O3 benefit (high, VSL valuation) 22.54 31.99 34.30 35.79 23.36 35.16 23.93 35.76
* Alternative estimates, not to be combined
AEA 29

In addition to the impacts by scenario, a key consideration is the location of the benefits that arise
from each scenario.

The following tables demonstrate the health benefits in Euros millions (high, VSL valuation) by
Member State against the baseline case in 2015 and 2020. These results are roughly twice as
large as the equivalent results based on use of the VOLY for mortality valuation. The main
purpose of these tables is to demonstrate the way that benefits are distributed across Member
States. An overview of the impact by scenario is provided below:

SEA 1
The impacts under SEA1 are primarily felt by those Member States that are closely located to the
North Sea/English Channel, with the greatest impacts being experienced in the United Kingdom,
Germany, France and Netherlands. This is in part driven by the SECA and the high population
densities of the Member States. There are also significant health benefits around Baltic Sea, the
other SECA region.

SEA 2
Under the SEA2 scenario the Mediterranean Sea also becomes a SECA region. This impact is
seen immediately with significant health benefits being seen in Italy, Greece and Spain. France
also experiences higher health benefits from the new SECA.

SEA 3
Under the SEA3 scenario, the Black Sea now becomes a SECA. The result of the scenario
increases the health benefits in Bulgaria, Romania, and Greece. However, the impact under SEA
3 is minimal to the change occurring under SEA1 and SEA2

AS1 and BS1
The AS1 scenario includes SEA 1 and the land based scenario (0.05% S). The difference against
the SEA1 scenario is a 400m increase, compared to the changes in sea scenario only. This
results in slightly higher health benefits in Germany, France, Italy, Netherlands, Belgium, Poland
and the United Kingdom.

The BS1 scenario includes SEA 1 and the more restrictive land based scenario (0.001% S). The
overall impact is almost double (approximately 800m increase) that of the impact of AS1 against
SEA1 (approximately 400m increase). The impact occurs primarily in the same countries as
identified under AS1.

AS4 and BS4 (2020 only)
The AS4 scenario includes SEA 4 and the land based scenario (0.05% S). The difference against
the SEA1 scenario is again minimal 400m increase, compared to the changes in sea scenario
only. This results in slightly higher health benefits in Germany, France, Italy, Netherlands,
Belgium, Poland and the United Kingdom.

The BS4 scenario includes SEA 4 and the more restrictive land based scenario (0.001% S). The
overall impact is almost is an increase of approximately 300m against the AS4 scenario. The
impact occurs primarily in the same countries as identified under AS4

AEA 30

Table 6.9 Member State Net Benefits by Scenario (Million) - (high, VSL valuation)
Member State SEA1_2015 SEA2_2015 SEA3_2015 AS1_2015 BS1_2015
Austria 121 176 176 134 151
Belgium 891 910 911 957 1,007
Bulgaria 30 127 202 33 37
Cyprus 0 8 8 0 0
Czech
Republic
192 250 250 209 227
Denmark 675 690 690 681 689
Estonia 34 36 36 35 35
Finland 73 77 77 73 75
France 1,827 2,372 2,370 1,972 2,147
Germany 3,605 3,782 3,786 3,870 4,145
Greece -6 1,133 1,166 8 24
Hungary 152 237 238 165 179
Ireland 89 91 91 94 100
Italy 501 3,714 3,716 593 676
Latvia 33 36 36 33 34
Lithuania 122 131 131 124 126
Luxembourg 11 12 12 12 13
Malta -2 45 45 -2 -2
Netherlands 1,774 1,800 1,798 1,829 1,881
Poland 1,074 1,222 1,226 1,125 1,186
Portugal 1 141 141 6 7
Romania 134 257 418 151 163
Slovakia 63 92 93 68 74
Slovenia 21 37 37 25 27
Spain 92 1,956 1,951 145 154
Sweden 470 483 483 471 481
United
Kingdom
4,500 4,530 4,536 4,586 4,702
Grand Total 16,480 24,345 24,624 17,399 18,337

AEA 31

Table 6.10 Member State Net Benefits by Scenario (Million) 2020, (high, VSL valuation)
Member
State
SEA1_20
20
SEA3_20
20
SEA4_20
20
SEA5_20
20
AS1_20
20
AS4_20
20
BS1_20
20
BS4_20
20
Austria 182 247 252 264 194 266 204 273
Belgium 1,197 1,220 1,290 1,291 1,247 1,341 1,282 1,378
Bulgaria 60 252 226 254 62 231 64 233
Cyprus 0 10 9 10 1 9 1 9
Czech
Republic
267 334 345 356 280 361 293 372
Denmark 851 866 886 889 856 892 861 897
Estonia 40 42 43 43 41 43 41 44
Finland 90 94 95 96 90 96 91 96
France 2,486 3,132 3,395 3,491 2,621 3,543 2,707 3,647
Germany 4,914 5,121 5,468 5,512 5,138 5,701 5,298 5,848
Greece 22 1,394 1,193 1,399 39 1,210 52 1,223
Hungary 192 295 289 303 208 300 213 307
Ireland 120 123 254 255 126 261 132 267
Italy 1,035 4,732 4,270 4,843 1,122 4,362 1,195 4,438
Latvia 39 42 43 43 40 43 40 43
Lithuania 153 166 172 175 154 174 156 175
Luxembo
urg
16 16 18 18 17 19 18 19
Malta -1 57 48 57 0 48 0 49
Netherlan
ds
2,378 2,408 2,540 2,545 2,427 2,592 2,456 2,623
Poland 1,421 1,594 1,647 1,676 1,472 1,698 1,502 1,739
Portugal 16 162 632 650 20 637 22 643
Romania 211 534 498 547 228 509 237 517
Slovakia 0 0 0 0 0 0 0 0
Slovenia 28 47 46 48 30 49 33 50
Spain 299 2,480 2,881 3,247 321 2,923 338 2,938
Sweden 577 594 605 611 583 610 585 614
United
Kingdom
5,860 5,909 7,032 7,038 5,951 7,110 6,017 7,188
Grand
Total
22,456 31,871 34,178 35,662 23,267 35,030 23,837 35,630

Results by Member State are shown graphically by Member State in Appendix 2.
AEA 32
Benefits outside of the EU27
In addition to the EU27 Member States the ALPHA model also includes the following 4 additional
countries:
Croatia
Norway
Switzerland
Turkey
The impact of the additional countries differs by scenario. However the additional 4 countries if
included in the analysis are able to contribute an additional 1.3billion in health benefits under the
SEA3 Scenario 2020, equating to an additional 9% benefit (low, VOLY valuation). The reason for
the additional health benefits under SEA3 in particular is the result of the additional SECA
regions; Mediterranean and Black Seas. This is combined with the high population densities that
exist in Turkey that result in a net benefit of approximately 1.27bn under this scenario.

The health impacts in Croatia, Norway and Switzerland are minor and are each comparable to
the results demonstrated by Slovakia. There is only a minor change in the results by scenario as
a result of having relatively small populations.

The table below demonstrates the additional impact for countries outside the European Union.
The results in this table can be compared against Table 6.10.

Table 6.11 Member State Net Benefits by Scenario (Million) 2020, (high, VSL valuation)
Member State SEA1_2020 SEA3_2020 SEA4_2020 SEA5_2020 AS1_2020 AS4_2020 BS1_2020 BS4_2020
Croatia 59 133 127 138 65 132 67 135
Norway 29 29 31 33 29 33 29 33
Switzerland 129 157 173 176 141 183 148 191
Turkey 141 2,171 1,882 2,183 159 1,893 169 1,904
Total 358 2,490 2,214 2,530 393 2,240 413 2,262

Total EU27 22,456 31,871 34,178 35,662 23,267 35,030 23,837 35,630
Total EU27 +4 22,814 34,361 36,392 38,193 23,661 37,270 24,250 37,892
Difference 1.6% 7.8% 6.5% 7.1% 1.7% 6.4% 1.7% 6.3%

AEA 33

Summary
A summary of the net health benefits, against the baselines are provided below. These figures of
course omit benefits arising from effects on other receptors including ecosystems, agriculture and
buildings including cultural heritage.

Table 6.12. The range of total health benefits relative to baseline for the EU27, billions
Run Name
Net Health Benefit Range
(2015)
Net Health Benefit Range
(2020)
SEA1 8 to 16 10 to 23
SEA2 11 to 24 -
SEA3 11 to 25 14 to 32
SEA4 - 15 to 34
SEA5 - 16 to 36
AS1 9 to 17 11 to 23
BS1 9 to 18 16 to 35
AS4 - 11 to 24
BS4 - 16 to 36
Note: Low = VOLY valuation, High = VSL Valuation

AEA 34
7 Cost analysis of the new IMO marine fuel
and engine standards and a potential
extension of emission control areas

This section of the report estimates the cost impact of the new IMO marine fuel and engine
standards and potential extension of the Sulphur Emission Control Areas. Furthermore, the costs for
stricter control of NO
x
emissions from shipping are explored.

The scenarios developed are aimed at illustrating the impact of varying the stringency of emission
controls in the sea regions under consideration. In the cost analysis, the scenarios represent
altering either: the SO
2
content of fuels (thereby changing the fuel costs) or introducing SO
2

abatement technology (scrubbers), as well as the introduction of NO
X
control technologies
classified into Tier I, II and III emission standards.

7.1 Cost calculation method description
The cost calculations for NO
X
and SO
2
abatement from shipping in this report have two parts. The
first part consists of calculating unit costs for the abatement option under consideration. In the
second part scenario specific costs are calculated - based on the penetration of low-NO
X
and low-
SO
2
technology options. The cost calculations are presented in low and high cost settings. In the
calculations, technologies that may, but are not certain to, reach the emission reduction levels
stipulated by the Tier standards are considered. The calculations also allow for scenario
comparison between closed loop scrubbers, open scrubbers and low sulphur fuels.

The unit costs are calculated per petajoule (PJ), corresponding to the heat of combustion of the
fuel used, for each fuel and engine category. The unit costs are based on the costs for
installation, maintenance, operation and fuel penalties for using the equipment. These costs take
into consideration the relevant technology-specific cost parameters. Furthermore, the
technologies and associated costs are split into the NO
X
emission standard categories Tier I, Tier
II and Tier III, as well as between retrofitted and new scrubbers. In the scenario where low
sulphur fuel is used instead of scrubbers, the costs for SO
2
emission abatement are calculated as
the difference in costs between the high and low fuel qualities. The costs are subject to scenario-
specific estimates on hours of use for the technologies. In this report scenario-specific operating
hours are of specific relevance for sulphur scrubbing options - because the regions considered as
SECA varies between scenarios. For NO
X
, the Tier III technologies are for all relevant scenarios
implemented in the Baltic, Mediterranean, and North Seas.

For each scenario, costs are calculated, specifying high and low costs estimates for NO
X
and SO
2

abatement. For SO
2
abatement, separate cost calculations are presented for scrubber costs and
low S-fuel costs.

All costs are given in
2005
. Adjustments of literature values and expert estimate values to
inflation and currency exchange rates have been made when necessary. In the central analysis, a
4 % investment interest rate has been used.

In obtaining the costs the following economic assumptions have been made:

The dynamics of supply / demand on price have not been considered.
The actual price of fuels or emission abatement equipment is dependent on the future market
price for the equipment. The costs estimates given in this study represent the costs that are found
today, and do not adjust for the future market prices.
AEA 35

The economies of scale are unknown and so have not been considered
The prices given in this report are all based on fairly small production volumes of abatement
equipment. Given that emission standards become more strict by 2016 and 2020, the market
volume for abatement technologies will increase, this may lead to lower unit prices due to
learning and economies of scale. This effect is not taken into account in this report.

No price effects on demand for transport or transport modal shifts have been considered.
This report does not consider any modal shifts away from marine shipping as a consequence of
increased shipping prices following costs for introducing emission abatement.

7.2 Abatement Technology Description
This section briefly describes the technical options for implementing regulations for SO
2

emissions to meet the standards corresponding to 0.1 and 0.5% S in the fuel. Further, different
options to reach Tier II and Tier III NO
X
emissions are described. The costs for the different
abatement options are discussed below.

Sulphur oxide abatement

Low sulphur fuel
The emission of sulphur dioxide is directly proportional to the sulphur content in the fuel, the
sulphur is oxidised in the combustion chamber and the mass of emitted SO
2
is twice that of the
sulphur (S) in the fuel used. Thus, a reduction in the S-content in the fuel can, in principle, meet
the lower emissions sought. Most of the high sulphur fuel (with a sulphur content of 1-3.5%) used
in ships today is heavy fuel oil or residual fuel. The fuel currently available with 0.1% S is typically
marine gasoil; it is likely that the fuel with this sulphur content, which will be required in SECA
areas after 2015, will be similar to it. Fuel with a sulphur content of 0.5% can be residual fuel that
has been de-sulphurised or marine diesel formulated from a mixture of residual oil and distillates.
Purvin & Gertz (2009) have described a number of scenarios in which the quality of the 0.5% S
fuel varies; the lowest priced fuel in that study had a high content of heavy oils.

Scrubbers
An alternative low sulphur fuel is to use an abatement technology which reduces the SO
2

emissions by an equivalent amount. While several technologies might be discussed well
established data is only available for wet scrubbing a technique widely used for SO2 reduction
from land based combustion plants. The principle of operation is that the exhaust gas is brought
into contact with water with a buffered alkalinity such that SO
2
is trapped and converted to
sulphate ions.

Scrubbers for marine applications can be divided into open (seawater) scrubbers and closed
(freshwater) scrubbers. The former utilises the natural alkalinity of sea water for the scrubbing
and neutralisation process, the residual SO
2
emission following scrubbing depend on the sulphur
content of the fuel, the pH of the seawater and the flows of exhaust and scrubber water. These
systems have been shown to achieve a high efficiency of SO
2
emissions reduction. There are
certain wash-water criteria which likely can be met in the future (and the sea water may require
treatment before being returned to sea). The systems may also be custom designed for specified
geographical areas and fuel qualities, i.e., areas with low alkalinity could be excluded and there
could be specified maximum sulphur content in the fuel used. The closed scrubbers use
freshwater with an added neutralising agent (normally caustic soda); these systems have been
demonstrated to meet the emission criteria and the wash-water criteria. In the future it is
anticipated that combined systems will be developed where seawater is used in open waters with
a strong buffering capacity together with a closed loop operational mode for use in ports,
estuaries and seas with brackish water.

AEA 36
Nitrogen oxides

Marine engine manufacturers will be able to meet the Tier II specifications from 2011 for new
engines by improving the control of fuel injection and other engine related modifications. There
are several, different, options for meeting the Tier III specifications - from 2016. The main
approaches are based either on selective catalytic reduction (SCR) or exhaust gas recirculation
(EGR) in combination with other measures. SCR has already been demonstrated to reach Tier III
emission levels on many ships while the EGR alternatives are still in the development phase.

SCR
In SCR a reducing agent, normally urea, is mixed with the exhaust gases in the presence of a
catalyst. For marine applications the catalyst is usually composed of vanadium and titanium
oxides. Under these conditions the urea is decomposed to ammonia which then selectivity reacts
with nitrogen oxides to produce nitrogen gas. The SCR systems require a minimum exhaust gas
temperature to operate effectively, this is normally around 300 C but the temperature depends
on the sulphur content of the fuel. Consequently SCR is less effective at low loads and for two-
stroke engines; in these cases it may be necessary to put the SCR between the cylinders and the
turbo in order to achieve the required temperature. The SCR operational costs include: urea
purchase and catalyst replacement; in practice catalysts need to be replaced after a few years
due to deactivation - likely to have been caused by: the formation of sulphur compounds,
impurities in the urea, or by thermal deactivation. On the other hand the reduction of nitrogen
oxides in the SCR is so effective that the engine may be tuned for low fuel consumption with
increased engine NO
X
emissions thus saving some fuel. The SCR system must, however, be
designed to avoid slip of ammonia with the exhaust - this can be achieved by means of an
oxidation catalyst after the SCR.

EGR/engine combinations
Alternatives to SCR are being developed to reach Tier III. While engines can be modified through
improved fuel injection systems and improved turbo systems these alone will not be enough to
reach Tier III and so additional methods are required. The main track seems to be to use exhaust
gas recirculation (EGR) in combination with other techniques. EGR means that some of the
exhaust is re-circulated into the engine leading to lower cylinder temperatures and reduce
formation/emission of NO
X
. The potential for NO
X
reduction is likely to be around 50%. The re-
circulated gas must not contain excessively high concentration of SO
2
or soot. Therefore a low
sulphur fuel (around 0.1%) would be needed or alternatively exhaust cleaning integrated with the
circulation system - this could be either a scrubber or a particulate filter (if low sulphur fuel is
used); EGR may also impose a fuel consumption penalty. In order to reach Tier III EGR may
have to be combined with some other technique - in all likelihood direct water injection (DWI), or
use of a fuel-water emulsion. These methods may require freshwater - implying a further fuel
penalty, there may also be a practical limit on the maximum sulphur content in the fuel.

Tier I-III
In the cost calculations two alternative means are used to reach Tier III - SCR and EGR in
combination with engine modifications and DWI (similar costs are expected with water emulsion
instead of DWI). Tier II is assumed to be reached by engine modifications. Retrofit costs for Tier0
engines produced between 1990 and 2000 to reach Tier I standards have also been estimated.

Compatibility
As mentioned above there are problems with using high sulphur fuel in combination with NO
X

abatement technologies. EGR requires very low sulphur content in the fuel or an internal
scrubber. SCR performance can fall away due to catalyst poisoning with high sulphur fuel and the
techniques using water injection or emulsion may also impose limits on the fuel sulphur content.
Consequently there are unresolved design issues with combining NO
X
reduction techniques with
scrubbers: if scrubbers are combined with SCR the exhaust gas may need to be reheated after
the scrubber to reach the operating temperature of the SCR, if the SCR is put in front of the
scrubber it will be exposed to a high sulphur gas.
AEA 37

7.3 Unit abatement costs
The abatement costs are presented as costs per petajoule of the heat value for the fuel used
(PJ); the costs are also presented as costs per tonne emission reduced - to facilitate comparison
with other cost estimates. The costs for the different control standards are dependent on which
technology can reach these standards. Abatement costs have been calculated for NO
X
and SO
2

abatement using previous literature

values (Entec 2005, UK Maritime and Coastguard Agency
2009, US-EPA 2009) and information resulting from direct communications with main engine and
equipment manufacturers,.

Sulphur oxides
In order to calculate costs for reducing sulphur emissions the following options have been
considered:

Table 7.1: SO2 abatement options
SO
2
Comment

New Scrubber
In combination with high
sulphur fuel (2.94 % )
Retrofit Scrubber
In combination with high
sulphur fuel (2.94 % )
Low sulphur HF, 0.5 % -
Low sulphur MD, 0.1 % -

The scenarios developed in this study separate costs for scrubbers from costs of low sulphur
fuels. The following text presents the costs for fuel shifts and scrubbers respectively.

Fuels
The costs of fuel shifts depend mainly on the projected future fuel prices for different sulphur
categories. The main parameters influencing the costs of fuel relate to sulphur content of crude
oils as well as the necessary investments in refinery capacities. The vessels using the fuels are
assumed to be subject to the relatively small costs of adapting to the different fuels. The fuel
costs used in this report are based on the fuel price projections in Purvin & Gertz (2009). The fuel
shift costs, specified as a cost per PJ of heat value in the fuel is given in the table below:

AEA 38
Table 7-2: Fuel costs in 2020, adapted from Purvin & Gertz, 2009 and corresponding SO2 associated
abatement cost.
Option
(%S in fuel)
Low
cost
[$ /
tonne]
High cost
[$ /
tonne]
Low cost
[k
2005
/ PJ]
High
cost
[k
2005
/
PJ]
Low SO
2

abatement
cost [ /
ton SO
2
]
Low SO
2

abatement
cost [ /
ton SO
2
]
Fuel shift
(2.94 1.5)
20 20 359 359 510 510
Fuel shift
(2.94 1)
30 30 538 538 568 568
Fuel shift
(2.94 0.5)
120 170 2 152 3 049 1 806 2 559
Fuel shift
(2.94 0.1)
280 330 4 510 5 370 3 621 4 268
Fuel shift
(0.5 0.1)
160 160 2 753 2 753 14 692 14 692

The fuel cost for heavy fuel with 2.94 % sulphur content is projected to be $ 420 / tonne in Purvin
& Gertz 2009. The fuel costs per tonne are converted to costs per PJ by considering the specific
heat value of the fuel and the conversion rate between euro and dollar.

Scrubbers
The most important parameters deciding the costs for scrubbers are whether the technology
would be installed in a new vessel or retrofitted to an existing vessel, and whether an open or
closed circuit scrubber system is chosen. The fuel penalty when using a scrubber is considered to
be 2% (Krystallion), which is used in the cost calculations. In closed loop systems, costs for
maintenance, purchase of NaOH and fresh water are also taken into consideration. For open
systems, only maintenance is considered relevant for the cost calculations in addition to the fuel
penalty. The cost for disposal of sludge is included in the cost estimates but is relatively small and
could decrease even further following construction of new sludge disposal facilities and higher
rates of usage of these facilities. For closed scrubber systems, the amount of fresh water and
NaOH used depend on the sulphur content of the fuel, and the costs will vary accordingly.

The market price for NaOH has varied over the last couple of years. In our central cost estimate,
a price for 2005 of 0.5 / litre 50% NaOH is used. This price estimate is high but engine
manufacturers estimate NaOH prices will range between 0.1 - 2 / litre, so the price is on the
lower boundary of other estimates. In 2007 the price for different grades of NaOH was 0.2 - 0.25
/ litre.

The price for urea is also subject to variation, but the engine manufacturer experts are closer in
their estimates on urea prices with a price range of 0.15 - 0.25 / litre. In our central estimate we
use an urea price of 0.2 / litre.

Cost parameters for the unit costs used in this report are given in Table 7.3.
Table 7.3: Central cost parameters
Tier operating hours at sea 4000 h/year
Scrubber operating hours at sea 4000 h/year
Urea price 0.2
2005 /
litre
NaOH price 0.5
2005 /
litre

AEA 39
Costs for different scrubber solutions are presented in Table 7.6. The costs in the Table are given
for medium sized SSD-RO engines running on residual oil fuel with 2.94 % sulphur content as
specified above. However, in the scenario calculations, costs for other engine categories are also
included. Low and high cost options are presented. These two options represent the costs
associated with either using fresh water in combination with NaOH, or sea water in the scrubber.
The investment costs (2005) are in the range 100 - 200 / kW for new installations and 200 - 400
/ kW for retrofit installations, which is in line with manufacturer estimates (Wrtsil 2009, Marine
and Energy Consulting 2009). The costs for scrubber operation vary with the required SO
2

content in the exhaust gases - and the different amounts of NaOH that will be needed. The costs
are therefore presented as costs for installations corresponding to certain sulphur content in the
fuel that would give the same SO
2
emission. For example, Scrubber-0.1 corresponds to
emissions from low sulphur fuels with 0.1 % sulphur content. The NaOH consumption needed to
reduce sulphur emissions from high sulphur fuel (2.94 %) to a level corresponding to 0.1 % S fuel
is estimated as 15 litres per MWh installed engine capacity. If trying to reach a 0.5 % S fuel
emission level, it is assumed that 12 litres / MWh are needed. For a 1.5 % S fuel emission level, 6
litres / MWh is assumed. The lifetime of the equipment is estimated to be the same as in the
ENTEC reports (ENTEC 2005).

Scrubber sludge disposal costs
ENTEC(2005) looked at the production of sludge when using a scrubber and the permitted
amount of sludge for disposal according to the standard tariff used at the port of Rotterdam.
ENTEC (2005) indicated that the amount of sludge produced would be within the current
allocations/ tariffs of a port like Rotterdam. Meech (2009) indicates that the costs associated with
sludge disposal could be small and Wrtsil (2009) do not take the costs of sludge disposal
specifically into consideration. Nevertheless, using the Gothenburg harbour port tariffs as a
reference, based on the Gross Tonnage (GT) of the vessel of maximum 0.4 / GT, and assuming
that vessels spend 500 hours between sludge discharge in ports, we have calculated that a
maximum cost of sludge disposal could be ~10 000 euro for a small vessel.

Table 7.4 Sludge disposal costs for Port of Gteborg
Scrubber sulphur sludge disposal cost estimates - based on the Gothenburg port charges in 2009
Vessel GT
SEK2009 /
GT
2005 /
GT
Number of
calls per year
Annual cost for sludge
disposal
Small 3,000 0.45 0.37 8.15 9,080
Medium 15,000 0.45 0.37 8.15 45,398
Large 50,000 0.45 0.37 8.15 151,326

These costs are much exaggerated since they allocate all sludge disposal costs to the scrubber
sludge, which would constitute only one part of total amount of sludge produced at the vessel.

Another price estimate for scrubber sludge disposal have been given by CE Delft (2006), who
used the contemporary fees at the port of Oslo (which were 0.12 / litre at the time), to calculate
the costs for sludge disposal. The study indicated that a large (40 MW installed power) container
ship would produce 0.8 tonne of scrubber sludge during 30 operating hours. This corresponds to
an hourly cost of 2.9 . A medium vessel (15 MW) would produce some 0.4 tonne over 40
operating hours, corresponding to a cost of ~1 / hour. No estimates are given for small vessels.
Wrtsil (2009) specifies that some scrubbers produce 1.3 litres sludge per MWh. By using these
numbers, the following sludge disposal costs are calculated for scrubbers,

AEA 40

Table 7.5. Sludge disposal costs.
Vessel type
Sludge production
[litres / MWh]
Sludge disposal cost /
litre []
Annual disposal costs
in the central analysis
[]
Small 1.3 0.12 1,600
Medium 1.3 0.12 5,300
Large 1.3 0.12 13,300

These costs correspond to 0.9 - 3.1 % of the total scrubber costs for a new scrubber (0.6 - 1.7 %
for retrofit).

The costs for sludge disposal are included in the O&M column in the table below. The table
presents the scrubber costs associated with a medium sized vessel reaching a sulphur content of
the flue gases corresponding to 1.5, 0.5, or 0.1 % S content in fuel.
Table 7.6: Scrubber costs used in this study
SCRUBBER COSTS
Tech Tech spec
Investment
[k/vessel]
Lifetime
[year]
O&M
[k/vessel]
Fuel cost
[k/vessel]
Annual
cost
[k]
Costs /
tonne
SO
2
[]
New
Scrubber-
1.5
open scrubber 1,148 15 28 41 167 1,035
New
Scrubber-
1.5
Closed
scrubber - high
cost
2,296 15 198 41 441 2,680
Retrofit
Scrubber-
1.5
open scrubber 2,296 12.5 28 41 301 1,840
Retrofit
Scrubber-
1.5
Closed
scrubber - high
4,592 12.5 198 41 708 4,290
New
Scrubber-
0.5
open scrubber 1,148 15 28 41 167 590
New
Scrubber-
0.5
Closed
scrubber - high
2,296 15 198 41 543 1,875
Retrofit
Scrubber-
0.5
open scrubber 2,296 12.5 28 41 301 1,045
Retrofit
Scrubber-
0.5
Closed
scrubber - high
4,592 12.5 198 41 811 2,785
New
Scrubber-
0.1
open scrubber 1,148 15 28 41 167 500
New
Scrubber-
0.1
Closed
scrubber - high
2,296 15 198 41 595 1,730
Retrofit
Scrubber-
0.1
open scrubber 2,296 12.5 28 41 301 885
Retrofit
Scrubber-
0.1
Closed
scrubber - high
4,592 12.5 198 41 862 2,505

Following the simulated impact of retrofit costs and the impact of NaOH use, the costs for
scrubbers vary with a factor of 5 between the least expensive open scrubber option and the most
expensive retrofit close circuit scrubber option. This large variation is a consequence of the
relatively expensive NaOH costs used in these cost estimates.

AEA 41
NO
X

The cost for reduction of NO
X
emissions in the scenarios studied is only applicable as a
replacement of Tier II technologies with Tier III technologies. There is currently much research
ongoing by engine manufacturers to reach Tier III standards by methods other than SCR. It is
likely that these other methods will be associated with comparatively lower costs but it is still
uncertain whether they will able to reach Tier III standards.

SCR
When calculating the costs for SCR, the cost assumed for urea is important. Its price, like that of
fuel, is heavily dependent on future supply and demand; recent fluctuations show that the future
price is uncertain. One potential advantage of SCR is that the emission reduction efficiency is
potentially so large that the engine can be modified for optimal fuel performance - the NO
X

emissions increase but remain within the capacity of the SCR system. However, this is not
considered in this study since there are different findings regarding this issue from different
engine manufacturer experts.

EGR (Exhaust Gas Recirculation) combinations
Lower cost options to SCR feature a combination of engine alterations (such as turbo charging)
and EGR and water injection. The latter come in several different forms - Humid Air Motors, HAM;
Direct Water Injection DWI; and Water Emulsion WE. The fuel efficiencies of the individual
measures act in opposing directions, EGR reduces fuel efficiency while turbo charging may
improve it. The net effect of EGR combinations in our study is assumed to imply a 2% fuel
penalty.

The Tier I and Tier II costs are based on the estimates from ENTEC (2005). In addition,
communication with manufacturers has indicated that these costs may be even lower and will
only require marginal additions for most engines.

The following list presents the emission standards and main technologies considered in the cost
calculations for NO
X
abatement.

Table 7.7: NOX control standards considered in this study
NOx % Removal Comment
Tier I retrofit 5 % NO
X

For ships constructed before
the year 2010
Tier II new 20 % NO
X
Built after 2011
Tier III new 80 % NO
X

Built after 2016 for NO
X
-
ECAs

For a typical vessel with a medium sized SSD-RO engine using residual oil with a 2.94 % sulphur
content, using the main and auxiliary engines at some 4500 hours per year (4000 hours at sea),
the following unit costs are obtained in the cost calculations.
AEA 42

Table 7.8: Tier cost estimates used in this study
Tier
Costs

Inv
[k/
vessel]
Lifetime
[year]
O&M
[k/vsl]
Fuel
cost
[k/vsl]
Annual
cost
[k]
Costs /
tonne
NO
X
[]
vintage Tech
Tech
spec

Retrofit Tier I Basic IEM 9 2.5 0 0 4 8
New
Tier
II
Adv. IEM 129 25 0 0 8 17
New
(0.1% S)
Tier
III
EGR+WIF 743 25 15 103 166 340
New
(0.5% S)
Tier
III
EGR+WIF 743 25 22 103 173 350
New
(1.45%S)
Tier
III
EGR+WIF 743 25 39 103 190 390
New
(2.7% S)
Tier
III
EGR+WIF 743 25 65 103 215 440
New
Tier
III
SCR 949 25 169 0 297 600

In the data set used for cost calculations, Tier I, Tier II and Tier III (SCR) are assumed to have the
same cost regardless of sulphur content of the fuel. The annual cost for SCR is, however,
dependent on the sulphur content of the fuel - because the sulphur in the exhaust gases
negatively affects the performance and lifetime of the catalytic elements of the SCR. The
additional costs when using high-sulphur fuels is not completely established as of yet. However,
in the scenario calculations performed here, SCR will only be used in combination with either
scrubbers or low sulphur fuels. The additional costs associated with high sulphur content in the
fuel are therefore not invoked in our calculations. For EGR + WIF solutions, the sulphur content in
the fuel will be of most concern because the exhaust gases re-circulated into the engine will be
unaffected by the installation of a main scrubber. In these options, an 'internal scrubber' is used
which de-sulphurises the exhaust gas that is re-circulated. The costs for this internal scrubber is
dependent on the amount of NaOH used which is in turn dependent on the sulphur content of the
fuel. This impact on costs is illustrated in the table above by the four different price estimates for
EGR + WIF.

The costs for Tier III equipment and consumables are within the range of the cost estimates in
ENTEC (2005) and by the USEPA (2009). The USEPA study presents significantly higher costs
per ton NO
X
abated with Tier III technology. The reason for this difference lies in the fact that the
USEPA study allocates the total cost of the abatement equipment to the relatively small amounts
of NO
X
that will actually be abated in the US coastal zone that is suggested for NO
X
-ECA.

In the scenario cost calculations used in this study are based on the cost per PJ. Having the cost
per PJ for each fuel/vessel/tech combination, scenario specific abatement costs can be
calculated based on the fuel and technology choices made for each region and year.

7.4 Scenario abatement costs
In step two of the cost calculations, scenario-specific fuel use and abatement technology
penetration is used to calculate the total costs for NO
X
and SO
2
abatement associated with the
scenarios.

The specifics of the scenarios analysed are as follows:

AEA 43
For SO
2
: the maximum sulphur content of marine fuels is 0.10% (by mass) for SECAs and 2.94
% (by mass) outside SECAs. A reduction of the maximum S-content to 0.50% outside SECAs will
also be taken into account for some 2020 scenarios. In the baseline, a sulphur content of 1.45 in
SECA and 2.6 outside SECA is considered.

For NO
X
: Marine diesel engines with a power output that exceeds 5,000 kW and a per cylinder
displacement at or above 90 litres are upgraded as follows:
Tier I standards for 30 % of the ships constructed between 1 January 1990 and 1
January 2000;
Tier I standards for ships constructed between 1 January 2000 and 1 January 2011;
Tier II standards for ships constructed on or after 1 January 2011; and
Tier III standards for ships constructed on or after 1 January 2016 and sailing in the Baltic
Sea, North Sea/English Channel and/or the Mediterranean Sea (only applicable to the
2020 scenarios)
7.5 Results
The following results are presented per scenario in aggregated and disaggregated form. First the
aggregated results are presented in tables 7.10 7.15, followed by region specific results in
tables 7. 16 7.20. Given the assumptions of the scenarios, it is assumed that scrubbers are not
used to reach the 1.45 % S content emission standards, therefore in the scrubber scenarios costs
for scrubbers will be associated with fuel cost savings that represent the difference in costs for
1.5 and 2.94 % S fuel for the Baltic Sea and North Sea.

Parameters of major importance for the cost results are as follows:

Table 7.9: Central parameters for cost estimates, 2020
Central parameters, 2020
Tier operating hours at sea 4000 h / year
Scrubber operating hours at sea 4000 h / year
Urea price 0.2 2005 / litre
NaOH price 0.5 2005 / litre

NOTE: 4000 operating hours per year corresponds to a bit less than 50 % of the vessels running time within a SECA or
NOX Emission Control Area.

The baseline scenario
The baseline scenario is described in Section 2 and is compared with the other scenarios (SEA1 -
SEA5).
Table 7.10: Scenario results - total baseline costs [million 2005]
BSL 2015 2020
Fuel 17 970 21 547

In the baseline scenario, no NO
X
removal is assumed. The costs for fuel represent the total fuel
costs for all sea regions considered in the report.

Readers should note that Table 7.10 is the only cost result table that indicates total annual costs.
In the following tables estimates all relate to the costs of the baseline scenario. In order to get the
total costs of each scenario, the costs of the baseline should be added to the costs presented in
the tables below.
AEA 44
Table 7.11: Scenario SEA1 - Total annual abatement costs on top of baseline [million 2005]
SEA1 2015 2015 2020 2020
low high low high

Alternative 1 Scrubber - scrubber
costs
683 2 003 682 2 088
SO2 Fuel cost compared to baseline -120 -120 -278 -278
Sum Alt 1 563 1 883 404 1 810
Alt 2: low S fuel cost 3 024 3 622 3 279 3 957

NOx Tier I - III 76 76 452 645

Following the introduction of Tier III NO
X
control in the Baltic, Mediterranean and the North sea,
the annual costs for NO
X
emission removal range between 452 and 645million
2005
. In the
scrubber scenario, the introduction of scrubbers in the Baltic and North Sea allows for the use of
high sulphur fuels, which are less expensive than the 1.5 % S fuel used in the baseline.
Therefore, the costs for high S fuel given in the table are negative. If reaching the SO
2
emission
levels by shifting to low sulphur fuels instead of using scrubbers, the costs are, at a minimum,
twice as high.

Table 7.12: SEA2 - Total annual abatement costs on top of baseline [million 2005]
SEA2
2015 2015 2020 2020
low high low high

Alternative 1 Scrubber
- scrubber costs
1 838 5 390
SO
2

Fuel cost compared to
baseline
-104 -104 n.c. n.c.
Sum Alt 1 1 734 5 286 n.c. n.c.
Alt 2: low S fuel cost 8 365 9 978 n.c. n.c.

NO
X
Tier I - III 76 76 n.c. n.c.
Note: n.c. not calculated

In SEA2, only covering the year 2015, the Mediterranean Sea is considered as a SECA. The fuel
costs increase significantly following the large fuel use in the Mediterranean. In the scrubber
scenario, the fuel savings are smaller relative to SEA1. The reason is that the marine diesel used
in the Mediterranean will contain 0.1% S rather than 0.5% S (as in the base case) which implies
extra costs.

AEA 45
SEA3
2015 2015 2020 2020
low high low high

- scrubber costs
1 898 5 566 1 899 5 807
SO
2

baseline
-104 -104 -261 -261
Sum Alt 1 1 794 5 462 1 638 5 547
Alt 2: low S fuel cost 8 642 10 308 9 634 11 521

NO
X
Tier I - III 76 76 393 645

In SEA3, both the Mediterranean and the Black Sea are considered as SECAs. In addition to this
the Mediterranean is considered as a NO
X
ECA. The low cost estimate on Tier III (EGR+WIF) is
lower than in SEA1 following that the Mediterranean is included in the SECA in SEA3, which
reduces the amount of NaOH needed for the internal scrubber.

SEA4
2015 2015 2020 2020
low high low high

- scrubber costs
2 432 7 069
SO
2

baseline
n.c. n.c. -278 -278
Sum Alt 1 n.c. n.c. 2 154 6 792
Alt 2: low S fuel cost n.c. n.c. 7 631 10 123

NO
X
Tier I - III n.c. n.c. 401 645

In SEA 4, the permitted sulphur content in the areas outside SECA is reduced to 0.5 %. The
Mediterranean and the Black Sea are again outside SECA. The Mediterranean Sea is together
with the Baltic and North Sea considered as a NO
X
Emission Control Area. The low cost
estimates for Tier III is lower than in SEA1 but higher than SEA3, following the allowed sulphur
content of 0.5 % (0.1% in SEA3).

AEA 46
SEA5
2015 2015 2020 2020
low high low high

- scrubber costs
2,432 7,328
SO
2

baseline
n.c. n.c. -261 -261
Sum Alt 1 n.c. n.c. 2,172 7,067
Alt 2: low S fuel cost n.c. n.c. 10,962 13,402

NO
X

In SEA5, all sea regions but the North East Atlantic are considered as SECAs (as in SEA3), and
the North East Atlantic has a maximum sulphur content of 0.5 %. This results in the highest
sulphur removal costs of all scenarios. The low cost estimates on Tier III are again following the
sulphur content of the low sulphur fuel scenario for the Mediterranean. The inclusion of the
Mediterranean and the Black Sea is not visible in the low scrubber estimates (sea water
scrubber) compared with SEA 4, since the costs for this type of scrubber is assumed to be similar
for removal to SO
2
emissions corresponding to either 0.1% or 0.5%.

Disaggregated into regions, the costs for the different scenarios are as follows:

AEA 47

Table 7.16: SEA1 - Regional annual abatement costs on top of baseline [million 2005]
SEA1 - specified per region
2015 2015 2020 2020
low high low high
Baltic SO
2

Alternative 1 Scrubber -
scrubber costs
204 597 203 623

baseline
-36 -36 -83 -83
Sum Alt 1 168 561 120 540
Alt 2: low S fuel cost 901 1,080 977 1,180
NO
X
Tier I - III 6 6 41 70
Black SO
2

scrubber costs
0 0 0 0

baseline
0 0 0 0
Sum Alt 1 0 0 0 0
Alt 2: low S fuel cost 0 0 0 0
NO
X
Tier I - III 2 2 2 2
Med SO
2

scrubber costs
0 0 0 0

baseline
0 0 0 0
Sum Alt 1 0 0 0 0
NO
X
Tier I - III 36 36 291 389
NE Atl SO
2

scrubber costs
0 0 0 0

baseline
0 0 0 0
Sum Alt 1 0 0 0 0
NO
X
Tier I - III 17 17 21 21
North SO
2

scrubber costs
479 1,406 479 1,465

baseline
-84 -84 -195 -195
Sum Alt 1 395 1,322 284 1,271
Alt 2: low S fuel cost 2,123 2,543 2,302 2,778
NO
X
Tier I - III 15 15 97 163
ALL REGIONS SO
2

scrubber costs
683 2,003 682 2,088

baseline
-120 -120 -278 -278
Sum Alt 1 563 1,883 404 1,810
NO
X
Tier I - III 76 76 452 645
AEA 48
SEA2 - specified per region 2015 2015 2020 2020
low high low high
Baltic SO
2

scrubber costs
204 597 n.c. n.c.

baseline
-36 -36 n.c. n.c.
Sum Alt 1 168 561 n.c. n.c.
Alt 2: low S fuel cost 901 1,080 n.c. n.c.
NO
X
Black SO
2

scrubber costs
0 0 n.c. n.c.

baseline
0 0 n.c. n.c.
Sum Alt 1 0 0 n.c. n.c.
Alt 2: low S fuel cost 0 0 n.c. n.c.
NO
X
Med SO
2

scrubber costs
1,155 3,387 n.c. n.c.

baseline
16 16 n.c. n.c.
Sum Alt 1 1,171 3,403 n.c. n.c.
Alt 2: low S fuel cost 5,341 6,356 n.c. n.c.
NO
X
NE Atl SO
2

scrubber costs
0 0 n.c. n.c.

baseline
0 0 n.c. n.c.
Sum Alt 1 0 0 n.c. n.c.
Alt 2: low S fuel cost 0 0 n.c. n.c.
NO
X
North SO
2

scrubber costs
479 1,406 n.c. n.c.

baseline
-84 -84 n.c. n.c.
Sum Alt 1 395 1,322 n.c. n.c.
NO
X
ALL REGIONS SO
2

scrubber costs
1,838 5,390 n.c. n.c.

baseline
-104 -104 n.c. n.c.
Sum Alt 1 1,734 5,286 n.c. n.c.
NO
X

AEA 49

Table 7.18: SEA3 - Regional annual abatement costs on top of baseline [million
2005
]
SEA3 specified per region 2015 2015 2020 2020
low high low high
Baltic SO
2

scrubber costs
204 597 203 623

baseline
-36 -36 -83 -83
Sum Alt 1 168 561 120 540
Alt 2: low S fuel cost 901 1,080 977 1,180
NO
X
Tier I - III 6 6 41 70
Black SO
2

scrubber costs
60 176 60 184

baseline
0 0 0 0
Sum Alt 1 60 177 60 184
NO
X

Tier I - III
2 2 2 2
Med SO
2

scrubber costs
1,155 3,387 1,157 3,535

baseline
16 16 17 17
Sum Alt 1 1,171 3,403 1,173 3,552
NO
X
Tier I - III 36 36 231 389
NE Atl SO
2

scrubber costs
0 0 0 0

baseline
0 0 0 0
Sum Alt 1 0 0 0 0
NO
X
Tier I - III 17 17 21 21
North SO
2

scrubber costs
479 1 406 479 1 465

baseline
-84 -84 -195 -195
Sum Alt 1 395 1,322 284 1,271
NO
X

Tier I - III
15 15 97 163
ALL REGIONS SO
2

scrubber costs
1,898 5,566 1,899 5,807

baseline
-104 -104 -261 -261
Sum Alt 1 1,794 5,462 1,638 5,547
NO
X
Tier I - III 76 76 393 645

AEA 50

Table 7.19: SEA4 - Regional annual abatement costs on top of baseline [million
2005
]
low high low high
Baltic SO
2

scrubber costs
n.c. n.c. 203 623

baseline
n.c. n.c. -83 -83
Sum Alt 1 n.c. n.c. 120 540
Alt 2: low S fuel cost n.c. n.c. 977 1,180
NO
X
Black SO
2

scrubber costs
n.c. n.c. 60 171

baseline
n.c. n.c. 0 0
Sum Alt 1 n.c. n.c. 60 171
Alt 2: low S fuel cost n.c. n.c. 149 212
NO
X

Tier I - III
n.c. n.c. 2 2
Med SO
2

scrubber costs
n.c. n.c. 1 157 3 290

baseline
n.c. n.c. 0 0
Sum Alt 1 n.c. n.c. 1,157 3,290
NO
X
NE Atl SO
2

scrubber costs
n.c. n.c. 534 1,520

baseline
n.c. n.c. 0 0
Sum Alt 1 n.c. n.c. 534 1,520
NO
X
North SO
2

scrubber costs
n.c. n.c. 479 1,465

baseline
n.c. n.c. -195 -195
Sum Alt 1 n.c. n.c. 284 1,271
NO
X

Tier I - III
n.c. n.c. 97 163
ALL REGIONS SO
2

scrubber costs
n.c. n.c. 2,432 7,069

baseline
n.c. n.c. -278 -278
Sum Alt 1 n.c. n.c. 2,154 6,792
NO
X

AEA 51

low high low high
Baltic SO
2

scrubber costs
n.c. n.c. 203 623

baseline
n.c. n.c. -83 -83
Sum Alt 1 n.c. n.c. 120 540
Alt 2: low S fuel cost n.c. n.c. 977 1,180
NO
X
Black SO
2

scrubber costs
n.c. n.c. 60 184

baseline
n.c. n.c. 0 0
Sum Alt 1 n.c. n.c. 60 184
Alt 2: low S fuel cost n.c. n.c. 313 373
NO
X
Med SO
2

scrubber costs
n.c. n.c. 1,157 3,535

baseline
n.c. n.c. 17 17
Sum Alt 1 n.c. n.c. 1,173 3,552
NO
X

Tier I - III
n.c. n.c. 231 389
NE Atl SO
2

scrubber costs
n.c. n.c. 534 1,520

baseline
n.c. n.c. 0 0
Sum Alt 1 n.c. n.c. 534 1,520
NO
X
North SO
2

scrubber costs
n.c. n.c. 479 1,465

baseline
n.c. n.c. -195 -195
Sum Alt 1 n.c. n.c. 284 1,271
NO
X
ALL REGIONS SO
2

scrubber costs
n.c. n.c. 2,432 7,328

baseline
n.c. n.c. -261 -261
Sum Alt 1 n.c. n.c. 2,172 7,067
NO
X

Tier I - III
n.c. n.c. 393 645

AEA 52
8 Comparing Costs and Benefits
The comparison of costs and benefits is carried out in two parts. First, a straightforward
comparison is made of the results from the previous sections. Results from this preliminary
comparison are then used to frame a more detailed comparison that takes more thorough
account of the uncertainties that are present.

8.1 Preliminary comparison of costs and benefits
Annual costs and benefits (low and high estimates) for the different scenarios are given in the
following tables, taking the results directly from the preceding analysis. The low bound for costs is
the lower estimate for fitting a scrubber to reduce SO
2
emissions together with the costs of NO
x

control. The upper bound for costs is for the high estimate for using low sulphur fuels to reduce
SO
2
emissions and the costs of NO
x
control. The range for benefits is based on use of either the
VOLY (lower bound) or VSL (upper bound) to value mortality.

Table 8.1 Summary of costs and quantified benefits for the 2015 scenarios (billions)

Table 8.2 Summary of costs and quantified benefits in the EU27 for the 2020 scenarios (billions)

In all cases there is no overlap between the estimates shown above: in other words that benefits
exceed costs for all of the cases shown. However, the ranges shown to this point are simply
indicative they do not account for all of the uncertainties that could affect the analysis,
particularly on the side of the benefits estimation. Without more detailed analysis of the
uncertainties that are present it is therefore unclear how robust the conclusion that benefits
exceed costs actually is (even though the results appear conclusive). To illustrate the problem,
the surplus of benefit over cost would grow were other benefits, to ecosystems, crops and the
built environment to be added in. However, if certain key aspects of the analysis are
overestimated, in particular elements relating to the quantification of benefits of reduced mortality
using the lower bound (VOLY) approach, benefits would in turn be overestimated. It is therefore
clearly appropriate to consider the uncertainties in the analysis in greater detail. This is done in
the next section which applies Monte Carlo analysis to examine more closely the effects of
different assumptions on the distribution of both costs and benefits. Table 8.3 summarises the
uncertainties accounted for and not accounted for in the ranges shown to this point, and how they
can be dealt with in the assessment that follows.
SEA1 SEA2 SEA3
Costs low 0.6 1.8 1.9
high 3.7 10 10

Benefits low 8 11 11
high 16 24 25
SEA1 SEA3 SEA4 SEA5
Costs low 0.9 2.0 2.6 2.6
high 4.6 12 11 14

Benefits low 10 14 15 16
high 23 32 34 36

AEA 53

Table 8.3 Uncertainties included and excluded from the ranges shown in Table 7.2
Element Uncertainty Accounted for?
Costs Response to legislation in terms of option applied Yes
Variability in operating costs Yes
Economies of scale No

Price effect on demand for transport or specific
transport modes
No

Price dynamics relating to supply / demand of
fuels or abatement equipment
No
Benefits Method for valuation of mortality Yes

Health incidence rates, response functions,
valuation data
No, can be assessed using
Monte Carlo

Exclusion of impacts on ecosystems and cultural
heritage (would increase benefits)
No

8.2 Detailed uncertainty assessment
It is important to be clear about the purpose of the more detailed comparison of costs and
benefits. The analysis in the preceding section already suggests that it is very likely that costs will
be less than benefits. The uncertainty analysis therefore needs to consider how robust this
conclusion is and whether there could be any grounds for coming to a different conclusion. As
will be shown, some uncertainties can be addressed quantitatively and in a lot of detail, whilst
others need to be addressed more qualitatively.

In some situations it is reasonable to estimate the possible spread of data by combining extreme
outcomes (e.g. adding or multiplying all high estimates together). However, when this involves the
combination of a large number of variables, or when it involves variables which have a large
range around them, it can provide only weak information about probable values for two reasons:
1. It does not account for the potential of errors in different parameters to cancel out against
each other
2. It does not account for the distribution of values within a range, whether they are all
equally likely or whether values somewhere towards the centre of the range are more
likely than others.
The result generated using these extreme outcomes may therefore be one possible outcome, but
it may also be extremely unlikely to happen. On this basis the use of extremes to generate ranges
is not helpful for informing policy development unless it demonstrates unequivocally that the
range for two parameters of interest (e.g. total cost and total benefit) do not overlap.

It has been observed that the ranges given in Tables 8.1 and 8.2 do not overlap. However, for the
benefits in particular the ranges shown account only partially for the uncertainties that are
present, as they show the effect of using a best estimate of the VOLY or a best estimates of the
VSL to value mortality. They do not account for uncertainty in these best estimates or in the other
parameters used in the quantification of impacts and monetary benefits.

This section factors in the uncertainties in the estimation of benefits relating to the population at
risk, the incidence of ill health, concentration-response functions and unit values for each of the
health impacts that have been quantified. This generates not only a range around the final
estimate of benefit, but also the probability distribution within that range. Extreme values may
arise, but they can be shown to be of very low probability compared to values closer to the centre
of the distribution.

The ranges described for costs in the previous chapter account for all available data accessed
during the project. These ranges are adopted for comparison with the newly-computed ranges for
AEA 54
benefits. However, we have only limited information to tell us how values are distributed within the
cost ranges. It is therefore necessary to make some assumptions on the shape of the
distributions. General options are shown in Figure 8.1, illustrating log normal, normal, triangular
and uniform distributions (note that the figure is illustrative and that the exact values and
probabilities shown are not important to the results of the present study). Other distributions are
possible, but are broadly similar to those shown. In cases where there is a reasonable amount of
information it may be appropriate to assume a detailed distribution like the lognormal or normal
case. In the present situation, however, we have only limited information and so it is appropriate
to adopt a simpler assumption with the distribution around cost estimates assumed to be
triangular or uniform. Choices made below reflect, in part whether a worst case or average/best
case situation is under investigation.

0.0
0.1
0.2
0.3
0.4
-
4
-
202468
1
0

0.0
0.1
0.2
0.3
0.4
-
4
-
202468
1
0

0.0
0.1
0.2
0.3
0.4
-
4
-
202468
1
0

0.0
0.1
0.2
0.3
0.4
-
4
-
202468
1
0
P
r
o
b
a
b
i
l
i
t
y
Value
Lognormal Normal Triangular Uniform

Figure 8.1 Examples of possible distributions for variables

Once distributions have been determined for the input variables the Monte Carlo analysis is
runover 10,000 iterations. At each iteration a value is selected at random from within each of the
defined probability distributions that have been input to the model. The resulting distributions can
be used to assess the extent of overlap in the distribution of cost and the distribution of benefit,
telling us the probability that quantified benefits will exceed costs.

The analysis is focused on the scenarios for the year 2020 rather than 2015 to account for more
complete implementation of the policies under investigation.
Describing the uncertainties in cost estimates
Three options have been considered for sulphur controls, the use of open or closed loop
scrubbers and the use of low sulphur fuels. For each technique a range of costs was derived
taking into account various uncertainties (see the preceding chapter), and these ranges were
used for the preliminary comparison of costs and benefits above. It should be pointed out that the
low range scrubber costs are for sea water scrubbers. It is still unclear under what regulations
these will operate (see Section 9). A likely scenario is that scrubbers operate with seawater at
open sea and as closed scrubbers or with low-sulphur fuel close to shore. It may also be that
seawater scrubbers are not used in sea regions with low alkalinity (i.e., parts of the Baltic Sea
and the Black Sea. Therefore, a probable cost for scrubbers could be in the middle or higher part
of the range given here for scrubbers. Various techniques were investigated for NO
x
control, as
listed in Table 7.6. Total annual costs for 2020 are shown in Table 8.4 from which the following
are apparent:
Costs are dominated by controls on sulphur, the costs of controls on nitrogen being an
order of magnitude less (with the partial exception of SEA1)
The extent of variation in the NO
x
estimates is small compared to variation in the SO
2

control costs
Despite a significant level of variation in costs for the two S control options, there is no
overlap in their ranges.
AEA 55

Table 8.4 Summary of annual abatement costs given in Chapter 6, million
SO2, scrubbers
SO2, low sulphur
fuel
NOx
SEA1, 2020 400 1,800 3,280 3,960 450 650
SEA3, 2020 4,640 5,550 9,630 11520 390 650
SEA4, 2020 2,150 6,790 7,630 10,120 400 650
SEA5, 2020 2,170 7,070 10,960 13,400 390 - 650

These observations imply that the main focus of the uncertainty analysis should be on the
variation in SO
2
control costs. They also raise questions about the appropriate shape for the
probability distribution to be associated with the costing, with the following variations considered:
Best case: Triangular or uniform distributions based on the sum of NO
x
control costs with
the low range for SO
2
control costs (using scrubbers).
Average Case: Triangular or uniform distributions across the full ranges shown, which
would assume that ship owners opt for a mixed approach to new regulations. A uniform
distribution is applicable where all values within a range are considered equally likely,
whilst a triangular distribution applies when it is thought most likely that the true value lies
towards the centre of the range.
Worst Case 1: Triangular or uniform distributions based on the sum of NO
x
control costs
with the high range for SO
2
control costs (using low sulphur fuel).
Worst Case 2: A point estimate for costs at the upper end of the range: if this worst case
generates a high probability of producing a net benefit there is no need to be concerned
about other parts of the cost range as they would simply further increase the probability.

For the first case we apply a triangular distribution with the most likely value assumed to be in the
centre of the range. For the second case we assume a triangular distribution with the most likely
estimate in the middle of the range on the assumption that some ships would use scrubbers and
some would use low sulphur fuel. For the third case we assume a uniform distribution, as this
seems likely to give a lower probability of attaining a net benefit (remembering that it is intended
as a worst case) given the observed relationship between costs and benefits from Tables 8.1 and
8.2. Resulting distributions can be compared for scenario SEA5 in Figure 8.2. The two worst
cases are shown to be significantly more pessimistic than the average case.

Triang(2.416, 4.99, 7.564)
0.00
0.15
0.30
0.45
0.60
0369
1
2
1
5
Uniform(11.513, 13.953)
0.00
0.15
0.30
0.45
0.60
0369
1
2
1
5
Uniform(11.513, 13.953)
0.00
0.15
0.30
0.45
0.60
0369
1
2
1
5
Triang(2.4, 8.2, 14)
0.00
0.15
0.30
0.45
0.60
0369
1
2
1
5
Value
1.0 1.0
1.0
0.45
0.30
0.15
0
1.0
Triangular (2.42, 4.99. 7.56) Triangular (2.42, 8.2, 14) Uniform (11.5, 13.9) Point estimate (13.9)
Cheapest Most expensive
Best case Average case Worst case 1 Worst case 2
1.0
0.45
0.30
0.15
0
1.0
0.45
0.30
0.15
0
1.0
0.45
0.30
0.15
0
0 3 6 9 12 15 0 3 6 9 12 15 0 3 6 9 12 15 0 3 6 9 12 15
P
r
o
b
a
b
i
l
i
t
y

d
e
n
s
i
t
y

Figure 8.2 Probability distributions used for the sensitivity cases on costs (examples are
all for scenario SEA5).

AEA 56

Uncertainties in benefit estimates
As noted above, account is taken here of uncertainties in the population at risk, the incidence of ill
health, concentration-response functions and unit values for each of the health impacts that have
been quantified. Ranges and associated probability distributions for each parameter are defined
by Holland et al (2009). Bringing these data together is found to generate a log-normal
distribution as shown in Figure 8.3 which takes the example of scenario SEA5 in 2020. The
results in this form provide us with a mean value and confidence intervals the 90% interval
being shown which covers a range from 9.23 billion/year to 26.9 billion/year. Using these
results it is now possible to compare against the cost ranges shown in Figure 8.2.

Value (billion/year)
P
r
o
b
a
b
i
l
i
t
y

d
e
n
s
i
t
y
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0.080
0.090
0.100
Mean=16.6
0 11 22 33 44 55 0 11 22 33 44 55
5% 90% 5%
9.23 26.9
Mean=16.5
Value (billion/year)
P
r
o
b
a
b
i
l
i
t
y

d
e
n
s
i
t
y
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0.080
0.090
0.100
Mean=16.6
0 11 22 33 44 55 0 11 22 33 44 55
5% 90% 5%
9.23 26.9
Mean=16.5
P
r
o
b
a
b
i
l
i
t
y

d
e
n
s
i
t
y
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0.080
0.090
0.100
Mean=16.6
0 11 22 33 44 55 0 11 22 33 44 55
5% 90% 5%
9.23 26.9
Mean=16.5

Figure 8.3 Calculated probability distribution for the benefits of Scenario SEA5 in 2020,
with mortality valued using the VOLY

Comparing costs and benefits accounting for uncertainties
Results from the Monte Carlo analysis showing the probability of gaining a net benefit for each
SEA scenario accounting for 8 sensitivity cases are shown in Table 8.5 and Figure 8.1
considering the set of uncertainties that have been addressed t this point. The following
observations are made:
:
Using the VSL to value mortality the probability of deriving a net benefit is more than 98%
for all SEA scenarios, even for Worst case 1 (the extreme worst case).
Using the VOLY for mortality valuation the lowest probability of gaining a net benefit is
65% for the SEA5 scenario (compare Worst Case 2 in Figure 8.2 with the distribution
calculated for benefits in Figure 8.3).
AEA 57
For the best case assumptions (lower range for S control costs with full range for N
control costs) there is a more than 99% probability of deriving a net benefit even when
applying the VOLY.
For SEA1 the probability of gaining a net benefit is greater than 98% in all cases.
Perhaps most importantly, for the average case scenario there is a probability greater
than 94% of gaining a net benefit for all SEA scenarios.

The precise probability estimates reported in the table are themselves a function of assumptions
made in the uncertainty analysis. What is important therefore is not the precise figures shown,
but the overall perspective that when further uncertainties are taken into account, the view that
benefits of the proposed actions are very likely to exceed the costs is unchanged.

Table 8.5 Probability of deriving a net benefit under various sensitivity cases
Worst 1 Worst 2 Average Best
VOLY estimates
SEA1 98.7% 99.2% 99.8% 100.0%
SEA3 66.3% 75.2% 95.6% 99.8%
SEA4 79.7% 88.3% 96.9% 99.5%
SEA5 65.3% 75.5% 94.8% 99.7%
VSL estimates
SEA1 99.8% 99.8% 99.9% 100.0%
SEA3 98.1% 98.8% 99.7% 99.9%
SEA4 99.0% 99.3% 99.7% 99.8%
SEA5 98.1% 98.7% 99.6% 99.8%

0
0.2
0.4
0.6
0.8
1
1.2
Worst 1 Worst 2 Average Best
P
r
o
b
a
b
i
l
i
t
y

o
f

b
e
n
e
f
i
t
>
c
o
s
t
VOLY estimates
SEA1
SEA3
SEA4
SEA5
VSL estimates
SEA1
SEA3
SEA4
SEA5

Figure 8.1 Probability of deriving a net benefit under various sensitivity cases.

AEA 58
8.3 Summary of the CBA results
Results presented in this section demonstrate that the probability of benefits exceeding costs for
all of the different scenarios investigated is high, in all cases greater than 65%. The IPCC
(Houghton et al, 2002) provides some guidance on the interpretation of probabilities, as follows:
<1% Extremely unlikely
110% Very unlikely
1033% Unlikely
3366% Medium likelihood
6690% Likely
9099% Very likely
>99% Virtually certain

For the cost ranges that we consider most reliable from the sensitivity analysis presented above
(the average or best cases, rather than the worst cases) the probability of benefit exceeding cost
can therefore be described as very likely / virtually certain, irrespective of the approach taken to
mortality valuation. With one slight exception (SEA5 with mortality valued using the VOLY) the
probability of benefits exceeding costs can be described as likely, very likely or virtually certain
even for the worst case assumptions on cost.

It is to be remembered that the probabilities calculated here are quantified only against benefits
for improved health. The inclusion of impacts to materials (including cultural heritage) and
ecosystems, in particular, would add to the benefits and further increase the already high
probabilities. Economies of scale, as the use of abatement techniques becomes more
widespread have also been excluded from the analysis. These would tend to reduce costs from
the levels used here which are based on data from current suppliers.

It is more difficult to make a qualitative appraisal of the effect of other uncertainties that have
been excluded from the current analysis, for example price effects on demand for transport and
the potential for modal shift in response for legislation, or price dynamics of supply/demand for
fuels, equipment, etc. Some account has been taken of such variation, for example in the costs of
abatement reagents such as NaOH and urea. It is, however, not beyond possibility that there
could be even more significant variation in cost than that considered here in the future, and that
this would lead to a net cost being incurred. However, from observations made here on the
balance of costs and benefits we consider this unlikely for the following reasons:
There is a high probability of gaining a net benefit even against the upper range for
abatement costs
The upper range for costs is about a factor 2 greater than the lower range
We consider it likely that actual costs would be in the lower range than the upper range

The following should also be considered:
Alternative methods for reaching the same goals have been identified. If the costs of one
option were to increase significantly it may therefore be appropriate to switch to a
different option
A conservative best estimate was adopted for the VOLY.

The challenge of accounting for all possible uncertainties in this analysis, given the number of
inputs and factors that need to be accounted for, is great. It would undoubtedly be possible to
make alternative assumptions in some areas to those taken here and this would change the
probability of gaining a net benefit. However, as the ranges etc. used here have been derived
from discussion with a number of manufacturers (on the cost side) and on several occasions with
stakeholders (on the benefits side), the principal conclusion reached, that benefits are very likely
to exceed costs, seems robust.
AEA 59
9 Emission abatement equipment
9.1 Summary
This section of the report discusses technical criteria for exhaust abatement equipment for the
reduction of emissions of sulphur dioxide from ships. The only technology available at present is
scrubbing. Scrubber units can be of two kinds, seawater (open) scrubbers and freshwater
(closed) scrubbers, or a combination of these. Criteria for the SO
2
to CO
2
ratio for emissions to air
have been presented by the IMO. Also criteria for the wash-water have been developed which
are discussed in this report.

At present there are criteria for wash-water but no regulations regarding emissions to air from
shipping (except for sulphur dioxide and nitrogen oxides); this is unfortunate from an
environmental point of view and better optimised abatement units could be developed if the
regulations covered both the emissions to air and to water.

The wash-water criteria developed by IMO covers pH, PAH content, turbidity and nitrate content.
While the criterion for pH seems to limit the amount of water that will be affected by low pH during
scrubber operation to a reasonable amount specific attention is needed for waters with low
alkalinity. The criteria for PAH is set in "phenanthrene equivalents"; it is not clear how this is
defined but it seems to mean that the concentration of phenanthrene is monitored. The criterion
corresponds to an emission of 2.2 mg/kWh of phenanthrene, this seems to be a high limit and
widespread use of scrubbers emitting wash-water with PAH concentrations close to this limit
could pose a risk to the environment. The criteria for turbidity can be a way to show that the
scrubber is operating correctly but the method has not been shown to be a reliable measure of
the concentration of particulate matter emitted with the wash-water. There is also a criterion for
nitrate content in the wash-water. A scrubber unit on a ship operating close to this limit could,
depending on the sea area, increase the ships' contribution to the nitrate load in the sea. It is
suggested that regulations are constructed so that this can be avoided in sensitive areas.

The IMO regulations allow two monitoring schemes. In one the SO
2
-emission to air is measured
continuously and in the other the scrubbers operational parameters are monitored to check its
efficiency of operation; it is suggested that the latter is applied because there will be scrubbers,
notably closed scrubbers, where the operation can be monitored without continuously measuring
the SO
2
concentration in the exhaust gas.
AEA 60
9.2 Introduction
Emission of sulphur dioxide from burning of fossil fuels is still the most significant source of
sulphur pollution to the atmosphere. These emissions have historically been the most important
source for the acidification of lakes and soils leading to deterioration of soils, collapsing
ecosystems and loss of forest production. Sulphur emissions also lead to the formation of
particulate matter. High levels of particulate matter are a major health concern in large parts of
Europe. In many parts of the world the emission of sulphur has recently decreased due to policy
measures such as lower use of coal-burning, cleaning of exhaust gases, sulphur limits on diesel
and gasoline for road transportation, etc.

One sector that, until recently, has been unaffected by regulations regarding sulphur emissions is
shipping. Here it is customary to use residual oil that contains high levels of sulphur in addition to
e.g., high levels of aromatics. In principle, all sulphur in the fuel ends up as emission of sulphur
oxides (SO
X
) after combustion in the engine. Some of this SO
2
is then oxidized further and may
then form particles. There is a clear connection between the sulphur content in the fuel and
particle emissions (Winnes 2009). Eyring et al. (2005) estimate that the worldwide annual SO
2

emission from shipping is 12.0 Mtonne (almost three times more than for all road traffic).

There is now a maximum allowed sulphur level of 4.5% agreed on worldwide. Further, so called
SECA areas, (from 2006 the Baltic Sea, and from 2007 also the North Sea and the English
Channel) have a limit of 1.5% sulphur. In view of the environmental and health impacts of sulphur
emission IMO have decided on future reduction in the sulphur content in marine fuels according
to Table 9.1.

Table 9.1. Sulphur limits in marine fuels
From year
Max S-content
(wt%) in fuel
Area
2005 4.5 Worldwide
2006 1.5 SECA
2010 July 1 1.0 SECA
2012 Jan 1 3.5 Worldwide
2015 Jan 1 0.1 SECA
2020 Jan 1* 0.5 Worldwide
*may be postponed to 2025

In addition to these regulations there are some regional limits for sulphur in marine fuels. Within
the European Union ships at berth and ships using inland waterways must use a fuel with less
than 0.1% sulphur from 2010. Ferries within the EU must use a fuel with less than 1.5% sulphur.
In the waters outside California ships have not been able use heavy fuel oil since July 1, 2009
and the sulphur content is restricted. In Sweden a system of lower fairway dues for shipping using
low-sulphur fuel has been in use for a number of years.

There are two principle approaches to reducing SO
2
emissions from shipping. The first alternative
is to use a low sulphur fuel - either desulphurised residual oil or marine gasoil. The second is to
remove the sulphur oxides by exhaust gas cleaning; this allows for the continued use of the
relatively less expensive high-sulphur heavy fuel oil together with an SO
2
abatement system. The
most practical abatement method is gas scrubbing, a technique where the sulphur oxides react
with water and are captured as sulphates. This is an established technique that has been used
for many years for land-based power plants etc. Scrubbers are also frequently used on large
tankers as a way to produce inert gas that is used while unloading the cargo. The water used for
scrubbing can be either re-circulated with a continuous addition of alkali, or alternatively, sea
water - where the natural alkalinity of sea water is utilised for the capture of SO
2
. The removal
efficiency is normally high but depends on the alkalinity of the water. The latter also decides the
acidity of the effluent water together with factors like the scrubber water flow in relation to exhaust
AEA 61
flow. The scrubber will also remove a significant part of the particles and a minor part of the
nitrogen oxides (NO
x
) from the exhaust. The seawater is by various means filtered before
disposed into the sea, producing sludge that should be treated on land.

The regulations stating the sulphur limits in Table 9.1 allow for the use of abatement
technologies, such as scrubbers, as an alternative to low-sulphur fuel. This is stipulated in
Regulation 4 of Annex VI to MARPOL 73/78 and in article 4C of 2005/33/EC. The "equivalence
clause" states that an alternative method should be at least as efficient in terms of emission
reduction as the use of low sulphur fuel would be. Thus, the use of abatement of SO
2
in
combination with HFO should be such that the environmental and health impacts are less than
what they would have been had low-sulphur fuel (according to Table 9.1) been used.

9.3 Abatement of sulphur oxides
Scrubber systems have been used since the 1930s in order to remove SO
2
from flue gases. The
basic principle is that SO
2

reacts with substances in a spray of water, which essentially keeps
pollutants in the exhaust from being emitted to air by capturing them in the liquid. In many land
based systems, SO
2
is treated with limestone to form gypsum but some facilities utilise the
natural alkalinity of sea water for sulphur scrubbing. On land, installations of SO
X
-scrubbers are
relatively common because land based industries do not have the same constraints of space and
complexity as ship based equipment. Alternatively freshwater, in combination with caustic soda,
can be used as the scrubbing agent. The solid particles that end up in the water are allowed to
settle in a sludge tank and the wash-water is treated before being disposed into the sea. If this
treatment is not carried out, particulates and other pollutants will reach the sea water.

The most important chemical reactions that take place during the desulphurisation process are
the following.

SO
2
(gas) + H
2
O + O
2
==> SO
4
2-
+ 2H
+

HCO
3
-
+ H
+
==> CO
2
+ H
2
O

SO
2
absorbed in seawater reacts with oxygen to form sulphate ions and hydrogen ions. Increased
concentration of hydrogen ions means increased acidity and decreased pH. Bicarbonate ions
(HCO
3
-
) from the seawater react with hydrogen ions and thereby neutralise the acidity and raise
the pH. SO
2
consumes oxygen to form sulphate and SO
2
concentration therefore corresponds to
a certain Chemical Oxygen Demand (COD).

In addition to the capture of SO
2
some NO
x
and CO
2
may be removed from the exhaust gas in the
scrubber. For NO
x
it is mainly the NO
2
fraction that is captured and it will be emitted with the
wash-water as nitrates. While some CO
2
may be captured and emitted with the wash-water this
gives no net capture of CO
2
because the balance between CO
2
in sea and air will be re-
established by CO
2
emission from the sea.

The alkalinity of the water is thus a key factor for the acidity of the effluent water. Furthermore,
the efficiency of the scrubber process increases with higher alkalinity. There are two principle
designs of marine scrubbers: seawater or open scrubbers and fresh water or closed loop
scrubbers. In a marine seawater scrubber the flue gas either passes through a spray of seawater
or is bubbled through seawater. The alkalinity of seawater is used to capture the sulphur and
neutralise the wash-water. The sulphur is discharged into the sea with the treated wash-water
which is acidic. In a freshwater scrubber the alkalinity is maintained through the addition of
caustic soda and a bleed off of the scrubber water is emitted to the sea after treatment. There are
also the possibility to combine the two types so that, for example, closed loop operation can be
used in waters with low alkalinity and seawater operation at open sea. A number of trials have
been undertaken with seawater and freshwater scrubbers onboard ships (see Skema 2009).
AEA 62

Three principle functions can be identified. 1. scrubbing of the flue gas where SO
2
, particles and
other constitutes are trapped in the seawater. 2. Filtering the seawater and dilution before
disposal overboard. This leaves a sludge that is kept onboard for disposal at shore. 3. Demisting
and heating of the exhaust gas after the scrubber unit in order to prevent a mist with low pH.

9.4 A comparison between reducing SO
2
emissions with low sulphur fuel or scrubbers
The intention of the equivalent clause described above is to allow for abatement technologies to
be used in combination with heavy fuel oil which may be economically advantageous. At the
same time, the environmental impact when using the abatement technology should be less or
equal to what would be the case if low-sulphur fuel were used. The writing specifically
emphasizes emissions of sulphur dioxide but there are also other factors to consider. It can be
useful to consider impacts which may be expected from the use of scrubbers in combination with
high-sulphur heavy fuel oil and compare to the impact from use of low sulphur fuel oil.

The abatement of SO
2
emissions required to be equivalent to certain sulphur content in fuel has
been calculated by IMO and is expressed as a ratio between SO
2
and CO
2
concentrations in the
exhaust gas. Thus, the SO
2
emissions for a ship using scrubbers, when fulfilling these criteria, will
be lower or equal to what it would be with the stipulated fuel sulphur content. However, in many
cases scrubber equipment would be able to reduce the SO
2
emissions significantly further, for
example when the ship is used outside SECA areas. Since there is a running cost for a scrubber
system, due to power consumption in pumps and, for closed loop systems, consumption of
NaOH, it is likely that the scrubbers will be used so that the regulations are just fulfilled.
Consequently there is potential for further reduction in SO
2
emissions that may be accessed
through policy initiatives.

In the production of low-sulphur fuel the sulphur from the refineries can be used for other
products. There is a fuel (and thus CO
2
) penalty in producing low sulphur fuel the magnitude of
which depends on a number of factors in the refinery design and the crude oil that is used. For
ship-based scrubbers the sulphur is emitted to the sea as sulphate ions; these will not constitute
any significant addition to the sulphates already present. However, the acidity of the effluent may
be an issue see above, its neutralisation is achieved either using the natural alkalinity of the
sea water (open scrubbers) or by an alkaline chemical (closed scrubbers). In both cases there
will be a net CO
2
emission, either during the production of NaOH or from the sea when carbonate
and bicarbonate ions neutralise the sulphuric acid.

The emission of particulates will be significantly reduced if the sulphur content in the fuel is low
because there will be less sulphate-based particulate matter produced. If at the same time there
is a switch from residual oil to distillate fuel then the PM emission will be reduced even further
since the high concentrations of polyaromatics and ashin the residual oils result in high emissions
of particles. A scrubber unit, as well as removing SO
2
, has the potential of capturing a large
fraction of the PM that would otherwise be emitted to the air; this approach reduces emission of
soot particles and a range of metals and PAH. If the scrubber wash-water were not treated,
however, the particulate matter would be discharged into the sea which is common for for
scrubbers used to produce inert gas for some ships. The level of wash-water contamination for
sulphur abatement scrubbers would be sufficiently high that it should be treated before being
discharged to the sea in order to avoid risk to the local marine environment. Treating the effluent
will produce sludge that must be taken ashore for safe disposal. However, since the regulations
focus on using scrubbers for sulphur abatement only, they are not well adapted to utilize the
potential of scrubbers as PM abatement systems; this is because a scrubber system that has not
been designed, specifically to abate exhaust gas particulate can more easily comply with the
wash-water criteria - less PM will be present in the wash-water and less treatment is needed. A
potential solution would be to combine wash-water criteria and PM exhaust emission criteria in
order to require that the scrubber units abate PM from the exhaust gas and treat the wash-water
effectively. The two main alternatives to comply with the sulphur emission regulations in a SECA
AEA 63
area after 2015 are to use marine gasoil with 0.1% sulphur or to use scrubbers; the latter option
then allows the use of high sulphur residual oil. The first alternative is likely to give a reduction in
PM emissions of about 90% (on mass basis) relative to residual oil with 1.5% sulphur. For outside
SECA areas after 2020 heavy fuel oil with a sulphur content of 0.5% is likely to be used. This will
give a reduction of around 65% relative to heavy fuel oil with around 3.5% sulphur. The reduction
of PM emissions for the scrubbers that have been tested for marine applications has varied
between 0 and 85%. The scrubbers were designed for SO
2
abatement and while a high PM
abatement efficiency may be possible it is not a trivial matter to measure the PM trapping
efficiency -when the temperature and humidity varies greatly between the measurements. In
summary, the use of marine gas oil with low sulphur content will significantly reduce the emission
of PM compared with the situation today. This will reduce the deposition to the seas as well as
the contribution to the concentration of PM in the air over land. The use of scrubbers can also
reduce the emissions of PM depending on the design of the unit. The emissions to water would
be higher if the effluents were not treated. With effective treatment sludge will be produced and
disposed at land. At present there are no real incentives to make scrubbers an effective
abatement technology for PM.

Regarding nitrogen oxides, scrubbers are likely to capture the NO
2
-fraction in the exhaust and
may be designed to capture NO as well. Doing so may cause an emission of nitrates into the sea
with the scrubber water which is unwanted in areas at risk of eutrophication. NO
x
emitted with
the exhaust gas is distributed over sea and over land so a scrubber that captures NO
x
may
redistribute the N deposition between the land and the sea, it would also decrease the
contribution to particle formation from nitrates.

Another difference between the two alternative methods of reducing SO
2
emissions - scrubbers
and low sulphur fuel - is that ships would carry different fuels. In the case of an accident or with
diffuse oil spills there could be either HFO or MGO present, each with somewhat different
properties and different harmful effects, likely more severe in the case of HFO. Furthermore, a
greater and more efficient use of scrubbers will produce sludge, for ports to handle, that is
potentially toxic.

9.5 Emission criteria for EGC
The emissions to air of SO
2
can be monitored as the SO
2
to CO
2
ratio from which corresponding
fuel-sulphur content can be obtained. Since there may be capture of CO
2
in the scrubber, or
release of CO
2
from the chemical reactions involved in the buffering, the CO
2
concentration
should be measured before the scrubber unit. All sulphur compounds emitted to air should be
measured and aggregated care should be taken to ensure that the more highly oxidised sulphur
compounds are accounted for. The SO
2
to CO
2
ratios corresponding to certain fuel-sulphur levels
have been calculated by IMO (see Table 9.2). No other emissions to air are regulated in the IMO
guidelines for sulphur abatement technologies.
Table 9.2. SO2/CO2 ratio corresponding to certain fuel-sulphur levels
Fuel-sulphur content (wt %) SO
2
(ppm)/CO
2
(%) ratio in exhaust
4.50 195.0
3.50 151.7
1.50 65.0
1.00 43.3
0.50 21.7
0.10 4.3
9.6 Wash-water criteria
The IMO Wash-water criteria are stipulated in section 10 of Annex 5 to MARPOL 73/78. We refer
here to the preliminary guidelines dated July 16, 2009. The values that are to be monitored
should include pH, PAH, turbidity and temperature. There are also limits on the nitrate content in
AEA 64
the discharged wash-water. The text in this version of the guidelines implies that the wash-water
criteria apply everywhere. However, in earlier versions the criteria were only applied to harbours,
estuaries and ports. It is uncertain, at present, if separate limits will be developed for operation at
open sea.

pH
pH is a measure of the acidic or basic (alkaline) nature of a solution. The concentration of the
hydrogen ion [H
+
] in a solution determines the pH. Ocean water usually has an excellent buffering
system, with interaction of carbon dioxide and water, and pH generally varies between 7.5 and
8.5. Neutral water has a pH of 7 while acidic substances are less than 7 (down to 1, which is
highly acidic) and alkaline substances are more than 7 (up to 14, which is highly alkaline).
Anything either highly acid or alkaline would severely affect marine life but the oceans are usually
very stable with regards to pH. According to USEPA the criteria for pH for effluents emitted to
marine waters should be 6.5-8.5, but no more than 0.2 units outside of normally occurring range.
Alkalinity is a measure of the buffering capacity of a solution and molecules, e.g. carbonates and
bicarbonates, take up hydrogen from a solution.

In the IMO wash-water criteria regarding pH it is stated that the wash-water should comply with
one of the following:

The discharge wash-water should have a pH of no less than 6.5 measured at the ships
overboard discharge with the exception that during manoeuvring and transit, the
maximum difference between inlet and outlet of 2 pH units is allowed measured at the
ships inlet and overboard discharge.

During commissioning of the unit(s) after installation, the discharged wash-water plume
should be measured externally from the ship (at rest in harbour) and the discharge pH at
the ships overboard pH monitoring point will be recorded when the plume at 4 metres
from the discharge point equals or is above pH 6.5. The discharged pH to achieve a
minimum pH units of 6.5 will become the overboard pH discharge limit recorded.

PAH
Polycyclic aromatic hydrocarbons (PAH) are the largest known group of carcinogenic substances
and include many individual chemical substances containing two or more condensed aromatic
rings. The biological effect of PAH is linked to the planar structure of the molecule and its ability to
affect DNA in the cell nucleus. PAH are formed when coal, oil, or other hydrocarbons are heated
under anoxic conditions, i.e. with insufficient oxygen, and are the products of incomplete
combustion. PAH are not usually present as individual compounds but occur in mixtures. PAH
are fat soluble and might be bio-accumulative (in fatty tissues). In the aquatic environments, PAH
are usually bound to particles which transport to sediment. Many PAH compounds accumulate
in invertebrate organisms. Fish eggs and fry exposed to PAH have been found to suffer
mutation, bleeding, heart conditions, reduced growth and increased mortality. A group of 16
PAHs are usually measured and analysed, but in the IMO wash-water criteria, PAH
phe
or
phenanthrene equivalence is used. Phenanthrene is a member of the PAH group and is
insoluble in water. PAH is sometimes used as an indicator of the total emissions of
hydrocarbons.

The IMO wash-water criteria regarding PAH:

The maximum PAH concentration in the wash-water should not be greater than 50 g/l
PAH
phe
above the inlet water PAH concentration. The PAH concentration should be
measured downstream of the water treatment equipment.

AEA 65
The 50 g/l limit is normalized for a wash water flow rate through the EGC unit of 45
tonne/MWh. The limit has to be adjusted upward for lower wash water flow rates per
MWh to get a limit of 2.25 mg/kWh of PAH
phe
.

For a 15-minute period in any 12-hour period, the PAH concentration may exceed the
limit described above by up to 100%.

Turbidity
Turbidity is a measure of the amount of particles in the water, or rather the cloudiness caused by
suspended solids. Turbidity in open water can be caused by several different reasons, e.g.
phytoplankton or high levels of sediment in the water. In lakes and shallow areas the turbidity can
decrease the amount of light reaching the lower depths and thereby affect submerged plants and
in the end also species that are dependent on the amount of plants. Turbidity is measured in FTU
(Formazin Turbidity Unit) or FNU (Formazin Nephelometric Units).

The IMO wash-water criteria regarding turbidity:

The maximum continuous turbidity in wash-water should not be greater than 25 FNU
or 25 NTU above the inlet water turbidity.

Nitrates
Nitrogen oxides, i.e. the sum of NO and NO
2
,

are produced during combustion at high
temperatures. The oxides can contribute to acid rain and other environmental effects. Nitrate
(NO
3
-
) is the most highly oxidised form of nitrogen and excess nitrate concentrations in aquatic
systems can lead to algae blooms and eutrophication.

The nutrient concentration in seawater usually decreases during springtime and all processes in
the nitrogen cycle are seasonally dependent. When oxygen is present, ammonia can be oxidised
to nitrate (via nitrite) in a process called nitrification.

The IMO Wash water criteria regarding nitrates:

Nitrates should be monitored. The wash-water treatment system should prevent the
discharge of nitrates beyond that associated with a 12% removal of NO
x
from the
exhaust, or beyond 60 mg/l normalized for wash-water discharge rate of 45
tonnes/MWh whichever is greater.
9.7 Environmental assessments
Behrends and Liebezeit (2003) presented an environmental impact analysis of sea water
scrubbers. Laboratory tests with mixtures of seawater and sulphuric acid solution at pH 4, to
mimic scrubber wash-water, were performed where the resulting pH was studied. A mixture of
1:10 in seawater was found to give a pH increase of around 0.1. This is compared to the EPA
limit of 0.2 that is considered safe. For brackish water a higher change in pH was observed
initially. In another report by Hufnagl et al. (2005) sampling was carried out of scrubber water and
water from ports where a trial with a seawater scrubber onboard the Pride of Kent was
undertaken. A decrease in pH of maximum 2 in the effluent was observed together with
increased nitrate content. No impacts on seawater or increased toxicity were reported.

Buhaug et al. (2006) discussed wash-water criteria for scrubbers and performed an
environmental analysis. They suggested using a PNEC (Predicted No Effect Concentration) value
for pH of 0.2. A set of criteria are suggested for three different levels of safety. The criteria are for
THC and a set of metals. They also propose that no limits are required for pH and COD. The
criteria are given in mass per volume of effluent water under the assumption that 45 kg
water/kWh is used in the scrubber. They also suggest that the PNEC value for total hydrocarbons
(THC) is set to 40 g/l. Values recalculated to mass per engine output are given in Table 9.3.
AEA 66
Buhaug et al. use the concept of near field, outside which no effect concentration should be
reached. This is set to 50 m from the stern for a moving ship and to 20 m for quay side
manoeuvring.
Table 9.3. Wash-water criteria after Buhaug et al. in mg/kWh
Component safety level 1 safety level 2 safety level 3 PNEC
THC 200 20 2 1.8
Cu - 1.8 0.2 0.13
Ni - 1.3 0.1 0.09
V - - 6.6 4.4
Pb - - 0.5 0.37

Karle and Turner (2007) analysed the amount of water needed for scrubbing of SO
2
as well as
the dilution of this water required for reaching a pH of 6.5 for different water types and
temperatures. They showed that the amount of water needed to capture the SO
2
is significantly
higher in river water than at open sea and with brackish water in between. They also showed that
the wash-water with a pH of 6.5 will quickly be neutralised to reach a delta pH of 0.2 and that both
the capture of SO
2
and the buffering (i.e. the dilution needed) are temperature dependent
processes.

GESAMP have commented on the wash water criteria in MEPC 59/4/19. No objections were
raised to the choice of parameters to be monitored but they recommended that dissolved oxygen
should be measured in addition. GESAMP also recommended that a generalised marine
environmental risk assessment should be developed in the future when it is possible that many
ships use scrubbers. According to GESAMP, the wash-water quality criteria appear to
adequately reflect the hazards posed by wash-water discharge to the local marine environment.
pH reflects the acidifying properties of dissolved SO
x
and its potential for acute effects on biota.
PAH represents any oily discharge, NO
x
reflects any tendency towards eutrophication and
turbidity is intended to safeguard against particulate matter (heavy metals and ash). GESAMP
stated that an environmental benefit in reducing SO
x
emissions to the atmosphere only can be
achieved if the discharge do not create further risks to the aquatic environment. One remark by
GESAMP was that IMO should consider the potential contribution to ocean acidification of the
large scale application of SO
2
capture from ships.

The reference value for a safe change in pH used in the studies is 0.2 which originally comes
from EPA. It has been found that wash water emissions from scrubbers is quickly diluted and
buffered to reach this level, both in measurements and modelling. No clear data is given for how
large zone around a ship is needed to reach this level (in ports or at sea) although the amount of
dilution needed can be obtained from Karle and Turner. The PAH criteria is intended as a
measure of oil and other hydrocarbons emitted with the wash water - this is the parameter
where Buhaug et al. obtained emission values that are high and close to the PNEC; it is not clear
from these reports how a PAH
phe
concentration is related to the emission of THC. Hufnagl et al.
report that phenanthrene is the dominating PAH in the wash water and thus the best candidate
for analysis. Turbidity is not discussed except that GESAMP regards it as a way to monitor
particulate matter. Elevated nitrate levels were observed by Hufnagl et al. and they actually note
that up to 20% of the NO
x
may be captured and released as nitrates. In addition GESAMP
suggests that COD should be monitored and Buhaug et al. suggest that emissions of Ni, V, Cu,
Pb and THC should be regulated.

9.8 Environmental impact related to the IMO wash water criteria
The potential impact of aqueous discharges to the marine environment require information on
both the potential toxic effects on representative marine organisms to different concentration
levels of the discharged water in sea water, and the sensitivity of the marine environment in the
region of concern. This should, furthermore, be related to the discharge conditions and the
AEA 67
resulting rate of dilution and spatial extent of the plume formed by the discharge after disposal to
the sea. There is a risk for locally high concentrations of a number of harmful substances if there
are no adequate wash-water criteria in force.

pH
Hufnagel et al. have analyzed the environmental impact of a seawater scrubber for two harbours,
Calais and Dover. They took samples from the seawater scrubber system and the samples
partially showed values lower than pH 3. The buffering system increased the pH. Experiments
showed that a mixture containing 40 % of seawater acidified to a pH of 4 (comparable to what
happens in a seawater scrubber) only changed pH about 0.2 units. The lowest pH measured in
the overboard discharge was 6.2. The US EPA, as mentioned earlier, has passed a guideline
concerning the introduction of acids which states that outside the initial mixing zone the pH
change is not allowed to be more than 0.2 units. Under the circumstances stated, the pH would
not reach a value that is harmful to the marine environment. This is dependent on the dilution,
volume of water and the specific area of concern. Buhaug et al. (2006) stated that discharges
from a single ship are rapidly diluted when the ship is moving, but expected to be much slower
when the ship is not moving.

Karle and Turner calculated the amount of water needed for scrubbing SO
2
and dilution to first
obtain pH of 6.5 and then a delta pH of 0.2. In addition they looked at further dilution to reach a
maximum of 1% reduction in oxygen concentration. They considered different water types and
temperatures for a 12 MW engine using a fuel with 3% sulphur. Adding all these factors gives an
indication of the amount of seawater affected by a scrubber. Table 9.4 gives an indication of
these amounts when scrubbing to a SO
2
emission corresponding to 0.1% fuel-S. The nominal
flow rate of scrubbing water used by MEPC is around 45 tonne/MWh giving about 540 tonne per
hour for a 12 MW ship. This should primarily be compared with the water needed for scrubbing in
Table 9.4
Table 9.4 Approximate amount of seawater affected by a scrubber for a 12 MW engine using 3% S-
fuel and with a seawater temperature of 10C for scrubbing to a SO2 emission corresponding to
0.1%S fuel. After Karle and Turner (2007).
Open sea Baltic Proper River freshwater
Water for scrubbing 700 900 2500 tonne/h
Water for dilution to pH
6.5
1400 1700 15000 tonne/h
Factor for dilution to delta
pH 0.2
3 3.5 2.5
Sum 6300 9100 44000 tonne/h

The results in table 9.4 show that almost 7 times more water for dilution is needed when
scrubbing takes place in fresh water, compared to open sea.

As an example we can look at the port of Hamburg. In 2000 115.5 ktonne of CO
2
and 1.9 ktonne
of SO
2
were emitted corresponding to a fuel consumption of about 36.3 ktonne with a sulphur
content of 2.6% (Whall et al. 2002). If all of these ships were to use open scrubbers around 21
tonnes/s of water would be needed for the scrubbing and buffering, this can be compared with
the flow in the Elbe, which is around 720 tonne/s.

PAH
The criteria for hydrocarbon emissions with the effluent water is given as PAH in phenanthrene
equivalents. The reason seems to be that measuring PAH is a means to monitor hydrocarbons
and phenanthrene was found to be the most abundant PAH in the analysis by Hufnagl et al.
(2005). However, the concept "phenanthrene equivalents" is not well established and needs to be
explained or replaced. Assuming that it simply means that the phenanthrene concentration is
measured, the criteria correspond to about 2.2 mg/kWh of PAH
phe
. Cooper et al. (1996) analysed
AEA 68
PAH in the exhaust from two ships and found phenanthrene levels in the gas to be 71 and 250
g/kWh respectively. The lower value refers to an engine using HFO. Thus, the wash-water
criterion seems to be set rather high - considering that the scrubber water should be treated
before the samples for the PAH analysis are taken. A 12 MW engine could, with the suggested
limit, emit around 26 g phenanthrene per hour and in addition unknown amounts of other PAHs.

According to Buhaug et al. (2006) OSPAR have developed Ecotoxicological Assessment Criteria
(EAC) for a large range of chemicals, including PAHs. This was done to make the evaluation of
monitoring data on toxic substances in the marine environment easier. The EAC can be applied
as PNEC (Predicted No Effect Concentration) values. The EAC-water for phenanthrene is
proposed to 0.5 5 (g/l) (provisional value). The PNEC value is the maximum permissible
concentration that can be sustained over time. The Predicted Environmental Concentration (PEC)
equals the proposed wash-water criteria multiplied by the dilution (if inlet concentration is
assumed to be zero). Since the wash-water criterion is set to 50 g/l PAH
phe
(for a scrubber water
flow of 45 l/kWh) the wash-water must be diluted 10 100 times to reach this PNEC
concentration. Furthermore, the background concentration should also be considered, which may
lead to further limits on the emissions. In Directive 2008/105/EC EQS (Environmental Quality
Standards) have been set for priority substances. The EQS are not applicable for PAH as a
group but for identified substances e.g. for Benzo(a)pyrene, the sum of Benzo(b)fluoranthene
and Benzo(k)fluoranthene, and the the sum of Benzo(g,h,i)perylene and Indeno(1,2,3-cd)pyrene;
the Directive does not include EQS for phenanthrene.

It is still to be established, furthermore, that the PAH criteria used will ensure that the levels of
total hydrocarbon compound (THC) emissions are at safe levels. The criteria for THC developed
by Buhaug et al. are 4500, 450 and 45 g/l for the three "Safety Margins" that they use. This is
under the assumption of a wash-water flow of 44 tonne/MWh. These levels would correspond to
200, 20 and 2 mg/kWh. Thus the IMO criteria which applies only to phenanthrene is set to about
the same as the THC value by Buhaug et al. for safety margin 3.

If we look at the port of Hamburg again, the amount of PAH
phe
emitted were all ships to use
scrubbers and allowing the scrubber water to be at the criteria limit would be around 400 kg
phenanthrene per year.

Turbidity
Turbidity is chosen as a means to measure the suspended solids in the water and because it a
simple method for on-line analysis. Turbidity can be a useful indicator of the status of the
scrubber, i.e. that suspended particles are not being formed in the scrubbing process or the wash
water treatment. However, turbidity is not a direct method of determining the number of exhaust
particles that end up in the wash water - there is no direct correlation between turbidity and
particle concentration. Further, turbidity values are strongly dependent on the size of particles in
the water and smaller particles are likely to have significantly less influence the measured
turbidity than larger ones. Studies are required to correlate the turbidity measurements with
concentrations of particulate matter, ash, metals etc in the wash water before this criteria can be
considered an indication of particulate levels in the wash-water. These uncertainties make it
hard to make an environmental assessment of the turbidity criteria.

Nitrates
The criteria for nitrates are motivated by the increased contribution to eutrophication that will be
the result of increased emissions of nitrates with the effluent water. Since scrubbers are likely to
capture the NO
2
in the exhaust the 12% maximum capture requirement of NO
x
will normally be
met. Hufnagl et al. (2005), however, report a capture fraction of up to 20%. To analyse what the
12 % criteria could mean it is of interest to look at the Baltic Sea, which is heavily trafficked, it is a
SECA area and it is highly affected by eutrophication and large efforts are made to decrease the
flow of nutrients into the Baltic. The total emission of NO
x
from shipping in the Baltic according to
a recent Helcom study (Jalkanen 2009) is around 400 ktonne per year at present. EMEP
AEA 69
modelling (Benedictow 2009) has calculated how much of the emissions from a certain area is
deposited in another area, from this work it can be concluded that about 13% of the NO
x
emitted
from shipping in the Baltic Sea is deposited in the Baltic Sea - the rest is deposited over land or
(to a smaller extent) in other sea regions. Consequently ~ 107 ktonne per year is deposited in the
Baltic Sea from shipping in the Baltic Sea calculated as nitrates. If all of these ships were to use
scrubbers and all of these scrubbers captured 12% of the NO
x
in the exhaust and released this as
nitrates in the scrubber water the total deposition would increase to 193 ktonne, i.e., an increase
of about 86 ktonne per year. This is, of course, an unrealistically high number considering that not
all ships will have scrubbers and that the capture fraction will, in practice, be lower. Further, a
certain fraction of the nitrates deposited over land would return to the Baltic Sea with rivers etc.; it
does give, however, a maximum possible contribution increase of nitrates. Further, of the total
load of nitrogen to the Baltic Sea 25% comes from deposition and the rest via rivers and direct
waterborne discharges. Thus the hypothetical maximum increase in load of nitrates, if all ships
were to use scrubber, would be about 2%.

COD, the chemical oxygen demand, is a measure of the theoretical oxygen consumption of a
water sample, i.e. the amount of oxygen needed to break down the organic matter in it.
According to Buhaug et al. there are no environmental standards for COD. Tolerance to reduced
dissolved oxygen varies between organisms, life stages and environmental conditions. While
Buhaug et al. concluded that COD was not a matter of concern under the circumstances of their
study. Karle and Turner used a criterion that the COD should be less than 1% of the available
free oxygen, which they found to be met when their criteria for pH were met.
9.9 Monitoring
In directive 2005/33/EC, article 4c it is stated that ships using abatement technology to achieve
low SO
2
emissions must be fitted with continuous monitoring equipment for SO
2
emissions in
order to establish that the emissions at all times are lower or equal to that which would have been
obtained by using fuel with the stipulated sulphur content.

In the "revised guidelines for exhaust gas cleaning systems" from MEPC dated 16 July 2009
there are two permissible schemes, Scheme A (unit certification with parameter and emission
checks) and Scheme B (continuous monitoring with parameter checks).

In Scheme A the unit should be certified to meet the limit values for SO
2
/CO
2
concentration in the
exhaust. The fuel that may be used, including sulphur level, must be specified together with other
operational parameters such as gas flow, water flow, and salinity of sea water or fresh water
elements necessary to provide adequate neutralisation agents. A SO
X
emission compliance
certification must be obtained from the administration. There must also be a certification
procedure established for use during surveys. The system must automatically record that the
abatement equipment is in use by monitoring wash water pressure and flow rate at a frequency of
at least 0.0035 Hz. Further, daily spot checks of the SO
2
/CO
2
ratio are "recommended". If the
manufacturer is unable to provide assurance that the unit will meet the certified value this way,
Scheme B is recommended.

In Scheme B compliance is demonstrated by continuous monitoring. Thus SO
2
and CO
2

concentrations are to be measured after the abatement equipment with a frequency of at least
0.0035 Hz. In this scheme parameters such as exhaust gas flow rate and water flow rate are to
be checked daily. There should be a technical manual where it is stated, for example, the
minimum and maximum water flow rate and minimum inlet water alkalinity and other parameters
for the operation of the unit. The manual should be approved by the administration.

The procedure for emission testing and data recording is described and the same for both
schemes. The recording equipment must be tamper proof. For both schemes the ship must have
a SO
X
emission compliance plan showing how the equipment should be used to comply and how
this is demonstrated.

AEA 70
In addition, both schemes will require continuous monitoring of the wash-water criteria discussed
above, i.e., pH, PAH concentration, turbidity and temperature.

Thus, in Scheme A the function of the abatement equipment is certified by continuously recording
operational parameters such as water flow and pressure as well as wash-water parameters. In
Scheme B the compliance is verified through direct measurements of the SO
2
emission together
with the wash-water parameters. Only Scheme B complies with the wording in Directive
2005/33/EC.

The advantage with continuous monitoring of the SO
2
/CO
2
ratio is that compliance with
regulations is directly monitored at all times. There are also some disadvantages: continuous
monitoring of exhaust from engines can be expensive and technically problematic, this is
discussed by CIMAC in a report from 2005 and by Johnsen et al, the measurement systems may
also be affected by the high temperatures and the high particle load in the exhaust. Further,
vibrations, variation in ambient temperature, spray of liquids and fluctuation in voltage may
influence the measurements and the monitoring systems may then require much maintenance
and calibration in order to function properly.

Continuous monitoring can be of great value especially if the conditions that may influence the
performance of the abatement equipment are likely to change. For sea water scrubbers this may
be when moving in areas with seawater of varying alkalinity. In waters with low alkalinity more
seawater is required to meet the required SO
2
emissions and further, more dilution will be
required to meet the wash water limit for pH.

For closed loop scrubbers the function is not dependent on the characteristics of the surrounding
seawater since the operation of the equipment involves fresh water and an added neutralising
agent.

The requirements in the MEPC guidelines make it possible to use Scheme A for closed
scrubbers, it may also be suitable for some sea water scrubbers. During certification the
feasibility of using Scheme A should also be investigated, i.e., by showing that the monitoring of
the system parameters gives evidence that the SO
2
emission and wash-water requirements are
met. There may well be cases where a scrubber is not certified, for example, for use in rivers or
brackish water systems and then the functionality can be expected to be more stable. If a system,
during certification, is found to have too low a capacity under certain conditions, such as
alkalinity, sea water temperature etc, it is possible to certify the unit using Scheme B. Since it is
possible that there will be a number of systems that can be monitored satisfactorily using Scheme
A it is recommended that the IMO regulations regarding monitoring are incorporated also in the
EU regulations.

With the use of either scheme it is of great importance that the SO
2
emissions and wash-water
criteria are controlled. This might be achieved via regular certification measurements required by
administrative provisions. Further surveillance of the SO
2
emissions can be made by authorities
using remote sensing of the SO
2
/CO
2
ratio in the exhaust plume from the land or from
airplanes, there is an ongoing program for the development of such systems at the JRC.
Mellqvist et al. (2008) have demonstrated such a method with surveillance from airplanes. The
uncertainty in such a measurement is expected to be around 10% and thus scrubber units that
are defective or turned off should be detectable. One problem may be that scrubbers can capture
some CO
2
and thus the SO
2
/CO
2
in the exhaust may show values that over-represent the actual
SO
2
emissions. This problem could be circumvented by listing the scrubber units' CO
2
capture
capacity during certification.
9.10 Conclusions and suggestions
The use of scrubbers to clean the exhaust from marine engines using residual oil could in
principle lead to high concentrations of a number of harmful compounds in the marine
environment around the ships. The use of scrubbers in combination with high sulphur residual oil
AEA 71
may be an economically attractive option in SECA areas and also worldwide from 2020 and so,
as a consequence, a large number of ships would use scrubbers; consequently it is of great
importance that measures are taken to assure that the marine environment is protected. There
should be mechanisms to prevent the use of scrubbers in certain areas if there is a risk to marine
ecosystems. The technology allows for the use of closed systems - making it possible to limit the
wash-water discharged in areas with limited water circulation or especially sensitive
environmental conditions. Scrubbing systems can be used to abate particulate emissions. The
present situation is somewhat incongruous in the sense that wash-water limits for scrubbers have
been set up while there are only limits for the gaseous emissions of SO
2
and NO
x
a situation
that does not reflect current concerns related to the health impacts of PM. It is recommended
that criteria are developed that promote the abatement of PM with, as a minimum, limits set to
allow either voluntary or market based reductions of PM emissions; it is suggested, furthermore,
that where/when possible, EU regulations are kept in line with the IMO guidelines.

The wash-water criteria include limits for PAH, pH, turbidity and nitrates. Of these all except the
nitrate criterion are intended for continuous monitoring onboard the ship. So, a major factor to be
considered when formulating the criteria is that they should be suitable for online monitoring.
However, it would also be possible to add additional criteria that are to be tested during
certification. This could give further assurance that the units are environmentally safe and would
not have to be constrained by the need to measure online. The wash-water criteria apply
everywhere but it is possible that additional, separate, criteria can be developed for open sea.

With the present criteria for pH it would appear that the dilution in the sea and natural buffering
will rapidly give acceptable levels - even in confined areas. It is recommended, however, that
further studies are undertaken for brackish and river water and that the criteria are then reviewed
accordingly. It is suggested that the IMO guidelines are adopted for the pH criteria while, at the
same time, allowing for revision as more data becomes available. If the criteria for open sea were
relaxed then more acidic scrubber water may be emitted to the sea - however, the same volume
of water would be affected because less water would be used for dilution within the ship.

The PAH criteria in the IMO guidelines are expressed as phenanthrene equivalents. This is not a
well established concept and should be further defined in the guidelines. The criteria correspond
to an emission of phenanthrene of 2.2 mg/kWh which could, in principle, lead to large emissions
of PAH from marine scrubber units. The criterion was probably chosen in order to be easy to
measure with continuous monitoring. While it may be appropriate to use phenanthrene for
monitoring purposes the criteria ought, nevertheless, to ensure a low discharge of a range of
PAHs so how the measured value relate to the PAH 16, for example, or its correlation to THC
needs to be established during certification; this correlation may differ for different scrubber
systems and fuels. It is suggested that the PAH criteria that are used are in line with Directive
2008/105/EC. The criteria for THC suggested by Buhaug et al. could be used for commissioning.
It is yet to be established that PAH
phe
is an adequate measure of the concentration of oil or THC
in the wash-water.

The criteria for turbidity may be suitable for establishing that unwanted solid material is not
formed in the scrubber unit but it is unlikely to be appropriate as a measure of the proportion of
particulate matter from the exhaust gas that captured and discharged with the effluent water.
There is no simple relation between particle concentration and turbidity and the turbidity measure
is strongly dependent on particle size. It is suggested that: the concentrations of particulate
matter, metals, and PAH are regulated and tested during commissioning, and that the correlation
to turbidity is investigated in order to allow continuous measurement onboard. Pending the
delivery of data from on-going trials it is suggested that the turbidity criteria from IMO are used,
alternatively the criteria suggested by Buhaug et al. could be applied.

The criteria for nitrates could in principle lead to large additional discharges of nitrates to sensitive
areas. The best way to deal with this issue would be to use exhaust gas NO
x
abatement
technology. Taking the Baltic example described above, an engine with an emission factor of 17
AEA 72
g NO
x
/kWh should reduce its emission to about 9.4 g/kWh (if 12% of the NO
x
is emitted as
nitrates with the scrubber water) in order not to increase the nitrate load when using a scrubber. It
is recommended that provisions are made when setting criteria for nitrate emissions so that
increased loads can be prevented in sensitive areas.

Since there is limited data on the performance of marine scrubbers for SO
2
, it is important to
follow on-going trials in order to ensure that thorough environmental impact assessments are
conducted when evaluating and developing wash-water criteria. A larger introduction of scrubbers
could lead, potentially, to high concentrations of e.g. PAH if the units actually emit PAH in
concentrations close to the criterion limit. There is also the opportunity of studying the impact of
inert gas scrubbers on the marine environment; these are frequently used though smaller than
exhaust gas scrubbers made to comply with sulphur emission regulations. It is also important to
obtain more data on the properties and amounts of the sludge produced. This will be needed in
order to develop facilities for handling of the sludge by ports.

It seems appropriate to adopt the IMO guidelines regarding monitoring, i.e., allowing for the
possibility of using alternatives to continuous monitoring of the SO
2
emissions on the basis that
there will be units that can be adequately monitored by other means.

AEA 73
10 Use of distillate fuels by ships at berth
10.1 Introduction
From 1 January 2010 ships using ports within the EU will be required to use low-sulphur fuel,
maximum 0.1 wt% sulphur, while at berth, with an exception for those making only short stops
(Directive 2005/33/EC). The regulations apply also to ships that are anchored. The fuel switch
should be done as soon as possible after arrival and as late as possible before departure. This
regulation applies to all engines and to all boilers, including auxiliary boilers.

These regulations will most likely make it necessary to use distillate fuel while at berth. There are
concerns that there are risks associated with switching from residual oil to distillate fuel in boilers.

This report has been drafted under the project Service contract for a Cost Benefit Analysis to
support the Impact Assessment accompanying the revision of Directive 1999/32/EC on the sulphur
content of certain liquid fuels, which was undertaken for the Commission by a team including
TNO, IVL and AEA under an AEA-led framework contract. It tries to clarify the risks associated
with the problem outlined above and aims to address the following questions:

Identify the type of ships where these risks may be relevant.
Estimate the number of such ships in 2010 and 2020.
Describe the nature of these risks.
Indicate technical solutions and assess the costs of these.

The study has been undertaken through literature searches and searches on the internet as well
as through interviewing representatives of industry and relevant authorities. We have had contact
with manufacturers of boilers and accessories, representatives of ship-owners, port authorities,
and maritime administrations. There are only a few reports that address the problem.
10.2 Outline of the problem
Boilers that are constructed for the use of Heavy Fuel Oil (HFO) cannot, in general, be used with
Marine Gas Oil (MGO) without modifications. A switch from HFO to MGO in a boiler that is
designed for HFO use could lead to operational problems with potential flame failure and an
increased risk of a boiler explosion. To avoid these problems, boilers may have to be modified
before a switch from HFO to MGO can take place in a safe way. The modifications needed must
be assessed for each boiler individually. The exact number of boilers that will need modification,
as well as the costs and time needed for these modifications, are difficult to assess because the
figures are not well known by the ship-owners.

Boilers are typically used at berth and tankers, in particular, use large boilers for heating the
cargo and for steam-driven cargo pumps.

A number of LNG-carriers use boilers for propulsion where boil-off cargo is used as fuel in
combination with HFO. These boilers, which are also used at berth for cargo transfer, need to be
modified before they can be used with MGO.

In order to comply with Directive 2005/33/EC, ship-owners must assess what boilers need to be
modified, to what extent, and have these modifications done before the end of 2009. Some ship-
owners have made the assessments and taken steps to do the necessary modifications but many
ships still need to be investigated. The process seems to have been started during 2008 and, in
most cases, activity intensified in 2009. The reason that the industry has not acted earlier seems
to be multifold. Some ship-owners anticipated that the EU directive would be brought in line with
the IMO directives of Marpol Annex 6 and that the 0.1 sulphur-limit would be implemented in 2015
- as it will be for SECA areas. It was also expected that the sulphur directive would be reviewed
AEA 74
before 2010, and some ship-owners seem to have anticipated, until relatively recently, that HFO
with 0.1% sulphur would be made available.
10.3 Types of ships where the risks may be relevant
Boilers are used on ships either for propulsion or as auxiliary boilers. Boilers for propulsion are
used in combination with steam turbines and, although once abundant, they are today relatively
uncommon because diesel engines have a higher thermal efficiency and are more fuel efficient.

An application where boilers are used for propulsion is for ships that transport LNG, the reason
being that some of the cargo boils off during transport and using it as fuel for the ship recovers a
part of the cargo that would otherwise be lost. For reasons of thermal efficiency, these boilers
operate at high pressure ~60 bar. Some newer ships instead use reliquefaction plants to collect
the boil-off gas. Boil-off gas (BOG) is sometimes used in combination with a liquid fuel and,
depending on power demand and operational mode, the final fuel mix can be either 100% BOG
or a combination of oil and BOG. The reasons for the dual fuel arrangements lie in safety and the
need for reliability for propulsion. For example, when manoeuvring - or for other purposes that
demand a rapid response in boiler power, the gaseous fuel alone may not be sufficient. In such
situations, dual fuel mode is used where oil and BOG is burnt simultaneously. A change in power
can then be taken up more slowly by using gas. Further, in cases when the gas supply to the
boilers is unintentionally reduced or interrupted, the system can use a fuel-oil boost up to ensure
continuity in power. There is also a limited turn down ratio for gas burners and so oil is used at
low loads too. The boiler powered LNG ships used today are designed for a combination of BOG
and HFO. Typically, a ship that would use 150 tonnes HFO per day if operated on HFO alone
uses around 10 tonnes of HFO in dual fuel mode. The boilers use small amounts of MGO during
flashing from cold when furnace temperatures are low (normally after refit) but cannot use MGO
during normal operation without modifications (see below).

Auxiliary boilers may be divided into:

1) small boilers used on ships to produce steam that can be used to heat residual oil,
to produce hot water and for the purpose of heating; and
2) larger boilers used on tankers for heating cargo and to drive steam turbine pumps.

Auxiliary boilers are used on practically all ocean going ships. These boilers can be designed to
be used with either HFO or distillate fuel. Some boilers can use both HFO and MGO. Typically,
boilers are used with HFO on ships that use this fuel for propulsion. The type of fuel used in
auxiliary boilers has not been investigated in detail since the fuel consumption is normally
assumed to be a minor part compared to main and auxiliary engines; it is difficult, therefore, to
establish the exact distribution of boilers that use only MGO, only HFO or a combination. The
vast majority of boilers, however, use HFO and this is certainly true for the larger auxiliary boilers
on tankers.
10.4 Relevant number of ships
The need for an assessment of the potential modifications to existing boilers to manage the
switch from HFO to MGO applies, in principle, to all ships by 2010. It is today not clear how many
of these will need modifications. Work by the leading manufacturer of boilers, Aalborg industries
(Aalborg 2009) and a study by the California Air Resource board (CARB 2008) indicates that
most boilers can at least temporarily use MGO without problems. However, both studies, as well
as a report by DNV (DNV 2005), state that MGO should not be used in a boiler designed for HFO
before it has been inspected. Further, a large number of boilers will need modifications before
they can be used safely with MGO. Although a number of ship-owners have assessed and
when necessarymodified their boilers, it seems unlikely that all boilers will be able to run on
MGO by the 1st of January 2010.

According to OCIMF at least 2 368 of 3 476 chemical and oil tankers that enter EU ports use
steam for cargo pumps or cargo heating. Other ships may use diesel engines or electrical motors
AEA 75
to drive the pumps, and some use electrical heating. These boilers are more likely to need
modifications since they use relatively large amounts of fuel, which affects valves and fuel pumps.
A number of tank owners will have difficulty making these modifications before the end of the
year, although the exact number is difficult to say. Modifications on all ships will, in all likelihood,
not be complete until late 2010.

The boilers on tankers are often used during the discharge of the cargo. Intertanko estimated
that in 2006 there were 8 000 - 10 000 port calls of tankers in the EU. With a fuel use per
discharge of 50 - 90 tonnes, this amounts in the EU to a total fuel use of around 400 - 500
ktonnes per year.

There are 262 LNG ships with steam turbine propulsion, of which most service EU ports.
Thirty two ships use diesel propulsion while 19 are dual fuel diesel electric. A number of LNG
ships are several decades old but the production of LNG tankers has increased a lot recently.
About 43% of the LNG ships were delivered between 2005 and 2008. The trend for new ships
seems to be that diesel engines in combination with reliquefaction plants are preferred to steam
propulsion. There is also a development towards dual fuel diesel engines that can burn fuel oil
(distillate or HFO) and gas. LNG tankers are refitted about every 2.5 years - or every 5 years for
newer ships. If modifications needed to comply with the sulphur directive are carried out during
refit it will be approximately 2014 before all ships are modified. The ships could also be taken out
of service for the modifications, which could be complete within approximately one - two years.
The latter alternative then implies extra costs when the ships are out of service for modifications.

There are approximately 690 deliveries by LNG tankers per year to the EU. At each port visit,
around 50 tonnes of HFO is used adding up to about 35 ktonnes per year.

By 2020 all ships will have had the time needed to make the necessary adjustments on their
boilers.
10.5 The nature of the risks
There have been reports of boiler explosions due to defects or improper operations. This can
occur if, for some reason, the flame goes out, a high pressure of fuel gas builds up in the burner,
and the control system is malfunctioning or disconnected. In such a case, the furnace must be
purged before ignition. Failure to do this may cause an explosion, as for example reported by the
British Department of Transport (1983). There are flame monitoring and control systems that will
automatically purge the furnace so that during normal operation the risks for explosions are small.

There are concerns that these risks may be higher when switching over from residual oil to
distillate fuel. This is noted by the IMO subcommittee on bulk liquids (2007), which states: "The
routine switching between marine distillate and Heavy Fuel Oil (HFO) has safety implications for
existing boilers. The boiler combustion systems may require to be modified to prevent the risk of
boiler furnace explosions in distillate operation". The risks are also noted in a recent US Coast
Guard maritime safety alert (2009) in relation to regulations for California which apply from 1 July
2009. These regulations stipulate that Marine Gas Oil (MGO) or Marine Distillate Oil (MDO) must
be used in all engines including boilers (with the exception of boilers for propulsion). The notice
points out that "proper procedures, training, and maintenance are essential for vessels to safely
switch between heavy/intermediate fuel oils and marine distillates". The notice is about all
engines and auxiliary boilers and suggests that each system should be assessed for
modifications and that training and routines should be established.

The reason for the special concerns regarding a fuel switch from HFO to distillates is that the
pipes and other parts are heated when using residual oil. When the distillate enters the furnace
there are concerns that it will fail to ignite, causing a build-up of gases that in turn is an explosion
risk. When distillate fuel is present in hot piping vapour locks may cause irregular fuel flows to the
burners (this may also be an issue for diesel engines). This irregular fuel and vapour flows/bursts
towards burners can cause flame extinction; the lighter the fuel, the easier the evaporation and
AEA 76
the larger the risk for an air/fuel mix which is potentially unsafe. There are concerns that the crew
on many ships may not be familiar with managing these situations safely.

The boilers normally run on MGO during start up; however, at this point the system has not yet
been heated. After the boiler pressure is up, the pipes can be emptied of MGO and then heated
and used for HFO. MGO is thus not introduced to the boiler through heated pipes at this stage.

In normal ship installations, "heat tracing" of pipes (by steam or electric heating) can be turned on
or off. In this kind of installation, proper operation may be sufficient to avoid the risk of an
explosion. In many cases, fuel switching is automatically managed by the ships hardware and
software (and there is no difference between fuel flows to engines and boilers in this respect) -
this may be the situation in modern ships that have been delivered during the recent years.
When fuel switching cannot be done automatically, however, the ships officers must follow very
precise operational instructions.

Besides concerns about explosions inside the boilers, there are other issues that need to be
addressed before a boiler designed for HFO can be used with MGO. These, reported by DNV
(DNV 2005) and Aalborg industries (Aalborg 2009), include:

Fuel pumps and valves may need to be replaced due to the different viscosity of the
fuels. There is an increased risk of wear and pump breakdown if the oil has a higher
viscosity and lower lubricity than what the pump is designed for.
There may be increased smoke emissions because the amount of fuel injected in the
burner is set for HFO which has a higher density and lower calorific value than MGO.
On certain types of burners (rotary cup) coke deposits may build up when using distillate
fuel. In steam atomising burners the MGO may be heated - causing vaporisation.

Other concerns raised, by the shipping industry, about the regulations from 1 January 2010
related to:

1) problems with the supply of fuel with 0.1% sulphur for ships at berth;
2) issues with ships not having enough fuel tanks to store the different kinds of fuels
that they will need; and
3) concerns that there is not enough time to modify existing boilers and train crews.

The objective of the present study is not to investigate possible problems with the supply of MGO.
However, according to a study by Entec (2002) c. 3 000 ktonnes of fuel was used for
manoeuvring and in port in the EU in the year 2000. The main portion of this fuel will likely have
to be MGO from 2010 and, considering the increase in shipping, a rough estimate is that around
4 000 ktonnes of MGO is needed.

For the LNG ships with steam propulsion, the issues are essentially similar to those described
above - with some additional concerns. Since the boilers are used for propulsion, it is essential
that the fuel supply is not interrupted. As described above, there must be a reliable supply of fuel
oil in addition to the BOG. The fuel oil systems in these ships are designed for HFO. They should
not be used with distillate fuel for a number of reasons:

1) There would be an increased risk of failure in fuel pumps and valves.
2) The distillate fuel could be unintentionally heated in pumps or in the burner leading to
the mentioned problems with evaporation.
3) Most of the burners used have fuel injectors of the concentric type where the
atomising steam is in between the tubes for fuel oil where distillate fuel may be
heated. Other burners have parallel tubes for steam and fuel oil. Here the heating of
the distillate fuel is less of a concern but temperature gradients when cold fuel oil is
used could cause distortion of the tubings used for oil supply.
AEA 77
4) The burner management system and flame supervision/monitoring systems are
designed for HFO.
10.6 Technical solutions and costs
For the safety and operational reasons discussed above, all boilers designed for the use of HFO
should be inspected before they are used with distillate fuels. Many boilers may be used
temporarily with gas oil without modifications, but for others modifications may be needed on fuel
pumps, burners or pipes, as well as flame supervision and adjustments in air/fuel ratio, fuel flow
and post-purging sequence. These modifications are further described in Aalborg (2009) and
DNV (2005). Modifications will have to be approved by a Classification Society. The solutions for
LNG tankers with boiler propulsion are discussed separately at the end of this section.

The risk of a boiler explosion, as discussed above, is associated with the vaporisation of
distillate fuel in combination with flame failure and improper purging. To avoid the heating of MGO
in the piping system, heaters should be by-passed or the tracing of fuel pipes turned off. The fuel
system should have time to cool to around 60C before MGO is introduced. The pipe system may
have to be converted to contain two parallel feeds to the burners. The fuel may also be
unintentionally heated in fuel pumps that then have to be adjusted or replaced. There are
potential problems with gassing in steam atomising burners, which may have to be modified.
Protective heat shields should be present on rotary cup burners to avoid deposit build-up and
unintended ignition after flame failure. The flame supervision system may have to be modified to
include another flame scanner - due to the different properties of HFO and MGO flames. The
control system may have to be modified to implement automatic post-purging sequence. This will
ascertain whether a potentially explosive gas mixture in the furnace has been purged in a safe
way.

Fuel pumps may have to be modified or replaced to be able to tolerate MGO, with lower viscosity
and lubricating properties than HFO, for a longer period. In some cases, it may be sufficient to
modify the control system.

Fuel pressure may have to be adjusted and/or burner nozzles replaced To avoid smoking, the
air/fuel ratio may also have to be adjusted.

The costs associated with these measures will vary depending on the modifications needed. The
exact distribution of costs for the existing boilers is not known. Boiler suppliers estimate costs in
the range of 5 000 to 25 000 per boiler when modifications are needed. For tankers with large
boilers, costs around 150 000 per ship have been quoted. There will also be costs for approval
and certification.

For LNG ships with steam turbine propulsion, the boil-off from the cargo is used in combination
with HFO. For the reasons described above, these ships should be modified before they are used
with MGO.

The boiler management system (BMS) settings will have to be adjusted, new flame scanners may
have to be installed, and a separate fuel supply system for MGO will be needed. The majority of
burners will have to be modified to avoid vaporisation. Further, fuel change-over procedures
have to be investigated. Some ships may also need larger tanks for MGO in order to
accommodate enough fuel for one port stay.

These modifications and actions would make it possible to use MGO rather than HFO at berth
and thus comply with the directive. There are still some concerns that there may be an increased
number of flame failures and that the main boiler manufacturers lack experience in burning MGO
during normal operation. After the modifications, however, these incidents would be handled in a
safe way by the BMS, although there may be interruptions and operational problems. The costs
for these measures have been estimated by SIGTTO. The costs are difficult to estimate and
figures ranging from 70 000 to 1 400 000 per ship have been cited with a median of around 600
AEA 78
000 . To this should be added costs for taking the ships out of service if the modifications are
done outside a regular refit.

An alternative solution would be to use 100% BOG while at berth. There are, however, concerns
about the stability of the boiler power at low loads. It is unclear if this can be implemented on all
ships and modifications to the control system and BMS would be needed. Further, there are
situations where 100% BOG cannot be used. This applies for relighting after a flame failure and in
cases where there is no LNG cargo, as when ships enter/leave refit/dry dock. Therefore, this
alternative does not comply in full with the EU directive. No cost estimates are available for this
alternative because the extent of modifications needed has not yet been investigated. This
alternative may require substantial modifications and may be more expensive than the previous
alternative. The problems with stability and non-compliance make this alternative less attractive
for the industry. Also, some ship-owners do not see this as a solution to comply with the SECA
rules coming in 2015 because 100% BOG will be unsuitable during some operational modes.

A third possibility which seems more attractive to ship-owners is to minimise the use of HFO and
use as much BOG as possible. The dual fuel combination of low sulphur HFO and BOG would
then be used in such a way that the emissions of SO
2
are lower than what they would have been
if 0.1% fuel oil were used at berth. If the scheme were set up in such a way that BOG together
with minimum amounts of low sulphur HFO is used for the complete port stay (including
manoeuvring) the resulting SO
2
emissions could be lowered further. This option would allow for
fuel boost with HFO in the case of flame failure. This option requires fewer modifications on most
ships and thus lower modification costs. It may not be suitable for older ships. Whether or not this
solution is in compliance with the EU regulations depends on whether the equivalence principle
outlined is regarded to be in line with paragraph 4c. In addtion, the ships would still need to use
HFO after refit. This option would be likely to require a monitoring system with either control of
fuel use and sulphur levels in the fuel or the installation of flue gas monitoring systems to monitor
the SO
2
emissions. The cost of the latter is in the order of 70 000 per ship.

An option of using exhaust gas scrubbing would not available by the end of the year because the
technology has not yet been tested for this application.
10.7 Conclusions
There are genuine safety risks associated with switching from HFO to MGO in existing boilers
that are supported by evidence gained through contacts with the industry and authorities. There
are, however, technical and operational ways to circumvent these risks.

Although many boilers can be used with MGO, all boilers should be inspected. The exact number
of boilers that need to be modified is uncertain at this point. There are 262 LNG ships with steam
propulsion that need to be modified. There are around 2500 tankers that use boilers for steam
driven pumps and/or cargo heating - and most are likely to need to be modified.

The main safety concern is that a distillate fuel may vaporise in hot boilers with heated pipes. If
the boiler system has not been modified, there isin the case of flame failure and improper
handlinga risk of an explosion. There are technical solutions to solve these issues where fuel
piping systems, burners, and flame inspection systems are modified. The control system may
have to be modified and new handling routines established. For LNG ships, modifications are
required. The costs range from 5 000 to 25 000 for small auxiliary boilers, around 150 000 per
ship for tankers and 70 000 to 1 400 000 for LNG tankers.

AEA 79
11 Requirements with regard to maximum
sulphur content of heating oil for
condensation boilers
11.1 Objective
The purpose of this section of the report is to make an inventory of the requirements with regard
to the maximum sulphur content of heating oil in condensation boilers for heating purposes.
Where relevant, the requirements for different categories of condensation boilers are given.

11.2 Introduction to oil fired condensation boilers
Until recent years there were only household heating boilers based on constant temperature and
low temperature systems. All boilers used the same standard heating oil. With the introduction of
condensing boilers a distinction was introduced between boilers that can use only low sulphur
heating oils and other boilers that can use the different heating oil available on the market.
Condensation technology is an efficient way to use the heating value of burned natural gas and
oil. In low temperature boilers condensation of flue gas water vapour on the heat exchange
surfaces must be avoided because of corrosion. Condensation technology recovers the latent
heat of the water vapour in the flue gases by condensation. Condensation of water vapour occurs
when flue gases are cooled down to temperatures below dew point of the gases. The dew point
temperature of flue gases from oil fired boilers is about 47 C. The flue gas temperature from
condensation boilers is lower than of conventional boilers resulting in a better use of sensible
heat in the flue gases as well. In this way heat efficiency of condensation boilers is about 6 %
higher (and CO2 emissions about 6 % less) than for conventional oil fired boilers at nominal
capacity. Another advantage of condensation boiler technology is that the water temperature in
the heating system can be regulated according to outside temperature - whereas conventional
boilers provide a constant water temperature. Especially in spring and autumn when heating
demand is much lower than nominal capacity, the efficiency of condensing boilers is much higher
than 6% compared to conventional boilers because the firing is modulated [1].

11.3 Heating oil quality and availability
According to Directive 1999/32/EC the content of sulphur in heating oils to be used in EU
Members States from 1st of January 2008 should not exceed 0.1% by mass (=1000 mg/kg).
Since July 2000 the permitted sulphur content is, at maximum, 0.2%. The sulphur content of the
diesel oil used for vehicle fuel has been reduced from 50 mg/kg to >10 mg/kg from 1st of January
2009. In this chapter sulphur content is given in mg/kg oil, in the literature sulphur content is
sometimes given in ppm (parts per million) while mg/kg is meant. In Germany the availability of
low sulphur oil (< 50 mg/kg), originally meant for vehicles, has facilitated the introduction and use
of condensation boilers. There is also an intermediate quality oil - sulphur reduced heating oil
(50-500 mg/kg) - available on the market which can be used in the EU countries to reduce
emissions of sulphur dioxide to the environment:

The use of gas oil for heating has been in general decline for many years, inland consumption fell
by about 22% over the period 1995 to 2007, most of this was in the residential commercial sector
and is the result of switching to natural gas. Overall a continuing decrease in heating oil use is
expected as the result of further natural gas substitution and improved equipment efficiency (i.e.
replacement of old heating systems) but may be sustained in some markets i.e. housing
developments outside cities (beyond the gas distribution system) [6]

AEA 80
According to a Concawe study [7] the weighted average sulphur content in the gas oil
(automotive diesel fuel, heating oil, industrial gas oil and marine diesel fuel) used in the EU has
fallen from 1100 mg/kg in 1998 to 900 mg/kg in 2002. Average weighted sulphur content of gas
oil varies with European region; in 2002 average content was highest (1150 mg/kg) in the
Mediterranean region and lowest (470 mg/kg) in the former EFTA countries (Norway, Sweden,
Finland, Austria and Switzerland) compared to Northwest Europe and the Atlantic.

11.4 Construction of condensation boilers
In most conventional low temperature boilers hot flue gases flow horizontally through the heat
changer and condensation must be avoided. In condensation boilers hot flue gases flow
downwards through the heat exchanger, water is allowed to condense, this also drains
downwards providing a self cleaning effect that prevents the accumulation of salts on the heat
exchange surfaces. Depending on the amount of sulphur in the oil burned, and the acids
produced as a result, the water can be acidic, with a pH between 1.5 and 3.5 [3]; standard
heating oil produces a condensate that can be very corrosive unless suitable construction
materials have been used. Extensive experience has shown that heat exchangers made of
stainless steel are most suitable but specific aluminium alloys can be an acceptable alternative.
A variety of stainless steel alloys are available (with e.g. chromium, nickel, molybdenum, titanium)
that are durable and which do not require additional surface treatment.

11.5 Condensation boilers and sulphur content of heating oil
Traditional heating oil (EL quality) contains up to 2000 mg/kg sulphur, its flue gas condensate
contains significant amounts of sulphurous and sulphuric acids and, if produced in large
amounts, should be neutralized before discharge to the sewer system; neutralisation is not
required if low sulphur oil is used. Neither can standard heating oil be used in all condensation
boilers on the market - some producers only allow the use of low sulphur heating oil.
Condensation boilers for standard heating oil are specifically constructed for this duty and tend to
have lower heat efficiencies [4].

Originally producers of condensation boilers in Germany were reluctant to specify the use of low
sulphur oil but concerns were allayed following extensive tests [5]. A distinction should be made,
irrespective of whether or not heating oil of differing sulphur content can be used, between three
categories of condensation boilers:
- air-to-water condensation boilers
o boilers in which flue gas cooling and condensation both take place on the heat
exchange surface integrated in the boiler;
o boilers that cool flue gases to the dew point using an internal heat exchanger
and where condensation then occurs on separate heat exchanger.
- air-to-air condensation boilers
o boilers in which the condensation of the flue gas system is used to preheat
incoming combustion air.

Air-to-water condensation boilers in which flue gas cooling and condensation takes place on a
heat exchange surface which is integrated in the boiler
This category of condensation boilers has been on the market for a long time. The heat
exchangers are exposed to the high temperatures of the flame and are susceptible to deposition
of the substances resulting from burning of high sulphur oils. Heat exchangers in this category
have to be made of corrosion resistant material like stainless steel. According to [3] also some
specific formulated alloys of aluminium can be used. To prevent deposition and maintain high
heat efficiency, if cleaned only once a year, these condensation boilers ideally need low sulphur
heating oil (< 50 mg/kg) [1] but can tolerate sulphur reduced heating oil (< 500 mg/kg) [2]. The
use of high sulphur heating oil (1000 mg/kg) in this type of condensation boilers results in faster
deposition and decrease of heat efficiency but this can be avoided by cleaning the heat
exchanger surface at least twice a year.
AEA 81

Air-to-water condensation boilers that cool flue gases to the dew point using an internal heat
exchanger and where condensation occurs on separate heat exchanger. High sulphur heating oil
(1000-2000 mg/kg sulphur) can be used in this category of condensation boilers because
condensation does not take place in the first (internal) heat exchanger [1; 2]. Combustion
residues (containing reaction products of sulphur) deposit on the heat exchanger in the
combustion chamber but because there is no condensation at this point there is no corrosion of
the surfaces. Hence the heat exchanger can be made of standard steel. In the second (add-on)
heat exchanger, where the condensation occurs, is made of stainless steel and there is little
deposition. Because high or medium sulphur heating oil may be used for this category of
condensation boilers the condensate can be aggressive and neutralisation of the condensate
may be required before discharge - depending on heating capacity of the boiler and wastewater
quality regulations.

Air-to-air condensation boilers in which condensation takes place in the flue gas system to
preheat incoming combustion air.
In these systems the internal heat exchanger and water flow are designed in such a way that
condensation does not take place. Flue gases reach the second heat exchanger at temperatures
far above dew point (about 100C). Condensation in this second heat exchanger is then governed
by control of flue gas and combustion air flow. Heating efficiency of this category of
condensation boilers is lower than for the air-to-water condensation boilers.

As a summary the suitability of different types of heating oil in different categories of
condensation boilers is given in the table below.

Table 11.1 Suitability of heating oils in different categories of condensation boilers

Boiler integrated heat
exchanger
Add-on heat exchanger
High sulphur heating oil
(1000 mg/kg)
Problematic, high deposits on
heat exchange surface
Allowed, moderate deposits
Sulphur reduced heating oil
(50 500 mg/kg)
Allowed, moderate deposits Allowed, minor deposits
Low sulphur heating oil
(< 50 mg/kg)
Allowed, minor deposits Allowed, no deposits

11.6 Conclusions
Heating oils with sulphur content up to the maximum allowed concentration of 1000 mg/kg can be
used in condensation boilers. Deposits on heat exchanger surfaces caused by use of high
sulphur heating oils are more problematic in condensation boilers with integrated heat exchanger
than in boilers with an add-on condensation heat exchanger. To maintain high efficiency
integrated heat exchangers have to be cleaned more frequently than add-on heat exchangers
and more frequently than once a year as is usual. Efficiency will be kept at a high level for both
configurations if low sulphur oil is used.

There are no technical obstacles which prevent the use of low sulphur heating oil in condensation
boilers designed for standard heating oil.

AEA 82
12 References:

Section 4

Amman et al, 2008. National Emission Ceilings for 2020 based on the 2008 Climate and Energy
Package. NEC Scenario Analysis report, nr 6. IIASA.

Chiffi, Fiorello, Schrooten, De Vlieger, 2007. EX-TREMIS - Exploring non road Transport
Emissions in Europe --Annex to Final Report. IPTS - Institute for Prospective Technological
Studies, DG-JRC, Seville, Spain (www.ex-tremis.eu).

Cofala et al., 2007. Analysis of Policy Measures to Reduce Ship Emissions in the Context of the
Revision of the National Emission Ceilings Directive. IIASA/MET.NO/ENTEC UK Ltd.

Roemer M. and others (2003). Ozone trends according to ten dispersion models. Eurotrac-2 ISS,
GSF-National Research Centre for Environment and Health, Munich, Germany.

Roemer M., T. van Harmelen and P. Builtjes (2006). Establishment of optimal control areas for
acidification, eutrophication and ground level ozone. TNO report 2006-A-R0251/B, TNO, Utrecht,
The Netherlands.

Schaap M., M. Roemer, F. Sauter, G. Boesen, R. Timmermans and P. Builtjes (2005). LOTOS-
EUROS: documentation. TNO report B&O-A R2005/297, TNO, Utrecht, The Netherlands.

Van Harmelen T., M. Roemer, J. Pesik and G. Boersen (2007). Analysis of the potential costs
savings and environmental and health implications of emission trading for sulphur dioxide and
nitrogen dioxide. TNO report 2007-A-R1256/B, TNO, Utrecht, The Netherlands.

Van Loon M.,M. Roemer and P. Builtjes (2004). Model intercomparison in the framework of the
review of the Unified EMEP model. TNO report R2004/282, TNO, Utrecht, The Netherlands.

Section 7

Purvin&Gertz, 2009, Avis M.J., Birch C.H.,
IMPACTS ON THE EU REFINING INDUSTRY & MARKETS OF IMO SPECIFICATION
CHANGES & OTHER MEASURES TO REDUCE THE SULPHUR CONTENT OF CERTAIN
FUELS

Entec, 2005, Service contract on ship emissions: Assignment, Abatement and Market-based
instruments

UK-Maritime and Coastgard Agency, 2009, Impact assessment of merchant Shipping (prevention
of Air pollution from ships (Amendment)) Regulations 2009

US-EPA, 2009, Proposal to Designate an Emission Control Area for Nitrogen Oxides, Sulfur
Oxides and Particulate Matter - Technical Support Document - Chapter 5 Costs, Assessment and
Standards Division Office of Transportation and Air Quality U.S. Environmental Protection Agency

Krystallion, Sea Water Scrubbing Does it Contribute to increased Global CO2 Emissions

AEA 83
www.purchasingdata.com

Wrtsil, 2009, General description of Wrtsil scrubber system

Marine and Energy Consulting limited, 2009, presentation at 2nd Bunker Fuel Management, 16th
& 17th September, London, UK

Delft 2006. Greenhouse Gas Emissions for Shipping and Implementation
Guidance for the Marine Fuel Sulphur Directive

Chiffi, Fiorello, Schrooten, De Vlieger, 2007. EX-TREMIS - Exploring non road Transport
Emissions in Europe --Annex to Final Report. IPTS - Institute for Prospective Technological
Studies, DG-JRC, Seville, Spain.

Section 9

Winnes, H and Fridell, E, Particle Emissions from Ships: Dependence on Fuel Type, J. Air &
Waste Manage. Assoc. 59:1391 (2009).

V. Eyring, H. W. Khler, J. van Aardenne and A. Lauer, Emissions from international shipping: 1.
The last 50 years. J. Geophys. Res. 110, D17305 (2005).

Skema 2009: Impact Study of the future requirements of Annex VI of the MARPOL Convention on
Short Sea Shipping, Skema 2009.

MEPC 59/WP.10, 16 July 2009

US EPA, 1976. Quality criteria for water. United States Environmental Protection Agency,
Washington D.C. 20460, The Red book, July 1976.

Polycyclic aromatic hydrocarbons (PAH), KEMI Swedish Chemicals Agency, www.kemi.se
Phenanthrene CAS Number 85-01-8, United States Environmental Protection Agency
www.epa.gov/waste/hazard/wastemin/minimize/factshts/phenanth.pdf

Billiard, S. M., Meyer, J. N., Wassenberg, D. M., Hodson, P. V., Di Giulio, R. T. 2008. Nonadditive
effects of PAHs on early vertebrate development: Mechanisms and implications for risk
assessment. Toxicol. Sci. 105: 5-23.

Behrends, B., & Liebezeit, G., 2003, Reducing SO
X
and NO
X
Emissions from Ships by a
Seawater Scrubber. Research Centre Terramare, Wilhelmshaven, Germany.

Hufnagel, M., Liebezeit, G., & Behrends, B., 2005. Effects of Sea Water Scrubbing, Final report.

Buhaug, ., Flgstad, H., Bakke T., 2006. MARULS WP3: Wash-water Criteria for seawater
exhaust gas-SO
X
scrubbers. MARINTEK report, for Norwegian Marine Technology.

Karle, I-M. and Turner, D., 2007. Seawater Scrubbing - reduction of SO
X
emissions from ship
exhausts. Department of Chemistry, Gteborg University, ISBN: 978-91-976534-1-1.

MEPC 59/4/19. Marine Environment Protection Committee, 59 th session Agenda item 4.
Prevention of air pollution from ships. Advice by GESAMP on the interim criteria for discharge of
wash-water from Exhaust Gas Cleaning Systems for removal of sulphur-oxides.

Whall et al., Quantification of emissions from ships associated with ship movements between
ports in the European Community, Entec 2002.

AEA 84
Cooper, D. A., Peterson, K., and Simpson, D., 1996, Hydrocarbon, PAH and PCB emissions from
ferries: a case study in the Skagerak-Kattegatt-resund region, Athmospheric Environment 30,
2463.

MEPC 56/INF.5, ANNEX 1, Page 22

Directive 2008/105/EC of the European Parliament and of the Council of 16 December 2008 on
environmental quality standards in the field of water policy, amending and subsequently repealing
Council Directives 82/176/EEC, 83/513/EEC, 84/156/EEC, 84/491/EEC, 86/280/EEC and
amending Directive 2000/60/EC of the European Parliament and of the Council.

Jalkanen, J. -P., Brink A., Kalli, J., Pettersson, H., Kukkonen, J. and Stipa, T., 2009, A modelling
system for the exhaust, emissions of marine traffic and its application in the Baltic Sea area,
Atmos. Chem. Phys. Discuss., 9, 1533915373.

Benedictow et al., 2009. Transboundary Acidification,Eutrophication and Ground Level Ozone
in Europe in 2007, EMEP Status Report.

Johnsen et al., (2006) Part C of Delft report Greenhouse Gas Emissions for Shipping and
Implementation Guidance for the Marine Fuel Sulphur Directive, DNV.

Mellqvist, J, Berg, N. and Ohlsson, D., 2008. Remote surveillance Remote surveillance of the
sulfur content and NO
x
emissions of ships, Presentation at Haqcc

Section 10

Aalborg 2009. Aalborg Industries. Aalborg Solutions No 12 January 2009. http://www.aalborg-
industries.com/marine_solutions/documents/12FuelchangetoMDOMGO-09JAN09.pdf.

British Department of Transport 1983. Explosion in boiler furnaces, Merchant Shipping Notice No.
M. 1083, August 1983.

CARB 2008. California Environmental Protection Agency Air Resource Board, Fuel sulfur and
other operational requirements for ocean-going vessels within California waters and 24 nautical
miles of the California baseline, June 2008.
http://www.arb.ca.gov/regact/2008/fuelogv08/isorfuelogv08.pdf.

DNV 2005. DNV, Regulations for the prevention of air pollution from ships, technical and
operational implications. February 2005. http://www.dnv.com/binaries/Marpol_Annex_VI_tcm4-
125462.pdf.

Entec 2002. C Whall et al., Quantification of emissions from ships associated with ship
movements between ports in the European Community, Entec, European Commission, 2002.

IMO subcommittee on bulk liquids, 2007. Review of MARPOL VI and the NO
x
technical code
December, 2007.

US coast guard Maritime Safety Alert, 2009. Avoiding propulsion loss from fuel switching:
American Petroleum Institute Technical Considerations. May 2009.

Section 11

Condensation technology Condensation technology for energy saving and a clean
environment.Documentation Viessmann (www.viessmann.com) [1]
AEA 85

Rogatty, W.: Modern oil-heating technology. IKZ-Haustechnik, Volume 8/2004, page 60 [2]
.
Brennwerttechnik. : http://www.energiesparmobil.de/downloads_infomaterial/pdf/haustechnik/brennwerttechnik.pdf
[3]

Frequently asked questions and answers in relation to the common declaration of the National
Government and the Mineral oil business on January 15th 2007. Bundesministerium fr Umwelt,
Naturschutz und Reaktorsicherheit. (http://www.bmu.de/energieeffizienz/gebaeude/doc/38680.php) [4]

Gerteeignung und Verfgbarkeit. Institut fr wirtschaftliche Oelheizung e.V.
http://www.iwo.de/sites/iwo_website/standard.jsp?nodeId=159355&open=x&pagename=TECH_HELSA_0 [5]

Impacts on the EU refining industry & markets of IMO specification changes & other measures to
reduce the sulphur content of certain fuels. Purvin & Gertz Inc., 30 June 2009 [6]

Post, L.: Sulphur dioxide emissions from oil refineries and combustion of oil products in Western
Europe and Hungary (2000). CONCAWE report no. 2/07, March 2007 [7]

Appendix 1

Cooper and Gustafsson, 2004 D. Cooper and T. Gustafsson, Methodology for Calculating
Emissions from Ships. 1. Update of Emission Factors Report Series for SMED and SMED&SLU,
SMHI, Norrkping, Sweden (2004) www.smed.se ISSN: 1652-4179.

IPCC 1995, Greenhouse gas inventory Reference manual. Intergovernmental Panel on Climate
Change.

Lloydss Register, 1995. Marine Exhaust Emissions Research Programme. Lloydss Register
Engineering Services, London.

TFEIP, 2009. EMEP/EEA Emission Inventory Guidebook, 2008 version, Task Force for Emission
Inventories and Projections (TFEIP). Available via http://www.eea.europa.eu/publications/emep-
eea-emission-inventory-guidebook-2009

AEA 86
13 Appendix 1 - Emission factors used

Table 1: Unabated emission factors used in the baseline scenario
kg/ton kg/GJ source notes
heavy fuel
oil

SO2-SECA 29 0.72 TFEIP 2009
1.45% S, after 2006 in North Sea and Baltic Sea
in baseline
SO2 54 1.34 TFEIP 2009 2,70% S, pre-2006
NOx 43 1.07 Ex-tremis
NMVOC 4.9 0.12 Ex-tremis
CO 5.5 0.14 IVL and Lloyds
PM 3.4 0.08 Ex-tremis
diesel oil
SO2 4 0.09 TFEIP 2009 0.2% S
NOx 43 0.99 Ex-tremis
NMVOC 4.9 0.11 Ex-tremis
CO 5.5 0.14 IVL and Lloyds
PM 2.6 0.06 Ex-tremis
Note: Since 1-1-2008, for seagoing ships on route between member states, the sulphur content of diesel fuel has
been lowered to 0.1%. Data were insufficient to make this distinction in the emission calculation.

Table 2: Unabated emission factors by fuel and motor speed
Fuel consumption NOx NMVOC
CO

g fuel/kWh kg/ton kg/GJ kg/ton kg/GJ kg/ton kg/GJ
GT-MG/DO 290 19.6 0.45 0.3 0.01 0.3 0.01
GT-RO 305 20.0 0.50 0.3 0.01 0.3 0.01
HSD-
MG/DO
203 59.1 1.36 0.9 0.02 5.5 0.13
HSD-RO 213 59.6 1.48 0.9 0.02 5.5 0.14
MSD-
MG/DO
203 61.2 1.41 2.5 0.06 5.5 0.13
MSD-RO 213 65.7 1.63 2.3 0.06 5.5 0.14
SSD-
MG/DO
185 92.2 2.13 3.2 0.07 7.1 0.16
SSD-RO 195 92.8 2.31 3.1 0.08 7.1 0.18
ST-MG/DO 290 6.9 0.16 0.3 0.01 0.7 0.02
ST-RO 305 6.9 0.17 0.3 0.01 0.7 0.02
Note: ]
GT-MG/DO: gas turbine - marine diesel oil/marine gasoil
GT-RO: gas turbine bunker fuel oil
HSD-MG/DO: high speed diesel - marine diesel oil/marine gasoil
HSD-RO: high speed diesel bunker fuel oil
MSD-MG/DO: medium speed diesel - marine diesel oil/marine gasoil
MSD-RO: medium speed diesel bunker fuel oil
SSD-MG/DO: slow speed diesel - marine diesel oil/marine gasoil
SSD-RO: slow speed diesel bunker fuel oil
ST-MG/DO: steam turbines - marine diesel oil/marine gasoil
ST-RO: steam turbines bunker fuel oil

Source:
TFEIP 2009 1.A.3.d Table 3-5, and (for CO) IVL and Lloyds (for Diesel engines) and Cooper
and Gustafsson, 2004 for GT and ST.

AEA 87

Table 3: Unabated PM emission factors by fuel and motor speed and sulphur
content of the fuel

Note: see table 2 for abbreviations;

Table 4: SO
2
emission factors in the scenarios
Outside SECAs Within SECAs Baltic
Black
Sea
Mediter-
ranean
North
Sea
Remaining
NE-atl
Scenar
io
S content
(mass%)
EF
(kg/GJ)
S content
(mass%)
EF
(kg/GJ)
IN
SECA?
IN
SECA?
IN
SECA?
IN
SECA?
IN SECA?
1 2.94% 1.46 0.10% 0.05 Y N N Y N
2 2.94% 1.46 0.10% 0.05 Y N Y Y N
3 2.94% 1.46 0.10% 0.05 Y Y Y Y N
4 0.50% 0.25 0.10% 0.05 Y N N Y N
5 0.50% 0.25 0.10% 0.05 Y Y Y Y N
BASE
LINE
2.70% 1.34 1.45% 0.72 Y N N Y N
Sulphur
content
0.10% 0.20% 0.50% 1.50% 2.70%
kg/ton kg/GJ kg/ton kg/GJ kg/ton kg/GJ kg/ton kg/GJ kg/ton kg/GJ
GT-MG/DO 0.017 0.000 0.023 0.001 0.039 0.001
GT-RO 0.059 0.001 0.064 0.002 0.078 0.002 0.126 0.003 0.183 0.005
HSD-
MG/DO
0.488 0.011 0.651 0.015 1.139 0.026
HSD-RO 0.843 0.021 0.910 0.023 1.112 0.028 1.787 0.044 2.596 0.065
MSD-
MG/DO
0.488 0.011 0.651 0.015 1.139 0.026
MSD-RO 0.843 0.021 0.910 0.023 1.112 0.028 1.787 0.044 2.596 0.065
SSD-
MG/DO
0.541 0.012 0.721 0.017 1.261 0.029
SSD-RO 2.416 0.060 2.609 0.065 3.189 0.079 5.121 0.127 7.440 0.185
ST-MG/DO 0.500 0.012 0.667 0.015 1.167 0.027
ST-RO 0.950 0.024 1.026 0.026 1.254 0.031 2.015 0.050 2.927 0.073

AEA 88

Table 5: NO
x
emission factors in scenarios
Tier I Tier II Tier III
kg/GJ

ships constructed
between 1-1-2000
and 1-1-2010
ships constructed
after 1-1-2011
for 2020 only:
ships constructed
after 1-1-2016
GT-MG/DO 0.45234 0.45234 0.45234
GT-RO 0.45234 0.45234 0.45234
HSD-MG/DO 1.06184 1.06184 0.21670
HSD-RO 1.06184 1.06184 0.21670
MSD-MG/DO 1.33494 1.07477 0.26699
MSD-RO 1.33494 1.07477 0.26699
SSD-MG/DO 1.84196 1.56025 0.36839
SSD-RO 1.84196 1.56025 0.36839
ST-MG/DO 0.15924 0.15924 0.15924
ST-RO 0.15924 0.15924 0.15924
Note: see table 2 for abbreviations

Table 6: Applicability of the Tiers in the various scenarios
Scenario
1 Tier I&II, for 2020 Tier III
2 Tier I& II
3 Tier I&II, for 2020 Tier III
4 Tier I,II & III (2020 only)
5 Tier I,II & III (2020 only)

Background on NO
x
limits to marine diesel engines and Emission Control Areas (ECAs)

NO
X
emissions are regulated under the International Maritime Organization (IMO), in the
International Convention for the Prevention of Pollution From Ships, 1973 as modified by the
Protocol of 1978 (in short MARPOL 73/78). NOx limits for marine diesel engines are set in Annex
VI of MARPOL and the revised NOx Technical Code.
In October 2008, the IMOs Marine Environmental Protection Committee (MEPC) adopted both
the revised Annex VI of MARPOL and the revised NOx Technical Code. The annex and the code
will enter into force 1 July 2010.

The new revision substantially tightens the NOx and Sulphur limits compared to the existing
annex, and also includes requirements governing NOx emissions from ships constructed from 1
January 1990 to 1 January 2000.

New NO
x
limits
The revised NOx regulations contain a three-Tier approach as follows:
Tier I (identical to todays limits)
For diesel engines installed on ships constructed from 1 January 2000 to 1 January 2011, the
allowable NOx emissions are:

AEA 89
17.0 g/kWh when n is less than 130 rpm
45.0*n(-0.2) g/kWh when n is 130 rpm or more but less than 2,000 rpm
9.8 g/kWh when n is 2,000 rpm or more

Tier II
For diesel engines installed on ships constructed on or after 1 January 2011, the allowable NOx
emissions are:

Tier III
Ships constructed on or after 1 January 2016 will have additional limitations when operating in an
Emission Control Area (ECA). No ECAs have yet been designated for NO
x
emissions, but it is
expected that both the Baltic Sea and North Sea will be designated as NO
x
ECAs well ahead of 1
January 2016. The allowable NOx emissions are:


AEA 90
14 Appendix 2 - Emission Figures by
Scenario and Area Baseline

Emission Figures by Scenario and Area Baseline
Baltic Sea Black Sea
Remaining NE Atlantic
ocean
x 1000
tonnes
2000 2015 2020 2000 2015 2020 2000 2015 2020
SO
2
177 149 169 53 82 93 472 730 829
NMVOC 10 15 17 3 5 5 26 40 46
CO 22 34 39 7 10 12 58 90 103
PM 9 15 17 6 10 11 56 88 100
NO
x
283 439 498 83 130 147 742 1153 1310

Mediterranean Sea North Sea
X 1000
tonnes
2000 2015 2020 2000 2015 2020
SO
2
1024 1582 1795 417 351 399
NMVOC 56 86 98 23 36 41
CO 125 194 220 52 81 92
PM 118 185 210 22 34 39
NO
x
1593 2469 2805 664 1031 1171

SEA1
Remaining
NE Atlantic
ocean
Mediterranean
Sea
North Sea
x 1000
tonnes
2015 2020 2015 2020 2015 2020 2015 2020 2015 2020
SO
2
10 12 90 102 795 903 1722 1955 24 28
NMVOC 15 17 5 5 40 46 86 98 36 41
CO 34 39 10 12 90 103 194 220 81 92
PM 11 13 10 12 88 100 185 210 26 30
NO
x
364 325 108 115 957 1,018 2,055 1,841 856 765


AEA 91
SEA2
Remaining
NE Atlantic
ocean
Mediterranean
Sea
North Sea
x 1000
tonnes
2015 2020 2015 2020 2015 2020 2015 2020 2015 2020
SO
2
10 12 90 102 795 903 59 67 24 28
NMVOC 15 17 5 5 40 46 86 98 36 41
CO 34 39 10 11 90 103 194 220. 81 92
PM 11 13 10 11 88 100 62 71 26 30
NO
x
364 388 108 115 957 1,018 2,055 2,187 856 911

SEA3
Remaining
NE Atlantic
ocean
Mediterranean
Sea
North Sea
x 1000
tonnes
2015 2020 2015 2020 2015 2020 2015 2020 2015 2020
SO
2
10 12 3 3 795 903 59 67 24 28
NMVOC 15 17 5 5 40 46 86 98 36 41
CO 34 39 10 12 90 103 194 220 81 92
PM 11 13 3 4 88 100 62 71 26 30
NO
x
364 325 108 115 957 1,018 2,055 1,841 856 765

SEA4
Remaining
NE Atlantic
ocean
Mediterranean
Sea
North Sea
x 1000
tonnes
2015 2020 2015 2020 2015 2020 2015 2020 2015 2020
SO
2
10 12 15 17 135 154 293 333 24 28
NMVOC 15 17 5 5 40 46 86 98 36 41
CO 34 39 10 12 90 103 194 220 81 92
PM 11 14 4 5 39 44 81 92 26 30
NO
x
364 325 108 115 957 1,018 2,055 1,841 856 765

SEA5
Remaining
NE Atlantic
ocean
Mediterranean
Sea
North Sea
x 1000
tonnes
2015 2020 2015 2020 2015 2020 2015 2020 2015 2020
SO
2
10 12 3 3 135 154 59 67 24 28
NMVOC 15 17 5 5 40 46 86 98 36 41
CO 34 39 10 12 90 103 194 220 81 92
PM 11 13 3 4 39 44 62 71 26 30
NO
x
364 325 108 115 957 1,018 2,055 1,841 856 765


AEA 92
Emissions land based sources, baseline and scenarios Baseline

2000 2010 2015 2020
x 1000 tonnes
SO
2
16,726 10,324 9,897 9,734
NMVOC 16,236 13,033 12,374 11,931
PM10 4,726 4,013 3,988 3,988
PM2.5 3,231 2,768 2,716 2,678
NO
x
16,582 13,839 12,691 11,312
Note: exclusive of emissions related to national maritime fuel consumption (e.g. fisheries), which are included in the
GAINS-totals
Source: GAINS Europe, NEC6, C&E package current policy

Land based A

2015 2020
x 1000 tonnes
SO
2
9,851 9694
NMVOC 12,374 11931
PM10 3,988 3,988
PM2.5 2,716 2,678
NO
x
12,691 11,312

Land based B

2015 2020
x 1000 tonnes
SO
2
9,806 9,665
NMVOC 12,374 11,931
PM10 3,988 3,988
PM2.5 2,716 2,678
NO
x
12,691 11,312


AEA 93
15 Appendix 3 - Relevant geographical
representations of the dispersion
calculations

In this appendix the most relevant results of the dispersion calculations are presented.
First some background information is given on the relevant background concentrations used in
the dispersion calculations and the effect of inter-annual changes due to the meteorology. As
states earlier we used the meteorology of 1997 given the fact that this is considered a
representative year for the European meteorology.
The absolute values of the calculated pollutant concentration are certainly relevant but in this cost
benefit analysis the emphasis is on the differences between the results for the different scenarios.

Sensitivity of results to inter-annual variability in meteorology and background ozone
concentrations.

Background ozone and changes therein are important to metrics such as AOT40, SOMO35 etc.
The increase of hemispheric ozone is under scientific debate and it is not easy to answer how
much more of ozone we can expect in 2020. Because the effect of IMO regulations on NO
x

emissions in the remaining North Atlantic beyond 10 W is not quantified we do not know how this
will affect background ozone concentrations. In general terms one can state; with each 1ug/m
3

increase of the background one can expect at the fringe of Europe, as a rough maximum
estimate, an increase of 2500 ug/m
3
.h in AOT40f, simply being the number of associated hours in
the AOT40f definition that will be increased with 1 ug/m
3
, plus a few hours that in addition will
exceed the threshold. At the interior of Europe this will be less, since part (say 70%) of the extra
ozone will be consumed by dry deposition, leaving an increase as a rule of thumb of 800 ug/m
3
.h.
Suppose the hemispheric increase of ozone between 2010 and 2020 will be 4 ug/m3, then the
AOT40f in continental Europe will increase by about 3200 ug/m3.h which is an increase of 10%
compared to 2010 levels.

Based on Laurila et al., 2004 we find, following their lowest SRES scenario, an increase of
AOT40f due to hemispheric changes of 2000 ug/m3.h from 2010 to 2020 on top of an AOT40f in
2010 of about 14000 ug/m3.h.

Uncertainty estimates in changes in hemispheric ozone are hard to give since the cause of the
changes is poorly understood; given this lack of knowledge we estimate an uncertainty of 100%.

The inter-annual variability in AOT40f can be as large as 8000 ug/m3.h fluctuation from one year
to another, but on average the fluctuation is typical in the other of 3000 ug/m3.h. For continental
sites as in Switzerland and Austria the aforementioned fluctuations are 30% and 12% of the long
term average of AOT40f in these countries. In the UK of the Netherlands (with much lower long-
term baseline AOT40f values) the fluctuations in relative terms are much larger, typically in the
order of 100-50%. More important are the inter-annual fluctuations of the scenarios. In absolute
terms (change in ug/m3.h) the fluctuations will be considerable, but once expressed in terms of
fractional changes with respect to the baseline AOT40f the inter-annual fluctuations are much
smaller, and more robust against the choice of meteorological year.
For other ozone metrics that are derived from larger portions of the ozone distribution, such as
SOMO35 (annual sum of daily max eight hour running mean of ozone exceeding 35 ppb) and
annual mean the fluctuations from year to year are smaller than for AOT40.


AEA 94
The atmosphere is never in equilibrium. The lifetime of NO
x
is short but due its effect on ozone
(which has a longer lifetime) still in the order of weeks to months. However, I do not see any
transient effects in the range of years due to NO
x
emission changes.

The interannual variation in PM concentrations is typically in the order of 10%. Again, scenario
changes expressed with respect to the baseline are less sensitive to interannual variability.

Based on the EMEP status report 2005/1 in which the exceedances of critical loads for either
acidification or eutrophication in 2002 and 2003 were compared (with 2003 being an extreme
year) the relative differences amount to about 20% from one year to another

Does the calculation grid size have an effect on the calculated benefits?
A smaller grid means a smaller grid for the dispersion calculations and for the
ecosystems/population data bases. Assuming that the concentration (deposition) in a 50x50 km
grid cell differs only slightly from the averaged concentration (deposition) of all associated 10x10
grid cells, then we can expect differences in the choice of grid cell only if the set 10x10 km
concentration data within 50x50 km bears some correlation with the set of 10x10 km ecosystem
(population) data. This will be specified for the themes separately.

Acidification. / Eutrophication
We have no information on the distribution of sensitive ecosystems (with regard to acidification or
eutrophication) at scales smaller than 50x50 km. Therefore, the question cannot be addressed.

SOMO35 (annual sum of daily max eighth hour running mean of ozone exceeding 35 ppb. )
In this study the SOMO35 values are multiplied with population numbers.
In rural areas with low NO
x
concentrations grid cells within a 50x50 km grid cell will not differ
much from the value in a 50x50 km cell. This part wont be affected by working on smaller scales.
In urban areas, characterised by high levels of NO
x
ozone concentrations (SOMO35) vary in
general from relatively low values in the city centres (high NO
x
levels, densely populated) to
relatively high values in residential suburban areas (lower NO
x
, less populated). The effect is to
put more weight on the low values, and consequently, a 10x10 km grid in an urban area results in
somewhat lower SOMO35*inhabitants values than a corresponding 50x50 km would.

PM10
For PM10 similar considerations apply as for SOMO35, although with a different sign. In rural
areas, there is no effect expected from reducing the grid size. In urban areas the primary PM10
emission (assuming it is proportional to population numbers) correlates with populations and, as a
result, so does the PM10 concentration (although the effect is smaller due to the damping effect
of the smoother sulphate and nitrate fields). The effect is that in urban areas the
PM10*inhabitants values are larger for a smaller grid than for a larger grid.


AEA 95


AEA 96

F
i
g
u
r
e

A
1
:

A
v
e
r
a
g
e

A
c
c
u
m
u
l
a
t
e
d

E
x
c
e
e
d
e
n
c
e

f
o
r

a
c
i
d
i
f
i
c
a
t
i
o
n

(
e
q
.
/
h
a
/
y
r
)

f
o
r

2
0
1
5

.

a
)

B
a
s
e
l
i
n
e

I
n

a
l
l

a
r
e
a
s
,

n
o
t

c
o
l
o
r
e
d

g
r
e
e
n
,

t
h
e
r
e

i
s

a
n

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

b
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A
1

c
o
m
p
a
r
e
d

t
o

b
a
s
e

l
i
n
e

T
h
e

m
a
p

s
h
o
w
s

c
l
e
a
r
l
y

t
h
e

r
e
d
u
c
t
i
o
n

o
f

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

a
l
l

r
o
u
n
d

t
h
e

N
o
r
t
h

S
e
a

a
n
d

t
h
e

B
a
l
t
i
c

d
u
e

t
h
e

d
e
s
i
g
n
a
t
i
o
n

o
f

t
h
e
s
e

w
a
t
e
r
s

a
s

S
E
C
A
.

c
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A

2

c
o
m
p
a
r
e
d

t
o

S
E
A

1

T
h
u
s

t
h
e

e
f
f
e
c
t

o
f

i
n
c
l
u
s
i
o
n

o
f

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

s
e
a

a
n
d

B
l
a
c
k

s
e
a

i
n

t
h
e

S
E
C
A

d
o
e
s

a
d
d

o
n
l
y

t
o

a

s
m
a
l
l

e
x
t
e
n
d

t
o

t
h
e

i
m
p
r
o
v
e
m
e
n
t

g
a
i
n
e
d

i
n

s
c
e
n
a
r
i
o

S
E
A
1

c
o
m
p
a
r
e
d

t
o

t
h
e

b
a
s
e
l
i
n
e
.

d
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A

3

c
o
m
p
a
r
e
d

t
o

S
E
A

3

R
e
d
u
c
t
i
o
n

o
f

t
h
e

S

c
o
n
t
e
n
t

o
u
t
s
i
d
e

t
h
e

S
E
C
A
s

d
o
e
s

l
o
w
e
r

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

i
n

t
h
e

U
K

a
n
d

I
R
.

S
e
e

a
l
s
o

t
a
b
l
e

7

i
n

a
p
p
e
n
d
i
x

4


AEA 97


AEA 98
F
i
g
u
r
e

A
2

A
v
e
r
a
g
e

A
c
c
u
m
u
l
a
t
e
d

E
x
c
e
e
d
e
n
c
e

f
o
r

a
c
i
d
i
f
i
c
a
t
i
o
n

(
e
q
.
/
h
a
/
y
r
)

f
o
r

2
0
1
5
.

.

a
)

B
a
s
e
l
i
n
e

I
n

a
l
l

a
r
e
a
s
,

n
o
t

c
o
l
o
r
e
d

g
r
e
e
n
,

t
h
e
r
e

i
s

a
n

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

b
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

B
b
a
s
e

c
o
m
p
a
r
e
d

t
o

B
a
s
e
l
i
n
e

T
h
e

m
a
p

s
h
o
w
s

t
h
e

p
o
s
i
t
i
v
e

e
f
f
e
c
t

o
n

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

f
o
r

a
c
i
d
i
f
i
c
a
t
i
o
n

w
h
e
n

l
o
w
e
r
i
n
g

t
h
e

S

c
o
n
t
e
n
t

o
f

f
u
e
l

o
n

l
a
n
d

t
o

0
.
0
0
1
%
.

T
h
i
s

m
a
p

i
n
d
i
c
a
t
e
s

t
h
e

m
a
x
i
m
u
m

e
f
f
e
c
t

o
f

l
o
w
e
r
i
n
g

t
h
e

S

c
o
n
t
e
n
t

i
n

l
a
n
d

b
a
s
e
d

f
u
e
l
.

c
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

B
S
E
A
1

c
o
m
p
a
r
e
d

t
o

S
E
A

1

T
h
e

m
a
p

s
h
o
w
s

t
h
e

s
a
m
e

p
i
c
t
u
r
e

a
s

m
a
p

b
)
,

t
h
e

m
i
n
o
r

d
i
f
f
e
r
e
n
c
e
s

a
r
e

c
a
u
s
e
d

b
y

t
h
e

f
a
c
t

t
h
a
t

i
n

t
h
e
s
e

s
c
e
n
a
r
i
o
s

t
h
e

N
o
r
t
h

S
e
a

/
B
r
i
t
i
s
h

c
h
a
n
n
e
l

a
n
d

t
h
e

B
a
l
t
i
c

a
r
e

a
l
r
e
a
d
y

d
e
s
i
g
n
a
t
e
d

a
s

S
E
C
A

a
n
d

t
h
u
s

e
x
c
e
e
d
e
n
c
e
s

w
e
r
e

a
l
r
e
a
d
y

r
e
d
u
c
e
d
.

T
h
e

g
e
n
e
r
a
l

c
o
n
c
l
u
s
i
o
n

o
f

t
h
e
s
e

m
a
p
s

i
s

t
h
a
t

t
h
e

e
f
f
e
c
t

o
n

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

f
o
r

a
c
i
d
i
f
i
c
a
t
i
o
n

w
h
e
n

l
o
w
e
r
i
n
g

t
h
e

S

c
o
n
t
e
n
t

o
f

f
u
e
l

o
n

l
a
n
d

i
s

m
a
r
g
i
n
a
l

a
p
p
r
o
x
i
m
a
t
e
l
y

1
.
3

t
o

2
.
8

%

r
e
d
u
c
t
i
o
n
.

B
e
c
a
u
s
e

a
c
i
d
i
f
i
c
a
t
i
o
n

i
s

t
h
e

p
a
r
a
m
e
t
e
r

e
f
f
e
c
t
e
d

t
h
e

m
o
s
t

b
y

s
u
l
f
u
r

e
m
i
s
s
i
o
n
s

t
h
i
s

a
n
a
l
y
s
i
s

i
s

n
o
t

p
e
r
f
o
r
m
e
d

f
o
r

t
h
e

o
t
h
e
r

p
a
r
a
m
e
t
e
r
s

S
e
e

a
l
s
o

t
a
b
l
e

7

a
n
d

9

i
n

a
p
p
e
n
d
i
x

4


AEA 99


AEA 100

F
i
g
u
r
e

A
3
:

A
v
e
r
a
g
e

A
c
c
u
m
u
l
a
t
e
d

E
x
c
e
e
d
e
n
c
e

f
o
r

e
u
t
r
o
p
h
i
c
a
t
i
o
n

(
e
q
.
/
h
a
/
y
r
)

f
o
r

2
0
1
5
.

.

a
)

B
a
s
e
l
i
n
e

I
n

a
l
l

a
r
e
a
s
,

n
o
t

c
o
l
o
r
e
d

g
r
e
e
n
,

t
h
e
r
e

i
s

a
n

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

b
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A
1

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
e

m
a
p

s
h
o
w
s

t
h
e

r
e
d
u
c
t
i
o
n

o
f

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

a
s

a

r
e
s
u
l
t

o
f

t
h
e

N
O
x

T
i
e
r

I

&

I
I

s
t
a
n
d
a
r
d
s

i
n

a
l
l

E
u
r
o
p
e
a
n

s
e
a
s
.

T
h
e

d
e
s
i
g
n
a
t
i
o
n

o
f

t
h
e

N
o
r
t
h

S
e
a

a
n
d

t
h
e

B
a
l
t
i
c

d
e
s

n
o
t

h
a
v
e

a
n

s
i
g
n
i
f
i
c
a
n
t

i
m
p
a
c
t

o
n

t
h
e

e
x
c
e
e
d
e
n
c
e
.

W
h
i
c
h

c
a
n

b
e

s
e
e
n

i
n

t
h
e

n
e
x
t

m
a
p

c
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A
2

c
o
m
p
a
r
e
d

t
o

S
E
A
1

T
h
u
s

t
h
e

e
f
f
e
c
t

o
f

i
n
c
l
u
s
i
o
n

o
f

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

s
e
a

a
n
d

B
l
a
c
k

s
e
a

i
n

t
h
e

d
o
e
s

n
o
t

a
d
d

s
i
g
n
i
f
i
c
a
n
t

t
o

t
h
e

i
m
p
r
o
v
e
m
e
n
t

g
a
i
n
e
d

i
n

s
c
e
n
a
r
i
o

S
E
A
1

c
o
m
p
a
r
e
d

t
o

t
h
e

b
a
s
e
l
i
n
e
.

d
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A
3

c
o
m
p
a
r
e
d

t
o

S
E
A
2

E
x
t
e
n
s
i
o
n

o
f

t
h
e

S
E
C
A

w
i
t
h

t
h
e

B
l
a
c
k

S
e
a

d
o
e
s

l
o
w
e
r

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

o
n
l
y

t
o

a

v
e
r
y

s
m
a
l
l

e
x
t
e
n
d
.

S
e
e

a
l
s
o

t
a
b
l
e

1
0

e
n

1
2

i
n

a
p
p
e
n
d
i
x

4


AEA 101


AEA 102

F
i
g
u
r
e

A
4

A
O
T
4
0
f

(
p
p
m
.
h
)

f
o
r

2
0
1
5
.

a
)

B
a
s
e
l
i
n
e

b
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

S
E
A
1

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
e

m
a
p

s
h
o
w
s

c
l
e
a
r
l
y

t
h
e

r
e
d
u
c
t
i
o
n

o
f

t
h
e

o
z
o
n
e

c
o
n
c
e
n
t
r
a
t
i
o
n

a
s

a
n

r
e
s
u
l
t

o
f

t
h
e

r
e
d
u
c
e
d

N
O
x

e
m
i
s
s
i
o
n
s

i
n

a
l
l

s
e
a
s

a
s

a

r
e
s
u
l
t

o
f

t
h
e

T
i
e
r

s
t
a
n
d
a
r
d
s
.

T
h
e

e
f
f
e
c
t

i
n

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

i
s

e
n
h
a
n
c
e
d

b
y

t
h
e

l
o
n
g
e
r

p
e
r
i
o
d
s

o
f

s
u
n
s
h
i
n
e
.

T
h
e

m
a
p
s

c

a
n
d

d

s
h
o
w

a

s
i
m
i
l
a
r

p
i
c
t
u
r
e

b
u
t

t
h
e
r
e

a
r
e

s
m
a
l
l

d
i
f
f
e
r
e
n
c
e
s
.

T
h
e
s
e

a
r
e

e
l
a
b
o
r
a
t
e
d

i
n

t
a
b
l
e

2
2

i
n

a
p
p
e
n
d
i
x

4
.

c
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

S
E
A
2

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

.

d
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

S
E
A
3

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

.

S
e
e

a
l
s
o

t
a
b
l
e

2
2

a
n
d

2
4

i
n

a
p
p
e
n
d
i
x

4


AEA 103


AEA 104

.

F
i
g
u
r
e

A
5

S
O
M
O
3
5

(
p
p
m
.
h
)

f
o
r

2
0
1
5
.

.

a
)

B
a
s
e
l
i
n
e

b
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

S
E
A
1

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
e

m
a
p

s
h
o
w
s

c
l
e
a
r
l
y

t
h
e

r
e
d
u
c
t
i
o
n

o
f

t
h
e

o
z
o
n
e

c
o
n
c
e
n
t
r
a
t
i
o
n

a
s

a
n

r
e
s
u
l
t

o
f

t
h
e

r
e
d
u
c
e
d

N
O
x

e
m
i
s
s
i
o
n
s

i
n

a
l
l

s
e
a
s

a
s

a

r
e
s
u
l
t

o
f

t
h
e

T
i
e
r

s
t
a
n
d
a
r
d
s
.

T
h
e

e
f
f
e
c
t

i
n

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

i
s

e
n
h
a
n
c
e
d

b
y

t
h
e

l
o
n
g
e
r

p
e
r
i
o
d
s

o
f

s
u
n
s
h
i
n
e
.

T
h
e

m
a
p
s

c

a
n
d

d

s
h
o
w

a

s
i
m
i
l
a
r

p
i
c
t
u
r
e

b
u
t

t
h
e
r
e

a
r
e

s
m
a
l
l

d
i
f
f
e
r
e
n
c
e
s
.

T
h
e
s
e

a
r
e

e
l
a
b
o
r
a
t
e
d

i
n

t
a
b
l
e

1
6

i
n

a
p
p
e
n
d
i
x

4
.

c
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

S
E
A
2

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

.

d
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

S
E
A
3

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

.

S
e
e

a
l
s
o

t
a
b
l
e

1
6

a
n
d

1
8

i
n

a
p
p
e
n
d
i
x

4


AEA 105


AEA 106

F
i
g
u
r
e

A
6

P
M
1
0

(
u
g
/
m
3
)

f
o
r

2
0
1
5

.

a
)

B
a
s
e
l
i
n
e

b
)

S
E
A

1

T
h
e

m
a
p

s
h
o
w
s

t
h
e

p
o
s
i
t
i
v
e

e
f
f
e
c
t

o
n

P
M

c
o
n
c
e
n
t
r
a
t
i
o
n

w
h
e
n

d
e
s
i
g
n
a
t
i
n
g

t
h
e

N
o
r
t
h

S
e
a

/

E
n
g
l
i
s
h

C
h
a
n
n
e
l

a
n
d

T
h
e

B
a
l
t
i
c

a
s

S
E
C
A
.

T
h
e

r
e
d
u
c
t
i
o
n

o
f

s
u
l
f
u
r

i
n

f
u
e
l
s

i
s

a
s
s
o
c
i
a
t
e
d

w
i
t
h

a
n

r
e
d
u
c
t
i
o
n

i
n

P
M

e
m
i
s
s
i
o
n
s
.

I
n

t
h
e

c
o
a
s
t
a
l

z
o
n
e
s

a
l
l

r
o
u
n
d

t
h
e

S
E
C
A

s
h
o
w

r
e
d
u
c
e
d

c
o
n
c
e
n
t
r
a
t
i
o
n
s

o
f

P
M

c
)

S
E
A

2

T
h
e

m
a
p

s
h
o
w
s

t
h
e

p
o
s
i
t
i
v
e

e
f
f
e
c
t

w
h
e
n

d
e
s
i
g
n
a
t
i
n
g

t
h
e

B
a
l
t
i
c

S
e
a
,

N
o
r
t
h

S
e
a

/

E
n
g
l
i
s
h

C
h
a
n
n
e
l
,

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

S
e
a

a
n
d

t
h
e

B
l
a
c
k

S
e
a

a
s

S
E
C
A
.

A
l
t
h
o
u
g
h

n
o
t

c
l
e
a
r
l
y

i
n
d
i
c
a
t
e
d

i
n

t
h
e

m
a
p

t
h
e

d
i
f
f
e
r
e
n
c
e
s

w
i
t
h

t
h
e

f
o
r
m
e
r

m
a
p

i
s

t
h
a
t

t
h
e

c
o
n
c
e
n
t
r
a
t
i
o
n
s

i
n

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

c
o
a
s
t
a
l

z
o
n
e
s

a
r
e

r
e
d
u
c
e
d
.

S
e
e

a
l
s
o

t
a
b
l
e

1
3

a
n
d

1
5

i
n

a
p
p
e
n
d
i
x

4

d
)

S
E
A

3

E
x
t
e
n
s
i
o
n

o
f

t
h
e

S
E
C
A

w
i
t
h

t
h
e

B
l
a
c
k

S
e
a

d
o
e
s

l
o
w
e
r

t
h
e

c
o
n
c
e
n
t
r
a
t
i
o
n

i
n

t
h
i
s

a
r
e
a
.
.


AEA 107


AEA 108
F
i
g
u
r
e

A
7

P
M
1
0

(
u
g
/
m
3
)

f
o
r

2
0
1
5

.

a
)

B
a
s
e
l
i
n
e

b
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A
1

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
e

n
e
x
t

m
a
p

s
h
o
w
s

t
h
e

p
o
s
i
t
i
v
e

e
f
f
e
c
t

o
n

P
M

c
o
n
c
e
n
t
r
a
t
i
o
n

w
h
e
n

d
e
s
i
g
n
a
t
i
n
g

t
h
e

N
o
r
t
h

S
e
a

/

E
n
g
l
i
s
h

C
h
a
n
n
e
l

a
n
d

T
h
e

B
a
l
t
i
c

a
s

S
E
C
A
.

T
h
e

r
e
d
u
c
t
i
o
n

o
f

s
u
l
f
u
r

i
n

f
u
e
l
s

i
s

a
s
s
o
c
i
a
t
e
d

w
i
t
h

a
n

r
e
d
u
c
t
i
o
n

i
n

P
M

e
m
i
s
s
i
o
n
s
.

c
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A
2

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
e

m
a
p

s
h
o
w
s

t
h
e

p
o
s
i
t
i
v
e

e
f
f
e
c
t

(
%

r
e
d
u
c
t
i
o
n

i
n

c
o
n
c
e
n
t
r
a
t
i
o
n
)

w
h
e
n

d
e
s
i
g
n
a
t
i
n
g

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

s
e
a

a
n
d

t
h
e

B
l
a
c
k

S
e
a

a
s

S
E
C
A
.

d
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A
3

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

E
x
t
e
n
s
i
o
n

o
f

t
h
e

S
E
C
A

w
i
t
h

t
h
e

B
l
a
c
k

S
e
a

r
e
d
u
c
e
s

t
h
e

P
M

c
o
n
c
e
n
t
r
a
t
i
o
n

i
n

t
h
i
s

a
r
e
a
.

S
e
e

a
l
s
o

t
a
b
l
e

1
3

a
n
d

1
5

i
n

a
p
p
e
n
d
i
x

4


AEA 109


AEA 110
F
i
g
u
r
e

A
8
:

A
v
e
r
a
g
e

A
c
c
u
m
u
l
a
t
e
d

E
x
c
e
e
d
e
n
c
e

f
o
r

a
c
i
d
i
f
i
c
a
t
i
o
n

(
e
q
.
/
h
a
/
y
r
)

f
o
r

2
0
2
0
.

.

a
)

B
a
s
e
l
i
n
e

I
n

a
l
l

a
r
e
a
s
,

n
o
t

c
o
l
o
r
e
d

g
r
e
e
n
,

t
h
e
r
e

i
s

a
n

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

b
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A
1

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
e

m
a
p

s
h
o
w
s

c
l
e
a
r
l
y

t
h
e

r
e
d
u
c
t
i
o
n

o
f

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

a
l
l

r
o
u
n
d

t
h
e

N
o
r
t
h

S
e
a

a
n
d

t
h
e

B
a
l
t
i
c

d
u
e

t
h
e

d
e
s
i
g
n
a
t
i
o
n

o
f

t
h
e
s
e

w
a
t
e
r
s

a
s

S
E
C
A
.

c
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A

3

c
o
m
p
a
r
e
d

t
o

S
E
A

1

T
h
u
s

t
h
e

e
f
f
e
c
t

o
f

i
n
c
l
u
s
i
o
n

o
f

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

s
e
a

a
n
d

B
l
a
c
k

s
e
a

i
n

t
h
e

S
E
C
A

d
o
e
s

a
d
d

o
n
l
y

t
o

a

s
m
a
l
l

e
x
t
e
n
d

t
o

t
h
e

i
m
p
r
o
v
e
m
e
n
t

g
a
i
n
e
d

i
n

s
c
e
n
a
r
i
o

S
E
A
1

c
o
m
p
a
r
e
d

t
o

t
h
e

b
a
s
e
l
i
n
e
.

d
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A

4

c
o
m
p
a
r
e
d

t
o

S
E
A

3

R
e
d
u
c
t
i
o
n

o
f

t
h
e

S

c
o
n
t
e
n
t

o
u
t
s
i
d
e

t
h
e

S
E
C
A
s

d
o
e
s

l
o
w
e
r

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

i
n

t
h
e

U
K

a
n
d

I
R
.

S
e
e

a
l
s
o

t
a
b
l
e

8

i
n

a
p
p
e
n
d
i
x

4


AEA 111


AEA 112

F
i
g
u
r
e

A
9

A
v
e
r
a
g
e

A
c
c
u
m
u
l
a
t
e
d

E
x
c
e
e
d
e
n
c
e

f
o
r

a
c
i
d
i
f
i
c
a
t
i
o
n

(
e
q
.
/
h
a
/
y
r
)

f
o
r

2
0
2
0
.

.

a
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

A
S
E
A

1

c
o
m
p
a
r
e
d

t
o

S
E
A

1

T
h
e

m
a
p

s
h
o
w
s

t
h
e

s
m
a
l
l

p
o
s
i
t
i
v
e

e
f
f
e
c
t

o
n

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

f
o
r

a
c
i
d
i
f
i
c
a
t
i
o
n

w
h
e
n

l
o
w
e
r
i
n
g

t
h
e

S

c
o
n
t
e
n
t

o
f

f
u
e
l

o
n

l
a
n
d

t
o

0
.
0
5
%
.

b
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

B
S
E
A

1

c
o
m
p
a
r
e
d

t
o

S
E
A

1

T
h
e

m
a
p

s
h
o
w
s

t
h
e

p
o
s
i
t
i
v
e

e
f
f
e
c
t

o
n

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

f
o
r

a
c
i
d
i
f
i
c
a
t
i
o
n

w
h
e
n

l
o
w
e
r
i
n
g

t
h
e

S

c
o
n
t
e
n
t

o
f

f
u
e
l

o
n

l
a
n
d

t
o

0
.
0
0
1
%
.

T
h
e
r
e

a
r
e

o
n
l
y

s
m
a
l
l

d
i
f
f
e
r
e
n
c
e

t
o

m
a
p

a

e
s
p
e
c
i
a
l
l
y

i
n

t
h
o
s
e

g
r
i
d

c
e
l
l
s

w
i
t
h

h
i
g
h

p
o
p
u
l
a
t
i
o
n

d
e
n
s
i
t
i
e
s
.

c
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

A
S
E
A

4

c
o
m
p
a
r
e
d

t
o

S
E
A

4

T
h
e

m
a
p

s
h
o
w
s

t
h
e

s
a
m
e

p
i
c
t
u
r
e

a
s

m
a
p

a
)
,

t
h
e

o
n
l
y

d
i
f
f
e
r
e
n
c
e

i
s

t
h
a
t

i
n

t
h
e
s
e

s
c
e
n
a
r
i
o
s

t
h
e

s
u
l
p
h
u
r

c
o
n
t
e
n
t

o
u
t
s
i
d
e

t
h
e

S
E
C
A

i
s

r
e
d
u
c
e
d

t
o

0
.
5

%
.

d
)

F
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

B
S
E
A

4

c
o
m
p
a
r
e
d

t
o

S
E
A

4

T
h
e

m
a
p

s
h
o
w
s

t
h
e

s
a
m
e

p
i
c
t
u
r
e

a
s

m
a
p

b
)
,

t
h
e

o
n
l
y

d
i
f
f
e
r
e
n
c
e

i
s

t
h
a
t

i
n

t
h
e
s
e

s
c
e
n
a
r
i
o
s

t
h
e

s
u
l
p
h
u
r

c
o
n
t
e
n
t

o
u
t
s
i
d
e

t
h
e

S
E
C
A

i
s

r
e
d
u
c
e
d

t
o

0
.
5

)
.

T
h
e

g
e
n
e
r
a
l

c
o
n
c
l
u
s
i
o
n

o
f

t
h
e
s
e

m
a
p
s

i
s

t
h
a
t

t
h
e

e
f
f
e
c
t

o
n

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

f
o
r

a
c
i
d
i
f
i
c
a
t
i
o
n

w
h
e
n

l
o
w
e
r
i
n
g

t
h
e

S

c
o
n
t
e
n
t

o
f

f
u
e
l

o
n

l
a
n
d

i
s

m
a
r
g
i
n
a
l

a
p
p
r
o
x
i
m
a
t
e
l
y

1
.
2

t
o

2
.
1

%

r
e
d
u
c
t
i
o
n
.

B
e
c
a
u
s
e

a
c
i
d
i
f
i
c
a
t
i
o
n

i
s

t
h
e

p
a
r
a
m
e
t
e
r

e
f
f
e
c
t
e
d

t
h
e

m
o
s
t

b
y

s
u
l
f
u
r

e
m
i
s
s
i
o
n
s

t
h
i
s

a
n
a
l
y
s
i
s

i
s

n
o
t

p
e
r
f
o
r
m
e
d

f
o
r

t
h
e

o
t
h
e
r

p
a
r
a
m
e
t
e
r
s

S
e
e

a
l
s
o

t
a
b
l
e

9

i
n

a
p
p
e
n
d
i
x

4


AEA 113


AEA 114

F
i
g
u
r
e

A
1
0
:

A
v
e
r
a
g
e

A
c
c
u
m
u
l
a
t
e
d

E
x
c
e
e
d
e
n
c
e

f
o
r

e
u
t
r
o
p
h
i
c
a
t
i
o
n

(
e
q
.
/
h
a
/
y
r
)

f
o
r

2
0
2
0
.

.

a
)

B
a
s
e
l
i
n
e

I
n

a
l
l

a
r
e
a
s
,

n
o
t

c
o
l
o
r
e
d

g
r
e
e
n
,

t
h
e
r
e

i
s

a
n

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

b
)

f
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A
1

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
e

m
a
p

s
h
o
w
s

c
l
e
a
r
l
y

t
h
e

r
e
d
u
c
t
i
o
n

o
f

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

a
l
l

r
o
u
n
d

t
h
e

N
o
r
t
h

S
e
a

a
n
d

t
h
e

B
a
l
t
i
c

d
u
e

t
h
e

d
e
s
i
g
n
a
t
i
o
n

o
f

t
h
e
s
e

w
a
t
e
r
s

a
s

S
E
C
A
.

A
l
s
o

t
h
e

e
x
c
e
e
d
e
n
c
e

i
n

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

c
o
u
n
t
r
i
e
s

i
m
p
r
o
v
e
s

a
s

a

r
e
s
u
l
t

o
f

t
h
e

N
O
X

T
i
e
r

s
t
a
n
d
a
r
d
s
.

c
)

f
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A

3

c
o
m
p
a
r
e
d

t
o

S
E
A

1

T
h
u
s

t
h
e

e
f
f
e
c
t

o
f

i
n
c
l
u
s
i
o
n

o
f

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

s
e
a

a
n
d

B
l
a
c
k

s
e
a

i
n

t
h
e

S
E
C
A

a
n
d

d
e
s
i
g
n
a
t
i
o
n

t
h
e
s
e

w
a
t
e
r
s

a
s

N
E
C
A

d
o
e
s

a
d
d

t
o

t
h
e

i
m
p
r
o
v
e
m
e
n
t

g
a
i
n
e
d

i
n

s
c
e
n
a
r
i
o

S
E
A
1

c
o
m
p
a
r
e
d

t
o

t
h
e

b
a
s
e
l
i
n
e
.

d
)

f
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A

4

c
o
m
p
a
r
e
d

t
o

S
E
A

3

R
e
d
u
c
t
i
o
n

o
f

t
h
e

S

c
o
n
t
e
n
t

o
u
t
s
i
d
e

t
h
e

S
E
C
A
s

d
o
e
s

l
o
w
e
r

t
h
e

e
x
c
e
e
d
e
n
c
e

o
f

t
h
e

c
r
i
t
i
c
a
l

l
o
a
d

o
n
l
y

t
o

a

s
m
a
l
l

e
x
t
e
n
d
.

T
h
e

r
e
m
a
i
n
d
e
r

r
e
d
u
c
t
i
o
n

i
n

S
E
A
4

s
t
e
m
s

f
r
o
m

t
h
e

N
E
C
A

i
n

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

a
n
d

t
h
e

B
l
a
c
k

S
e
a

S
e
e

a
l
s
o

t
a
b
l
e

1
1

i
n

a
p
p
e
n
d
i
x

4


AEA 115


AEA 116

F
i
g
u
r
e

A
1
1

P
M
1
0

(
u
g
/
m
3
)

f
o
r

2
0
2
0
.

.

a
)

B
a
s
e
l
i
n
e

b
)

S
E
A

1

T
h
e

m
a
p

s
h
o
w
s

t
h
e

p
o
s
i
t
i
v
e

e
f
f
e
c
t

o
n

P
M

c
o
n
c
e
n
t
r
a
t
i
o
n

w
h
e
n

d
e
s
i
g
n
a
t
i
n
g

t
h
e

d
e
s
i
g
n
a
t
i
n
g

t
h
e

N
o
r
t
h

S
e
a

/

E
n
g
l
i
s
h

C
h
a
n
n
e
l

a
n
d

T
h
e

B
a
l
t
i
c

a
s

S
E
C
A
.

T
h
e

r
e
d
u
c
t
i
o
n

o
f

s
u
l
f
u
r

i
n

f
u
e
l
s

i
s

a
s
s
o
c
i
a
t
e
d

w
i
t
h

a
n

r
e
d
u
c
t
i
o
n

i
n

P
M

e
m
i
s
s
i
o
n
s
.

I
n

t
h
e

c
o
a
s
t
a
l

z
o
n
e
s

a
l
l

r
o
u
n
d

t
h
e

S
E
C
A

s
h
o
w

r
e
d
u
c
e
d

c
o
n
c
e
n
t
r
a
t
i
o
n
s

o
f

P
M

c
)

S
E
A

3

T
h
e

m
a
p

s
h
o
w
s

t
h
e

p
o
s
i
t
i
v
e

e
f
f
e
c
t

w
h
e
n

d
e
s
i
g
n
a
t
i
n
g

t
h
e

B
a
l
t
i
c

S
e
a
,

N
o
r
t
h

S
e
a

/

E
n
g
l
i
s
h

C
h
a
n
n
e
l
,

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

S
e
a

a
n
d

t
h
e

B
l
a
c
k

S
e
a

a
s

S
E
C
A
.

A
l
t
h
o
u
g
h

n
o
t

c
l
e
a
r
l
y

i
n
d
i
c
a
t
e
d

i
n

t
h
e

m
a
p

t
h
e

d
i
f
f
e
r
e
n
c
e
s

w
i
t
h

t
h
e

f
o
r
m
e
r

m
a
p

i
s

t
h
a
t

t
h
e

c
o
n
c
e
n
t
r
a
t
i
o
n
s

i
n

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

c
o
a
s
t
a
l

z
o
n
e
s

a
r
e

r
e
d
u
c
e
d
.

S
e
e

a
l
s
o

t
a
b
l
e

1
4

i
n

a
p
p
e
n
d
i
x

4

d
)

S
E
A

5

T
h
e

m
a
p

s
h
o
w
s

t
h
a
t

r
e
d
u
c
i
n
g

t
h
e

S

c
o
n
t
e
n
t

o
u
t
s
i
d
e

t
h
e

S
E
C
A

h
a
s

a

p
o
s
i
t
i
v
e

e
f
f
e
c
t

o
n

P
M
1
0

p
a
r
t
i
c
u
l
a
r
l
y

i
n

t
h
e

U
K

a
n
d

I
R

(
c
o
m
p
a
r
e

t
o

m
a
p

b
)
.

F
o
r

m
o
r
e

d
e
t
a
i
l
s

s
e
e

a
l
s
o

t
a
b
l
e

1
4

i
n

a
p
p
e
n
d
i
x

4


AEA 117

AEA 118

F
i
g
u
r
e

A
1
2

P
M
1
0

(
u
g
/
m
3
)

f
o
r

2
0
2
0
.

.

a
)

B
a
s
e
l
i
n
e

b
)

f
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A

1

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
e

n
e
x
t

m
a
p

s
h
o
w
s

t
h
e

p
o
s
i
t
i
v
e

e
f
f
e
c
t

o
n

P
M

c
o
n
c
e
n
t
r
a
t
i
o
n

w
h
e
n

d
e
s
i
g
n
a
t
i
n
g

t
h
e

d
e
s
i
g
n
a
t
i
n
g

t
h
e

N
o
r
t
h

S
e
a

/

E
n
g
l
i
s
h

C
h
a
n
n
e
l

a
n
d

T
h
e

B
a
l
t
i
c

a
s

S
E
C
A
.

T
h
e

r
e
d
u
c
t
i
o
n

o
f

s
u
l
f
u
r

i
n

f
u
e
l
s

i
s

a
s
s
o
c
i
a
t
e
d

w
i
t
h

a
n

r
e
d
u
c
t
i
o
n

i
n

P
M

e
m
i
s
s
i
o
n
s
.
.

c
)

f
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A

3

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
e

m
a
p

s
h
o
w
s

t
h
e

p
o
s
i
t
i
v
e

e
f
f
e
c
t

(
%

r
e
d
u
c
t
i
o
n

i
n

c
o
n
c
e
n
t
r
a
t
i
o
n
)

w
h
e
n

d
e
s
i
g
n
a
t
i
n
g

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

s
e
a

a
n
d

t
h
e

B
l
a
c
k

S
e
a

a
s

S
E
C
A

a
n
d

t
h
e

B
a
l
t
i
c

S
e
a
,

N
o
r
t
h

S
e
a

/

E
n
g
l
i
s
h

C
h
a
n
n
e
l

a
n
d

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

S
e
a

a
s

N
E
C
A
.

d
)

f
r
a
c
t
i
o
n
a
l

c
h
a
n
g
e

(
%
)

o
f

S
E
A

4

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
e

m
a
p

s
h
o
w
s

t
h
a
t

r
e
d
u
c
i
n
g

t
h
e

S

c
o
n
t
e
n
t

o
u
t
s
i
d
e

t
h
e

S
E
C
A

(
M
e
d
i
t
e
r
r
a
n
e
a
n

S
e
a

a
n
d

B
l
a
c
k

S
e
a

a
r
e

i
n

t
h
i
s

s
c
e
n
a
r
i
o

n
o
t

S
E
C
A
)

a
l
s
o

h
a
s

a

p
o
s
i
t
i
v
e

e
f
f
e
c
t

o
n

P
M
1
0

p
a
r
t
i
c
u
l
a
r
l
y

i
n

t
h
e

U
K

a
n
d

I
R

(
c
o
m
p
a
r
e

t
o

m
a
p

b
)
.

S
e
e

a
l
s
o

t
a
b
l
e

1
4

i
n

a
p
p
e
n
d
i
x

4


AEA 119


AEA 120

F
i
g
u
r
e

A
1
3

A
O
T
4
0
F

(
p
p
m
.
h
)

f
o
r

2
0
2
0
.

.

a
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

A
S
E
A

1

c
o
m
p
a
r
e
d

t
o

S
E
A

1

;

b
)

d
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

B
S
E
A

1

c
o
m
p
a
r
e
d

t
o

S
E
A

1

T
h
e

m
a
p

s
h
o
w
s

c
l
e
a
r
l
y

t
h
e

r
e
d
u
c
t
i
o
n

o
f

t
h
e

o
z
o
n
e

c
o
n
c
e
n
t
r
a
t
i
o
n

a
s

a
n

r
e
s
u
l
t

o
f

t
h
e

r
e
d
u
c
e
d

N
O
x

e
m
i
s
s
i
o
n
s

i
n

a
l
l

s
e
a
s

a
s

a

r
e
s
u
l
t

o
f

t
h
e

T
i
e
r

s
t
a
n
d
a
r
d
s
.

c
)

d
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

A
S
E
A

4

c
o
m
p
a
r
e
d

t
o

S
E
A

4

T
h
u
s

t
h
e

e
f
f
e
c
t

o
f

i
n
c
l
u
s
i
o
n

o
f

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

S
e
a

a
n
d

B
l
a
c
k

s
e
a

i
n

t
h
e

S
E
C
A

d
o
e
s

n
o
t

a
d
d

t
o

t
h
e

i
m
p
r
o
v
e
m
e
n
t

g
a
i
n
e
d

i
n

s
c
e
n
a
r
i
o

S
E
A
1

c
o
m
p
a
r
e
d

t
o

t
h
e

b
a
s
e
l
i
n
e
.

d
)

d
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

B
S
E
A

4

c
o
m
p
a
r
e
d

t
o

S
E
A

4

R
e
d
u
c
t
i
o
n

o
f

t
h
e

S

c
o
n
t
e
n
t

o
u
t
s
i
d
e

t
h
e

S
E
C
A
s

d
o
e
s

n
o
t

a
d
d

t
o

a

r
e
d
u
c
t
i
o
n

o
f

o
z
o
n
e

c
o
m
p
a
r
e
d

t
o

S
E
A
1
.

S
E
A
1

i
s

c
o
m
p
a
r
a
b
l
e

t
o

S
E
A
4
.

A
l

t
h
e

m
a
p
s

s
h
o
w

l
i
t
t
l
e

d
i
f
f
e
r
e
n
c
e
s

a
s

o
n
e

c
o
u
l
d

e
x
p
e
c
t

b
e
c
a
u
s
e

t
h
e

d
i
f
f
e
r
e
n
c
e
s

s
h
o
w
n

i
n

t
h
e

m
a
p
s

a
r
e

t
h
e

r
e
s
u
l
t

o
f

l
o
w
e
r
i
n
g

t
h
e

S

c
o
n
t
e
n
t

i
n

f
u
e
l

u
s
e
d

o
n

l
a
n
d
.

B
e
c
a
u
s
e

t
h
e

d
i
f
f
e
r
e
n
c
e
s

a
r
e

t
h
a
t

s
m
a
l
l

p
l
e
a
s
e

s
e
e

a
l
s
o

2
4

a
n
d

2
7

i
n

a
p
p
e
n
d
i
x

4


AEA 121

AEA 122

F
i
g
u
r
e

A
1
4

A
O
T
4
0
F

(
p
p
m
.
h
)

f
o
r

2
0
2
0
.

.

a
)

B
a
s
e
l
i
n
e

b
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

S
E
A
1

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
e

m
a
p

s
h
o
w
s

c
l
e
a
r
l
y

t
h
e

r
e
d
u
c
t
i
o
n

o
f

t
h
e

o
z
o
n
e

c
o
n
c
e
n
t
r
a
t
i
o
n

a
s

a
n

r
e
s
u
l
t

o
f

t
h
e

r
e
d
u
c
e
d

N
O
x

e
m
i
s
s
i
o
n
s

i
n

a
l
l

s
e
a
s

a
s

a

r
e
s
u
l
t

o
f

t
h
e

T
i
e
r

s
t
a
n
d
a
r
d
s
.

c
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

S
E
A
3

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
u
s

t
h
e

e
f
f
e
c
t

o
f

i
n
c
l
u
s
i
o
n

o
f

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

S
e
a

a
n
d

B
l
a
c
k

s
e
a

i
n

t
h
e

S
E
C
A

d
o
e
s

n
o
t

a
d
d

t
o

t
h
e

i
m
p
r
o
v
e
m
e
n
t

g
a
i
n
e
d

i
n

s
c
e
n
a
r
i
o

S
E
A
1

c
o
m
p
a
r
e
d

t
o

t
h
e

b
a
s
e
l
i
n
e
.

d
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

S
E
A
4

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

R
e
d
u
c
t
i
o
n

o
f

t
h
e

S

c
o
n
t
e
n
t

o
u
t
s
i
d
e

t
h
e

S
E
C
A
s

d
o
e
s

n
o
t

a
d
d

t
o

a

r
e
d
u
c
t
i
o
n

o
f

o
z
o
n
e

c
o
m
p
a
r
e
d

t
o

S
E
A
1
.

S
E
A
1

i
s

c
o
m
p
a
r
a
b
l
e

t
o

S
E
A
4
.

S
e
e

a
l
s
o

t
a
b
l
e

2
3

i
n

a
p
p
e
n
d
i
x

4


AEA 123


AEA 124

F
i
g
u
r
e

A
1
5

S
O
M
O
3
5

(
p
p
m
.
h
)

f
o
r

2
0
2
0
.

.

a
)

B
a
s
e
l
i
n
e

b
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

S
E
A

1

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
e

m
a
p

s
h
o
w
s

c
l
e
a
r
l
y

t
h
e

r
e
d
u
c
t
i
o
n

o
f

t
h
e

o
z
o
n
e

c
o
n
c
e
n
t
r
a
t
i
o
n

a
s

a
n

r
e
s
u
l
t

o
f

t
h
e

r
e
d
u
c
e
d

N
O
x

e
m
i
s
s
i
o
n
s

i
n

a
l
l

s
e
a
s

a
s

a

r
e
s
u
l
t

o
f

t
h
e

T
i
e
r

s
t
a
n
d
a
r
d
s
.

c
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

S
E
A

3

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

T
h
u
s

t
h
e

e
f
f
e
c
t

o
f

i
n
c
l
u
s
i
o
n

o
f

t
h
e

M
e
d
i
t
e
r
r
a
n
e
a
n

s
e
a

a
n
d

B
l
a
c
k

s
e
a

i
n

t
h
e

S
E
C
A

d
o
e
s

n
o
t

a
d
d

t
o

t
h
e

i
m
p
r
o
v
e
m
e
n
t

g
a
i
n
e
d

i
n

s
c
e
n
a
r
i
o

S
E
A
1

c
o
m
p
a
r
e
d

t
o

t
h
e

b
a
s
e
l
i
n
e
.

d
)

D
i
f
f
e
r
e
n
c
e

(
p
p
m
.
h
)

o
f

S
E
A

4

c
o
m
p
a
r
e
d

t
o

b
a
s
e
l
i
n
e

R
e
d
u
c
t
i
o
n

o
f

t
h
e

S

c
o
n
t
e
n
t

o
u
t
s
i
d
e

t
h
e

S
E
C
A
s

d
o
e
s

n
o
t

a
d
d

t
o

a

r
e
d
u
c
t
i
o
n

o
f

o
z
o
n
e

c
o
m
p
a
r
e
d

t
o

S
E
A
1
.

S
E
A
1

i
s

c
o
m
p
a
r
a
b
l
e

t
o

S
E
A
4
.

S
e
e

a
l
s
o

t
a
b
l
e

1
7

i
n

a
p
p
e
n
d
i
x

4


AEA 125
16 Appendix 4 - Results per EU member
states
In this appendix the results of the dispersion modelling are aggregated to the level of the
individual countries. In order to make the results more comparable the rough results were
multiplied by either the area of the country or the number of inhabitants.

For each parameter first the result for 2015 qnd 2020 are presented. A third table shows the
absolute and percentage difference between different scenarios and the trend between 2015 and
2020 on the level of total EU 27+2.

Table 1: Sulphur deposition in 2015 (Meq/year)

area
(km2)
base SEA1 SEA2 SEA3 ASEA1 BSEA1 A B
AT Austria 83756 913 877 820 819 869 851 904 887
BE Belgium 30520 1028 913 908 908 880 851 995 966
BG Bulgaria 110857 2061 2062 1944 1912 2056 2050 2055 2049
CH Switzerland 41084 434 422 392 392 418 414 430 426
CY Cyprus 5120 48 48 39 39 48 48 47 47
CZ Czech republic 78600 1216 1167 1151 1150 1159 1149 1207 1198
DE Germany 355693 5826 5161 5100 5100 5054 4933 5719 5598
DK Denmark 44270 728 334 331 331 331 326 725 720
EE Estonia 45263 445 313 310 310 311 310 444 442
ES Spain 497559 6749 6878 6085 6085 6853 6831 6724 6702
FI Finland 334395 1617 1258 1251 1251 1255 1244 1613 1603
FR France 547219 5928 5433 4914 4914 5316 5215 5810 5709
GR Greece 132022 2178 2236 1459 1452 2210 2188 2152 2130
HU Hungary 92857 1189 1173 1120 1119 1169 1163 1185 1180
IE Ireland 68128 781 745 743 743 732 721 768 757
IT Italy 301129 4358 4465 2897 2896 4430 4399 4323 4293
LT Lithuania 64612 866 785 778 777 783 780 863 860
LU Luxembourg 2589 45 43 42 42 41 39 43 41
LV Latvia 64207 658 543 536 535 540 537 656 653
MT Malta 319 6 6 1 1 6 6 6 6
NL Netherlands 35076 869 650 647 647 637 624 855 843
NO Norway 295987 1845 1386 1377 1377 1381 1375 1839 1832
PO Poland 310641 6797 6448 6395 6393 6407 6366 6756 6715
PT Portugal 88648 1329 1366 1320 1320 1360 1356 1324 1319
RO Romania 237394 3821 3791 3656 3615 3778 3764 3807 3794
SE Sweden 446266 2893 1858 1845 1845 1848 1832 2880 2865
SK Slovakia 48857 805 789 771 770 786 782 802 798
SL Slovenia 20672 377 378 320 320 374 371 373 370
UK
United
kingdom
243251 3886 3185 3172 3172 3167 3145 3867 3845

EU27+2 4626991 59697 54714 50323 50237 54197 53670 59174 58646


AEA 126
Table 2: Sulphur deposition in 2020 (Meq/year)

Area
(km2)
base SEA1 SEA3 SEA4 SEA5 ASEA1 BSEA1 ASEA4 BSEA4
AT Austria 83756 897 857 791 790 780 849 838 782 771
BE Belgium 30520 1002 872 866 852 852 843 824 824 805
BG Bulgaria 110857 2033 2033 1864 1885 1861 2028 2023 1879 1875
CH Switzerland 41084 427 414 381 376 371 410 408 372 370
CY Cyprus 5120 40 41 31 32 31 41 41 32 32
CZ
Czech
republic
78600 1178 1122 1104 1098 1095 1115 1110 1090 1085
DE Germany 355693 5914 5160 5090 5030 5020 5068 5007 4939 4877
DK Denmark 44270 782 335 332 323 323 333 329 321 318
EE Estonia 45263 456 306 303 300 299 305 304 299 297
ES Spain 497559 5569 5716 4817 3912 3784 5697 5684 3893 3880
FI Finland 334395 1641 1234 1227 1209 1208 1232 1225 1206 1199
FR France 547219 5729 5169 4582 4262 4179 5072 5007 4165 4100
GR Greece 132022 2177 2243 1356 1479 1352 2219 2202 1455 1438
HU Hungary 92857 1145 1126 1065 1069 1060 1122 1119 1065 1062
IE Ireland 68128 772 731 729 629 629 719 711 618 609
IT Italy 301129 4446 4568 2792 3011 2756 4534 4511 2977 2954
LT Lithuania 64612 883 792 783 779 778 790 788 777 775
LU Luxembourg 2589 44 41 40 40 39 39 38 38 36
LV Latvia 64207 683 552 543 539 538 549 548 536 535
MT Malta 319 7 7 1 2 1 7 7 2 2
NL Netherlands 35076 889 641 638 627 627 630 622 616 608
NO Norway 295987 1904 1386 1375 1277 1275 1381 1377 1272 1268
PO Poland 310641 6718 6324 6261 6243 6234 6287 6261 6206 6180
PT Portugal 88648 1177 1218 1166 775 767 1213 1210 770 766
RO Romania 237394 3880 3846 3647 3667 3640 3833 3824 3655 3645
SE Sweden 446266 3040 1870 1854 1805 1803 1861 1851 1796 1786
SK Slovakia 48857 800 782 761 761 758 780 777 758 755
SL Slovenia 20672 380 381 315 322 313 377 375 319 316
UK
United
kingdom
243251 3783 2991 2976 2542 2540 2975 2961 2526 2512

EU27+2 4626991 58396 52759 47691 45635 44912 52310 51981 45186 44858


AEA 127
Table 3: Differences between scenarios and years for sulphur depositions (EU27+2 total)

2015 2020
abs
(Meq/yr)
%
abs
(Meq/yr
)
%
SEA1-base -4983 -8.3% -5638 -10%
SEA2-base -9374 -16% n.c n.c
SEA3-base -9460 -16% -10706 -18%
SEA4-base n.c n.c -12761 -22%
SEA5-base n.c n.c -13484 -23%
BSEA1-
ASEA1
-527 -1.0% -329 -0.6%
ASEA1-SEA1 -517 -0.9% -449 -0.9%
BSEA1-SEA1 -1044 -1.9% -778 -1.5%
A-SEA4-
ASEA1
n.c n.c -7123 -14%
BSEA4-
BSEA1
n.c n.c -7123 -14%
A-base -522 -0.9% n.c n.c
B-base -1050 -1.8% n.c n.c
B-A -528 -0.9% n.c n.c
SEA2-SEA1 -4391 -8.0% n.c n.c
SEA3-SEA2 -86 -0.2% n.c n.c
SEA3-SEA1 n.c n.c -5068 -11%
SEA4-SEA1 n.c n.c -7124 -14%
SEA5-SEA3 n.c n.c -2779 -5.8%
SEA5-SEA4 n.c n.c -723 -1.6%
ASEA1-base -5500 -9.2% n.c n.c
BSEA1-base -6027 -10% n.c n.c
n.c - not calculated

AEA 128
Table 4: Nitrogen deposition in 2015 (Meq/year)

area
(km2)
AT Austria 83756 4160 4127 4115 4115 4127 4123 4159 4156
BE Belgium 30520 3076 3040 3039 3039 3036 3032 3073 3071
BG Bulgaria 110857 4128 4097 4071 4065 4097 4096 4127 4127
CH Switzerland 41084 2021 2008 2000 2000 2008 2007 2021 2021
CY Cyprus 5120 132 131 131 131 131 131 132 132
CZ
Czech
republic
78600 4256 4230 4224 4224 4229 4228 4255 4254
DE Germany 355693 23806 23592 23573 23573 23576 23557 23789 23769
DK Denmark 44270 2099 2059 2058 2058 2059 2059 2099 2099
EE Estonia 45263 818 791 789 789 790 790 817 817
ES Spain 497559 18212 17864 17698 17698 17860 17857 18208 18205
FI Finland 334395 3180 3068 3063 3063 3066 3065 3179 3177
FR France 547219 28449 28101 28009 28009 28075 28052 28418 28393
GR Greece 132022 3740 3605 3511 3510 3603 3602 3738 3736
HU Hungary 92857 4053 4030 4018 4018 4029 4028 4052 4051
IE Ireland 68128 3624 3603 3602 3602 3602 3601 3623 3621
IT Italy 301129 16276 16006 15681 15681 15994 15984 16263 16253
LT Lithuania 64612 2163 2143 2139 2139 2143 2142 2163 2163
LU Luxembourg 2589 198 196 196 196 195 195 197 196
LV Latvia 64207 1468 1431 1426 1426 1430 1430 1467 1467
MT Malta 319 6 6 5 5 6 6 6 6
NL Netherlands 35076 3819 3772 3770 3770 3769 3766 3817 3814
NO Norway 295987 4183 4025 4018 4018 4023 4021 4181 4180
PO Poland 310641 15721 15602 15580 15580 15599 15596 15718 15714
PT Portugal 88648 3305 3229 3213 3213 3228 3228 3304 3304
RO Romania 237394 9751 9697 9663 9658 9695 9693 9749 9747
SE Sweden 446266 5821 5546 5537 5537 5544 5541 5818 5816
SK Slovakia 48857 2124 2108 2100 2100 2107 2106 2123 2123
SL Slovenia 20672 1248 1236 1225 1225 1235 1235 1248 1248
UK
United
kingdom
243251 11978 11759 11747 11747 11756 11753 11976 11973

EU27+2 4626991 183814 181100 180202 180188 181012 180921 183722 183633

129

Table 5: Nitrogen deposition in 2020 (Meq/year)

area
(km2)
AT Austria 83756 3974 3918 3909 3904 3902 3918 3917 3904 3902
BE Belgium 30520 2949 2895 2894 2890 2889 2891 2888 2886 2883
BG Bulgaria 110857 4080 4010 3977 3981 3974 4010 4010 3980 3980
CH Switzerland 41084 1913 1890 1884 1882 1880 1890 1890 1882 1882
CY Cyprus 5120 132 129 128 128 128 129 129 128 128
CZ Czech republic 78600 4070 4030 4025 4021 4020 4030 4029 4020 4019
DE Germany 355693 23205 22886 22869 22848 22844 22874 22865 22835 22826
DK Denmark 44270 2022 1937 1936 1934 1934 1937 1938 1934 1935
EE Estonia 45263 811 761 759 757 757 761 761 757 757
ES Spain 497559 17062 16409 16244 16141 16104 16406 16404 16137 16135
FI Finland 334395 3084 2899 2894 2886 2885 2898 2897 2884 2883
FR France 547219 27267 26678 26598 26578 26560 26658 26646 26559 26546
GR Greece 132022 3695 3382 3288 3304 3284 3381 3380 3302 3300
HU Hungary 92857 3991 3950 3938 3936 3934 3949 3949 3936 3935
IE Ireland 68128 3562 3527 3526 3523 3522 3526 3525 3521 3521
IT Italy 301129 15634 15019 14702 14748 14680 15009 15001 14734 14725
LT Lithuania 64612 2156 2118 2114 2112 2111 2118 2118 2112 2112
LU Luxembourg 2589 186 184 183 183 183 183 182 183 182
LV Latvia 64207 1439 1375 1370 1367 1366 1374 1374 1366 1366
MT Malta 319 6 5 5 5 5 5 5 5 5
NL Netherlands 35076 3832 3758 3757 3755 3755 3756 3754 3753 3752
NO Norway 295987 4081 3816 3809 3789 3788 3815 3814 3787 3786
PO Poland 310641 15292 15098 15075 15063 15059 15095 15094 15060 15059
PT Portugal 88648 3235 3112 3099 3055 3052 3112 3111 3055 3054
RO Romania 237394 9695 9592 9553 9554 9546 9591 9590 9552 9551
SE Sweden 446266 5695 5222 5213 5190 5188 5221 5219 5188 5186
SK Slovakia 48857 2034 2008 2001 1999 1998 2008 2007 1999 1998
SL Slovenia 20672 1192 1165 1156 1156 1154 1165 1165 1156 1156
UK United kingdom 243251 11379 11027 11015 10998 10996 11025 11023 10996 10995

EU27+2 4626991 177674 172803 171922 171688 171499 172734 172685 171611 171559

130

Table 6: Differences between scenarios and years for nitrogen depositions (EU27+2 total)

2015 2020
abs
(Meq/yr)
%
abs
(Meq/yr)
%
SEA1-base -2714 -1.5% -4871 -2.7%
SEA2-base -3613 -2.0% n.c n.c
SEA3-base -3626 -2.0% -5752 -3.2%
SEA4-base n.c n.c -5986 -3.4%
SEA5-base n.c n.c -6175 -3.5%
BSEA1-
ASEA1
-90 0.0% -49 0.0%
ASEA1-SEA1 -88 0.0% -69 0.0%
BSEA1-SEA1 -179 -0.1% -118 -0.1%
A-SEA4-
ASEA1
n.c n.c
-1123 -0.7%
BSEA4-
BSEA1
n.c n.c
-1126 -0.7%
A-base -92 -0.1% n.c n.c
B-base -182 -0.1% n.c n.c
B-A -90 0.0% n.c n.c
SEA2-SEA1 -899 -0.5% n.c n.c
SEA3-SEA2 -14 0.0% n.c n.c
SEA3-SEA1 n.c n.c -881 -0.5%
SEA4-SEA1 n.c n.c -1115 -0.6%
SEA5-SEA3 n.c n.c -423 -0.2%
SEA5-SEA4 n.c n.c -189 -0.1%
BSEA1-base -2893 -1.6% n.c n.c

131

Table 7: Acidification (AAE) in 2015 (Meq/year)

area
(km2)
AT Austria 83756 1 1 1 1 1 1 1 1
BE Belgium 30520 28 21 20 20 20 19 26 25
BG Bulgaria 110857 0 0 0 0 0 0 0 0
CH Switzerland 41084 3 3 2 2 3 3 3 3
CY Cyprus 5120 0 0 0 0 0 0 0 0
CZ
Czech
republic
78600 5 4 4 4 4 4 5 5
DE Germany 355693 558 474 471 471 464 452 547 534
DK Denmark 44270 3 1 1 1 1 1 3 3
EE Estonia 45263 0 0 0 0 0 0 0 0
ES Spain 497559 0 0 0 0 0 0 0 0
FI Finland 334395 12 6 6 6 6 6 12 11
FR France num 61 55 55 55 53 51 59 56
GR Greece 132022 0 0 0 0 0 0 0 0
HU Hungary 92857 0 0 0 0 0 0 0 0
IE Ireland 68128 6 6 5 5 5 5 6 6
IT Italy 301129 0 0 0 0 0 0 0 0
LT Lithuania 64612 0 0 0 0 0 0 0 0
LU Luxembourg 2589 0 0 0 0 0 0 0 0
LV Latvia 64207 0 0 0 0 0 0 0 0
MT Malta 319 0 0 0 0 0 0 0 0
NL Netherlands 35076 214 184 183 183 182 180 212 210
NO Norway 295987 422 270 267 267 268 267 421 419
PO Poland 310641 5 5 5 5 5 5 5 5
PT Portugal 88648 6 6 6 6 6 6 6 6
RO Romania 237394 0 0 0 0 0 0 0 0
SE Sweden 446266 220 113 112 112 112 111 219 217
SK Slovakia 48857 2 2 2 2 2 2 2 2
SL Slovenia 20672 0 0 0 0 0 0 0 0
UK
United
kingdom
243251 187 158 157 157 157 156 186 185

EU27+2 4626991 1734 1305 1296 1296 1287 1266 1712 1686

132

Table 8: Acidification (AAE) in 2020 (Meq/year)

area
(km2)
AT Austria 83756 1 0 0 0 0 0 0 0 0
BE Belgium 30520 24 17 17 16 16 16 15 15 15
BG Bulgaria 110857 0 0 0 0 0 0 0 0 0
CH Switzerland 41084 2 2 2 2 2 2 2 2 2
CY Cyprus 5120 0 0 0 0 0 0 0 0 0
CZ
Czech
republic
78600 2 2 1 1 1 2 2 1 1
DE Germany 355693 551 448 446 442 442 441 436 434 429
DK Denmark 44270 4 0 0 0 0 0 0 0 0
EE Estonia 45263 0 0 0 0 0 0 0 0 0
ES Spain 497559 0 0 0 0 0 0 0 0 0
FI Finland 334395 12 5 5 5 5 5 5 5 5
FR France 547219 45 41 40 39 39 39 38 38 37
GR Greece 132022 0 0 0 0 0 0 0 0 0
HU Hungary 92857 0 0 0 0 0 0 0 0 0
IE Ireland 68128 5 5 5 4 4 5 5 4 4
IT Italy 301129 0 0 0 0 0 0 0 0 0
LT Lithuania 64612 0 0 0 0 0 0 0 0 0
LU Luxembourg 2589 0 0 0 0 0 0 0 0 0
LV Latvia 64207 0 0 0 0 0 0 0 0 0
MT Malta 319 0 0 0 0 0 0 0 0 0
NL Netherlands 35076 223 183 182 181 181 181 180 179 178
NO Norway 295987 423 239 235 223 223 238 237 222 221
PO Poland 310641 4 3 3 3 3 3 3 3 3
PT Portugal 88648 3 3 3 0 0 3 3 0 0
RO Romania 237394 0 0 0 0 0 0 0 0 0
SE Sweden 446266 235 107 106 103 103 106 105 102 101
SK Slovakia 48857 1 1 1 1 1 1 1 1 1
SL Slovenia 20672 0 0 0 0 0 0 0 0 0
UK
United
kingdom
243251 137 108 107 90 90 107 107 90 89

EU27+2 4626991 1673 1165 1155 1112 1111 1151 1140 1098 1087

133

Table 9: Differences between scenarios and years for acidification (AAE) (EU27+2 total)

2015 2020
abs
(Meq/yr)
%
abs
(Meq/yr)
%
SEA1-base -429 -25% -508 -30%
SEA2-base -438 -25% n.c n.c
SEA3-base -438 -25% -518 -31%
SEA4-base n.c n.c -561 -34%
BSEA1-
ASEA1
-21 -1.7% -11 -0.9%
ASEA1-SEA1 -19 -1.4% -14 -1.2%
BSEA1-SEA1 -40 -3.1% -25 -2.1%
A-SEA4-
ASEA1
n.c n.c
-53 -4.6%
BSEA4-
BSEA1
n.c n.c
-53 -4.6%
A-base -22 -1.3% n.c n.c
B-base -48 -2.8% n.c n.c
B-A -26 -1.5% n.c n.c
SEA2-SEA1 -9 -0.7% n.c n.c
SEA3-SEA2 0 0.0% n.c n.c
SEA3-SEA1 n.c n.c -10 -0.8%
SEA4-SEA1 n.c n.c -53 -4.5%
SEA5-SEA3 n.c n.c -45 -3.9%
SEA5-SEA4 n.c n.c -1 -0.1%
ASEA1-base -447 -26% n.c n.c
BSEA1-base -468 -27% n.c n.c

134

Table 10: Eutrophication (AAE) in 2015 (Meq/year)

area
(km2)
AT Austria 83756 1240 1218 1212 1212 1218 1215 1239 1236
BE Belgium 30520 461 442 441 441 440 437 459 458
BG Bulgaria 110857 21 20 19 17 20 20 21 21
CH Switzerland 41084 29 29 28 28 29 29 29 29
CY Cyprus 5120 31 31 31 31 31 31 31 31
CZ
Czech
republic
78600 345 334 332 332 334 333 344 344
DE Germany 355693 9042 8869 8855 8855 8856 8840 9028 9011
DK Denmark 44270 21 20 20 20 20 20 21 21
EE Estonia 45263 2 2 2 2 2 2 2 2
ES Spain 497559 2037 1921 1853 1853 1919 1918 2035 2034
FI Finland 334395 43 31 31 31 31 31 43 43
FR France 547219 5330 5108 5069 5069 5088 5073 5306 5288
GR Greece 132022 57 47 39 39 47 47 57 57
HU Hungary 92857 139 136 134 134 136 135 139 138
IE Ireland 68128 0 0 0 0 0 0 0 0
IT Italy 301129 1495 1443 1357 1357 1437 1432 1489 1484
LT Lithuania 64612 57 56 56 56 56 56 57 57
LU Luxembourg 2589 43 42 42 42 42 41 43 42
LV Latvia 64207 5 5 5 5 5 5 5 5
MT Malta 319 0 0 0 0 0 0 0 0
NL Netherlands 35076 873 852 851 851 850 848 872 870
NO Norway 295987 0 0 0 0 0 0 0 0
PO Poland 310641 1958 1900 1889 1889 1899 1897 1956 1955
PT Portugal 88648 145 131 128 128 130 130 145 145
RO Romania 237394 140 135 132 131 135 135 140 139
SE Sweden 446266 35 27 27 27 27 27 35 35
SK Slovakia 48857 158 154 153 152 154 154 158 157
SL Slovenia 20672 190 184 180 180 184 184 190 190
UK
United
kingdom
243251 58 50 50 50 50 50 58 58

EU27+2 4626991 23955 23188 22935 22932 23139 23088 23903 23850

135

Table 11: Eutrophication (AAE) in 2020 (Meq/year)

area
(km2)
AT Austria 83756 1121 1087 1082 1078 1077 1086 1085 1078 1076
BE Belgium 30520 410 384 384 383 383 382 381 381 379
BG Bulgaria 110857 19 17 14 15 14 17 17 15 15
CH Switzerland 41084 22 21 21 21 21 21 21 21 21
CY Cyprus 5120 31 29 29 29 29 29 29 29 29
CZ
Czech
republic
78600 267 254 252 251 250 254 253 250 250
DE Germany 355693 8582 8332 8319 8304 8302 8322 8315 8294 8286
DK Denmark 44270 18 16 16 16 16 16 16 16 16
EE Estonia 45263 2 1 1 1 1 1 1 1 1
ES Spain 497559 1539 1345 1276 1253 1240 1343 1343 1251 1250
FI Finland 334395 36 20 20 20 20 20 20 20 20
FR France 547219 4633 4301 4271 4266 4259 4288 4279 4253 4244
GR Greece 132022 47 29 23 24 23 29 29 24 24
HU Hungary 92857 129 124 123 123 123 124 124 123 123
IE Ireland 68128 0 0 0 0 0 0 0 0 0
IT Italy 301129 1283 1183 1119 1128 1116 1177 1173 1122 1117
LT Lithuania 64612 58 57 57 57 57 57 57 57 57
LU Luxembourg 2589 35 34 34 34 34 34 33 34 33
LV Latvia 64207 5 5 5 5 5 5 5 5 5
MT Malta 319 0 0 0 0 0 0 0 0 0
NL Netherlands 35076 884 853 853 852 852 852 851 851 850
NO Norway 295987 0 0 0 0 0 0 0 0 0
PO Poland 310641 1693 1607 1596 1590 1588 1606 1606 1589 1588
PT Portugal 88648 132 111 110 99 98 111 111 99 98
RO Romania 237394 131 122 118 119 118 122 122 118 118
SE Sweden 446266 33 21 21 20 20 21 21 20 20
SK Slovakia 48857 134 128 127 127 126 128 128 126 126
SL Slovenia 20672 164 154 151 151 150 154 154 151 150
UK
United
kingdom
243251 26 21 21 20 20 21 21 20 20

EU27+2 4626991 21436 20259 20043 19984 19941 20222 20195 19946 19918

136

Table12: Differences between scenarios and years for eutrophication (AAE) (EU27+2 total)

2015 2020
abs
(Meq/yr)
%
abs
(Meq/yr)
%
SEA1-base -768 -3.2% -1178 -5.5%
SEA2-base -1020 -4.3% NC NC
SEA3-base -1023 -4.3% -1393 -6.5%
SEA4-base n.c n.c -1453 -6.8%
SEA5-base n.c n.c -1495 -7.0%
BSEA1-ASEA1 -50 -0.2% -27 -0.1%
ASEA1-SEA1 -49 -0.2% -37 -0.2%
BSEA1-SEA1 -99 -0.4% -64 -0.3%
A-SEA4-ASEA1 NC NC -276 -1.4%
BSEA4-BSEA1 NC NC -277 -1.4%
A-base -53 -0.2% n.c n.c
B-base -106 -0.4% n.c n.c
B-A -53 -0.2% n.c n.c
SEA2-SEA1 -252 -1.1% n.c n.c
SEA3-SEA2 -3 0.0% n.c n.c
SEA3-SEA1 n.c n.c -215 -1.1%
SEA4-SEA1 n.c n.c -275 -1.4%
SEA5-SEA3 n.c n.c -102 -0.5%
SEA5-SEA4 n.c n.c -43 -0.2%
137

Table 13: Particulate matter inhabitant weighted in 2015 (g PM10/m3*Minh.)
inhabitants base SEA1 SEA2 SEA3 ASEA1 BSEA1 A B
AT Austria 7994855 36 34 34 34 34 34 36 35
BE Belgium 10065306 100 91 91 91 90 90 100 99
BG Bulgaria 8392665 37 36 35 35 36 36 37 36
CH Switzerland 6994898 30 29 29 29 29 29 30 29
CY Cyprus 437692 1 1 1 1 1 1 1 1
CZ
Czech
republic
10362569 61 59 58 58 59 59 61 61
DE Germany 81558728 490 459 458 458 457 455 488 486
DK Denmark 4714330 23 17 17 17 17 17 23 23
EE Estonia 1460788 4 3 3 3 3 3 4 4
ES Spain 36886896 193 192 179 179 192 192 193 193
FI Finland 4900677 5 5 5 5 5 5 5 5
FR France 56954432 371 354 349 349 353 351 369 368
GR Greece 9640492 33 33 27 27 33 33 33 33
HU Hungary 10213844 52 51 50 50 50 50 52 52
IE Ireland 3472169 13 12 12 12 12 12 13 12
IT Italy 56077436 363 359 331 331 358 357 362 361
LT Lithuania 3684786 15 14 14 14 14 14 15 15
LU Luxembourg 404906 3 3 3 3 3 3 3 3
LV Latvia 2553183 8 7 7 7 7 7 8 8
MT Malta 326771 1 1 0 0 1 1 1 1
NL Netherlands 15059911 140 122 121 121 121 121 140 139
NO Norway 3874886 6 6 6 6 6 6 6 6
PO Poland 38535612 295 284 283 283 284 283 294 294
PT Portugal 9137149 41 41 40 40 41 41 41 41
RO Romania 22628702 122 121 120 118 121 120 122 122
SE Sweden 8315852 18 14 14 14 14 14 18 18
SK Slovakia 5318256 26 25 25 25 25 25 26 26
SL Slovenia 2023350 8 8 8 8 8 8 8 8
UK
United
kingdom
56343596 385 352 351 351 351 350 384 383

EU27+2 478334737 2878 2732 2670 2668 2724 2716 2870 2863

138

Table 14: Particulate matter inhabitant weighted in 2020 (g PM10/m3*Minh.)
inhabitants base SEA1 SEA3 SEA4 SEA5 ASEA1 BSEA1 ASEA4 BSEA4
AT Austria 7994855 32 31 30 30 30 31 31 30 30
BE Belgium 10065306 96 84 84 83 83 83 83 83 82
BG Bulgaria 8392665 36 36 34 34 34 36 36 34 34
CH Switzerland 6994898 27 26 26 25 25 26 26 25 25
CY Cyprus 437692 1 1 1 1 1 1 1 1 1
CZ
Czech
republic
10362569 55 52 52 52 52 52 52 52 51
DE Germany 81558728 464 423 422 419 419 422 420 417 416
DK Denmark 4714330 23 16 16 16 16 16 16 16 16
EE Estonia 1460788 4 3 3 3 3 3 3 3 3
ES Spain 36886896 173 171 155 151 149 171 171 151 151
FI Finland 4900677 5 4 4 4 4 4 4 4 4
FR France 56954432 339 317 312 309 308 316 315 308 307
GR Greece 9640492 33 33 26 27 25 33 33 26 26
HU Hungary 10213844 48 46 46 46 45 46 46 45 45
IE Ireland 3472169 12 11 11 10 10 11 11 10 10
IT Italy 56077436 337 328 298 302 297 328 327 301 300
LT Lithuania 3684786 15 14 13 13 13 14 14 13 13
LU Luxembourg 404906 2 2 2 2 2 2 2 2 2
LV Latvia 2553183 8 7 7 7 7 7 7 7 7
MT Malta 326771 1 1 0 1 0 1 1 1 1
NL Netherlands 15059911 135 112 112 111 111 112 112 110 110
NO Norway 3874886 6 5 5 5 5 5 5 5 5
PO Poland 38535612 275 262 260 260 259 261 261 259 259
PT Portugal 9137149 39 39 38 34 34 39 39 34 34
RO Romania 22628702 123 122 119 119 119 121 121 119 119
SE Sweden 8315852 18 13 13 13 13 13 13 13 13
SK Slovakia 5318256 24 23 23 23 23 23 23 23 23
SL Slovenia 2023350 8 7 7 7 7 7 7 7 7
UK
United
kingdom
56343596 361 318 318 307 307 317 317 307 306

EU27+2 478334737 2701 2509 2437 2414 2403 2502 2497 2407 2402

139

Table 15: Differences between scenarios and years for particulate matter ( inhabitant weighted,
EU27+2 total)
2015 2020

abs
(g
PM10/m3*Minh.)
%
abs
(g PM10/m3*Minh.)
%
SEA1-base -146 -5.1% -192 -7.1%
SEA2-base -208 -7.2% NC NC
SEA3-base -210 -7.3% -264 -10%
BSEA1-ASEA1 -8 -0.3% -5 -0.2%
ASEA1-SEA1 -8 -0.3% -7 -0.3%
BSEA1-SEA1 -16 -0.6% -12 -0.5%
A-SEA4-ASEA1 n.c n.c -95 -3.8%
BSEA4-BSEA1 n.c n.c -96 -3.8%
A-base -7 -0.3% n.c n.c
B-base -15 -0.5% n.c n.c
B-A -7 -0.3% n.c n.c
SEA2-SEA1 -62 -2.3% n.c n.c
SEA3-SEA2 -2 -0.1% n.c n.c
SEA3-SEA1 n.c n.c -72 -3.0%
SEA4-SEA1 n.c n.c -95 -3.8%
SEA5-SEA3 n.c n.c -34 -1.4%
SEA5-SEA4 n.c n.c -11 -0.5%

140

Table 16: SOMO35 in 2015 (ppm.d.*Minh.)
AT Austria 7994855 14 14 14 14 14 14 14 14
BE Belgium 10065306 13 13 13 13 13 13 13 13
BG Bulgaria 8392665 14 14 14 14 14 14 14 14
CH Switzerland 6994898 15 14 14 14 14 14 15 15
CY Cyprus 437692 1 1 1 1 1 1 1 1
CZ
Czech
republic
10362569 15 14 15 15 14 14 15 15
DE Germany 81558728 126 124 125 125 124 124 126 126
DK Denmark 4714330 7 7 7 7 7 7 7 7
EE Estonia 1460788 1 1 1 1 1 1 1 1
ES Spain 36886896 101 98 99 99 98 98 100 100
FI Finland 4900677 5 5 5 5 5 5 5 5
FR France 56954432 95 94 94 94 94 94 95 95
GR Greece 9640492 28 27 28 28 27 27 28 28
HU Hungary 10213844 19 19 19 19 19 19 19 19
IE Ireland 3472169 3 3 3 3 3 3 3 3
IT Italy 56077436 183 180 181 181 180 180 183 184
LT Lithuania 3684786 3 3 3 3 3 3 3 3
LU Luxembourg 404906 1 1 1 1 1 1 1 1
LV Latvia 2553183 3 3 3 3 3 3 3 3
MT Malta 326771 1 1 1 1 1 1 1 1
NL Netherlands 15059911 20 20 20 20 20 20 20 20
NO Norway 3874886 5 4 4 4 4 4 5 5
PO Poland 38535612 49 48 48 48 48 48 49 49
PT Portugal 9137149 28 27 27 27 27 27 28 28
RO Romania 22628702 31 31 31 31 31 31 31 31
SE Sweden 8315852 12 12 12 12 12 12 12 12
SK Slovakia 5318256 9 9 9 9 9 9 9 9
SL Slovenia 2023350 5 5 5 5 5 5 5 5
UK
United
kingdom
56343596 65 64 64 64 64 64 65 65

EU27+2 478334737 873 860 864 864 860 860 873 873

141

Table 17: SOMO35 in 2020 (ppm.d.*Minh.)
AT Austria 7994855 13 13 13 13 13 13 13 13 13
BE Belgium 10065306 14 14 14 14 14 14 14 14 14
BG Bulgaria 8392665 13 13 13 13 13 13 13 13 13
CH Switzerland 6994898 13 13 13 13 13 13 13 13 13
CY Cyprus 437692 1 1 1 1 1 1 1 1 1
CZ
Czech
republic
10362569 14 13 13 13 13 13 13 13 13
DE Germany 81558728 124 121 121 121 121 121 121 121 121
DK Denmark 4714330 7 7 7 7 7 7 7 7 7
EE Estonia 1460788 1 1 1 1 1 1 1 1 1
ES Spain 36886896 99 94 95 96 96 94 94 96 96
FI Finland 4900677 5 4 4 4 4 4 4 4 4
FR France 56954432 89 86 86 86 86 86 86 86 86
GR Greece 9640492 27 26 26 26 26 26 26 26 26
HU Hungary 10213844 17 17 17 17 17 17 17 17 17
IE Ireland 3472169 3 3 3 3 3 3 3 3 3
IT Italy 56077436 173 165 166 166 166 165 165 166 166
LT Lithuania 3684786 3 3 3 3 3 3 3 3 3
LU Luxembourg 404906 1 1 1 1 1 1 1 1 1
LV Latvia 2553183 3 2 2 2 2 2 2 2 2
MT Malta 326771 1 1 1 1 1 1 1 1 1
NL Netherlands 15059911 21 21 21 21 21 21 21 21 21
NO Norway 3874886 4 4 4 4 4 4 4 4 4
PO Poland 38535612 46 44 45 45 45 44 44 45 45
PT Portugal 9137149 26 25 26 26 26 25 25 26 26
RO Romania 22628702 30 29 29 29 29 29 29 29 29
SE Sweden 8315852 12 11 11 11 11 11 11 11 11
SK Slovakia 5318256 9 9 9 9 9 9 9 9 9
SL Slovenia 2023350 5 5 5 5 5 5 5 5 5
UK
United
kingdom
56343596 68 65 65 65 65 65 65 65 65

EU27+2 478334737 843 811 815 816 817 811 811 816 816

142

Table 18: Differences between scenarios and years for SOMO35 (EU27+2 total)

2015 2020
abs
(ppm.d.*Minh.)
%
abs
(ppm.d.*Minh.)
%
SEA1-base -13 -1.5% -33 -3.9%
SEA2-base -9 -1.0% n.c n.c
SEA3-base -9 -1.1% -28 -3.3%
SEA4-base n.c n.c -27 -3.2%
SEA5-base n.c n.c -26 -3.1%
BSEA1-ASEA1 0 0.0% 0 0.0%
ASEA1-SEA1 0 0.0% 0 0.0%
BSEA1-SEA1 0 0.0% 0 0.0%
A-SEA4-ASEA1 n.c n.c 6 0.7%
BSEA4-BSEA1 n.c n.c 6 0.7%
A-base 0 0.0% n.c n.c
B-base 0 0.0% n.c n.c
B-A 0 0.0% n.c n.c
SEA3-SEA1 NC NC 5 0.6%

143

Table 19: AOT40c in 2015 (ppm.d.*Minh.)
AT Austria 7994855 43 41 42 42 41 41 43 43
BE Belgium 10065306 31 31 31 31 31 31 31 31
BG Bulgaria 8392665 28 28 28 28 28 28 28 28
CH Switzerland 6994898 33 33 33 33 33 33 33 33
CY Cyprus 437692 3 2 2 2 2 2 3 3
CZ
Czech
republic
10362569 40 39 40 40 39 39 40 40
DE Germany 81558728 296 288 291 291 288 288 296 296
DK Denmark 4714330 23 22 22 22 22 22 23 23
EE Estonia 1460788 5 5 5 5 5 5 5 5
ES Spain 36886896 241 231 237 237 231 231 241 241
FI Finland 4900677 14 14 14 14 14 14 14 14
FR France 56954432 186 183 183 183 183 183 187 187
GR Greece 9640492 83 79 81 81 79 80 83 83
HU Hungary 10213844 53 52 53 53 52 52 53 53
IE Ireland 3472169 15 15 15 15 15 15 15 15
IT Italy 56077436 592 574 583 583 575 575 592 593
LT Lithuania 3684786 7 6 6 6 6 6 7 7
LU Luxembourg 404906 1 1 1 1 1 1 1 1
LV Latvia 2553183 7 7 7 7 7 7 7 7
MT Malta 326771 5 5 5 5 5 5 5 5
NL Netherlands 15059911 43 43 43 43 43 43 43 43
NO Norway 3874886 11 10 11 11 10 10 11 11
PO Poland 38535612 162 160 161 161 160 160 162 162
PT Portugal 9137149 77 74 75 75 74 74 77 77
RO Romania 22628702 74 74 74 74 74 74 74 74
SE Sweden 8315852 42 40 40 40 40 40 42 42
SK Slovakia 5318256 30 30 30 30 30 30 30 30
SL Slovenia 2023350 21 21 21 21 21 21 21 21
UK
United
kingdom
56343596 162 157 157 157 157 157 162 162

EU27+2 478334737 2328 2263 2290 2290 2263 2264 2329 2329

144

Table 20: AOT40c in 2020 (ppm.d.*Minh.)
Inhabitants base SEA1 SEA3 SEA4 SEA5 ASEA1 BSEA1 ASEA4 BSEA4
AT Austria 7994855 37 33 34 34 35 34 34 34 34
BE Belgium 10065306 30 30 30 30 30 30 30 30 30
BG Bulgaria 8392665 23 22 23 23 23 23 23 23 23
CH Switzerland 6994898 28 27 27 28 28 27 27 28 28
CY Cyprus 437692 2 2 2 2 2 2 2 2 2
CZ
Czech
republic
10362569 34 30 32 31 32 30 30 31 31
DE Germany 81558728 276 256 259 259 259 256 256 259 259
DK Denmark 4714330 22 20 20 20 20 20 20 20 20
EE Estonia 1460788 4 4 4 4 4 4 4 4 4
ES Spain 36886896 230 207 215 217 219 207 207 217 217
FI Finland 4900677 13 11 11 11 11 11 11 11 11
FR France 56954432 169 159 160 160 160 159 159 160 160
GR Greece 9640492 79 71 73 72 73 71 71 72 72
HU Hungary 10213844 45 42 43 43 43 42 42 43 43
IE Ireland 3472169 14 13 13 13 13 13 13 13 13
IT Italy 56077436 534 490 499 498 501 490 490 499 499
LT Lithuania 3684786 6 6 6 6 6 6 6 6 6
LU Luxembourg 404906 1 1 1 1 1 1 1 1 1
LV Latvia 2553183 6 6 6 6 6 6 6 6 6
MT Malta 326771 5 4 4 4 4 4 4 4 4
NL Netherlands 15059911 43 42 42 42 42 42 42 42 42
NO Norway 3874886 10 9 9 9 9 9 9 9 9
PO Poland 38535612 142 135 137 137 137 135 135 137 137
PT Portugal 9137149 73 68 69 71 71 68 68 71 71
RO Romania 22628702 65 63 64 64 64 63 64 64 64
SE Sweden 8315852 40 35 35 35 35 35 35 35 35
SK Slovakia 5318256 26 25 25 25 25 25 25 25 25
SL Slovenia 2023350 18 17 17 17 18 17 17 17 17
UK
United
kingdom
56343596 163 146 146 147 147 146 146 147 147

EU27+2 478334737 2138 1975 2006 2008 2016 1976 1977 2009 2010

145

Table 21: Differences between scenarios and years for AOT40c (EU27+2 total)

2015 2020
abs
(ppm.d.*Minh.)
%
abs
(ppm.d.*Minh.)
%
SEA1-base -66 -2.8% -163 -7.6%
SEA3-base -38 -1.6% -133 -6.2%
SEA4-base n.c n.c -130 -6.1%
SEA5-base n.c n.c -122 -5.7%
BSEA1-ASEA1 1 0.0% 0 0.0%
ASEA1-SEA1 1 0.0% 1 0.0%
BSEA1-SEA1 1 0.0% 1 0.1%
A-SEA4-ASEA1 n.c n.c 33 1.7%
BSEA4-BSEA1 n.c n.c 33 1.7%
B-A 1 0.0% n.c n.c
n.c. - not calculated

146

Table 22: AOT40f in 2015 (ppm.d.*Minh.)
inhab. base SEA1 SEA2 SEA3 ASEA1 BSEA1 A B
AT Austria 7994855 60 59 60 60 59 59 60 60
BE Belgium 10065306 54 54 54 54 54 54 54 54
BG Bulgaria 8392665 50 49 49 49 49 49 50 50
CH Switzerland 6994898 59 59 59 59 59 59 59 59
CY Cyprus 437692 4 3 4 4 3 3 4 4
CZ
Czech
republic
10362569 49 48 49 49 48 48 49 49
DE Germany 81558728 463 453 455 455 453 453 463 463
DK Denmark 4714330 32 31 31 31 31 31 32 32
EE Estonia 1460788 5 5 5 5 5 5 5 5
ES Spain 36886896 448 431 442 442 431 431 448 448
FI Finland 4900677 18 17 17 17 17 17 18 18
FR France 56954432 387 378 379 379 378 378 387 387
GR Greece 9640492 133 128 130 130 128 128 133 133
HU Hungary 10213844 78 77 78 78 77 77 78 78
IE Ireland 3472169 16 15 15 15 15 15 16 16
IT Italy 56077436 1022 994 1004 1004 994 995 1022 1022
LT Lithuania 3684786 10 10 10 10 10 10 10 10
LU Luxembourg 404906 2 2 2 2 2 2 2 2
LV Latvia 2553183 10 10 10 10 10 10 10 10
MT Malta 326771 8 8 8 8 8 8 8 8
NL Netherlands 15059911 71 70 70 70 70 70 71 71
NO Norway 3874886 18 17 17 17 17 17 18 18
PO Poland 38535612 202 199 200 200 199 199 202 202
PT Portugal 9137149 130 125 127 127 125 125 130 130
RO Romania 22628702 104 104 105 105 104 104 104 104
SE Sweden 8315852 58 56 56 56 56 56 58 58
SK Slovakia 5318256 42 41 42 42 41 41 42 42
SL Slovenia 2023350 30 30 30 30 30 30 30 30
UK
United
kingdom
56343596 293 284 284 284 284 284 293 293

EU27+2 478334737 3856 3756 3791 3792 3757 3757 3856 3856

147

Table 23: AOT40f in 2020 (ppm.d.*Minh.)
AT Austria 7994855 53 49 50 50 50 49 49 50 50
BE Belgium 10065306 55 55 55 55 55 55 55 55 55
BG Bulgaria 8392665 44 41 41 41 41 41 41 41 41
CH Switzerland 6994898 52 50 50 50 50 50 50 50 50
CY Cyprus 437692 4 3 3 3 3 3 3 3 3
CZ
Czech
republic
10362569 43 39 40 40 40 39 39 40 40
DE Germany 81558728 442 415 418 418 418 415 415 418 418
DK Denmark 4714330 31 28 28 28 28 28 28 28 28
EE Estonia 1460788 5 4 4 4 4 4 4 4 4
ES Spain 36886896 433 395 408 410 414 395 395 410 410
FI Finland 4900677 16 14 14 14 14 14 14 14 14
FR France 56954432 345 323 324 326 326 323 323 326 326
GR Greece 9640492 127 115 117 116 117 115 115 116 117
HU Hungary 10213844 66 62 64 64 64 63 63 64 64
IE Ireland 3472169 14 13 13 13 13 13 13 13 13
IT Italy 56077436 929 859 870 869 872 859 859 870 870
LT Lithuania 3684786 9 9 9 9 9 9 9 9 9
LU Luxembourg 404906 2 2 2 2 2 2 2 2 2
LV Latvia 2553183 10 9 9 9 9 9 9 9 9
MT Malta 326771 9 7 8 7 8 7 7 7 7
NL Netherlands 15059911 72 71 71 71 71 71 71 71 71
NO Norway 3874886 16 14 14 14 14 14 14 14 14
PO Poland 38535612 178 170 171 171 172 170 170 171 171
PT Portugal 9137149 122 114 116 119 119 114 114 119 119
RO Romania 22628702 93 91 92 92 92 91 91 92 92
SE Sweden 8315852 55 49 49 49 49 49 49 49 49
SK Slovakia 5318256 37 35 35 35 35 35 35 35 35
SL Slovenia 2023350 26 25 25 25 25 25 25 25 25
UK
United
kingdom
56343596 296 271 271 272 272 271 271 272 272

EU27+2 478334737 3579 3331 3371 3378 3388 3332 3333 3379 3380

148

Table 24: Differences between scenarios and years for AOT40f (EU27+2 total)

2015 2020
abs
(ppm.d.*Minh.)
%
abs
(ppm.d.*Minh.)
%
SEA1-base -100 -2.6% -248 -6.9%
SEA3-base -64 -1.7% -208 -5.8%
SEA4-base n.c n.c -201 -5.6%
SEA5-base n.c n.c -191 -5.3%
BSEA1-ASEA1 1 0.0% 0 0.0%
ASEA1-SEA1 1 0.0% 1 0.0%
BSEA1-SEA1 1 0.0% 1 0.0%
A-SEA4-ASEA1 NC NC 47 1.4%
BSEA4-BSEA1 NC NC 47 1.4%
B-A 1 0.0% n.c n.c
SEA3-SEA1 n.c n.c 40 1.2%

149

Table 25: O3avg in 2015 (ppm.d.*Minh.)
AT Austria 7994855 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2
BE Belgium 10065306 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2
BG Bulgaria 8392665 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3
CH Switzerland 6994898 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2
CY Cyprus 437692 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
CZ
Czech
republic
10362569 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3
DE Germany 81558728 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0
DK Denmark 4714330 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
EE Estonia 1460788 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
ES Spain 36886896 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
FI Finland 4900677 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
FR France 56954432 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5
GR Greece 9640492 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3
HU Hungary 10213844 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3
IE Ireland 3472169 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
IT Italy 56077436 1.7 1.7 1.7 1.7 1.7 1.7 1.7 1.7
LT Lithuania 3684786 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
LU Luxembourg 404906 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
LV Latvia 2553183 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
MT Malta 326771 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
NL Netherlands 15059911 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3
NO Norway 3874886 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
PO Poland 38535612 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
PT Portugal 9137149 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3
RO Romania 22628702 0.7 0.7 0.7 0.7 0.7 0.7 0.7 0.7
SE Sweden 8315852 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2
SK Slovakia 5318256 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2
SL Slovenia 2023350 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
UK
United
kingdom
56343596 1.1 1.1 1.1 1.1 1.1 1.1 1.1 1.1

EU27+2 478334737 12 12 12 12 12 12 12 12

150

Table 26: O3avg in 2020 (ppm.d.*Minh.)
Inhabitants base SEA1 SEA3 SEA4 SEA5 ASEA1 BSEA1 ASEA4 BSEA4
AT Austria 7994855 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2
BE Belgium 10065306 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2
BG Bulgaria 8392665 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3
CH Switzerland 6994898 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2
CY Cyprus 437692 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
CZ
Czech
republic
10362569 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3
DE Germany 81558728 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1 2.1
DK Denmark 4714330 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
EE Estonia 1460788 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
ES Spain 36886896 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
FI Finland 4900677 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
FR France 56954432 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5
GR Greece 9640492 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3
HU Hungary 10213844 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3
IE Ireland 3472169 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
IT Italy 56077436 1.7 1.7 1.7 1.7 1.7 1.7 1.7 1.7 1.7
LT Lithuania 3684786 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
LU Luxembourg 404906 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
LV Latvia 2553183 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
MT Malta 326771 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
NL Netherlands 15059911 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3
NO Norway 3874886 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
PO Poland 38535612 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
PT Portugal 9137149 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3
RO Romania 22628702 0.7 0.7 0.7 0.7 0.7 0.7 0.7 0.7 0.7
SE Sweden 8315852 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2
SK Slovakia 5318256 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2
SL Slovenia 2023350 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
UK
United
kingdom
56343596 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2

EU27+2 478334737 13 13 13 13 13 13 13 13 13

151

Table 27: Differences between scenarios and years for O3avg (EU27+2 total)
2015 2020

abs
(ppm.d.*Minh.)
%
abs
(ppm.d.*Minh.)
%
SEA1-base 0.008 0.06% 0.006 0.05%
SEA2-base 0.018 0.14% n.c n.c
SEA3-base 0.019 0.15% 0.023 0.18%
SEA4-base n.c n.c 0.027 0.21%
SEA5-base n.c n.c 0.023 0.18%
BSEA1-ASEA1 0.005 0.04% 0.008 0.06%
ASEA1-SEA1 -0.006 -0.05% -0.001 -0.01%
BSEA1-SEA1 -0.001 -0.01% 0.007 0.06%
A-SEA4-ASEA1 n.c n.c 0.014 0.11%
BSEA4-BSEA1 n.c n.c 0.010 0.08%
A-base -0.002 -0.02% n.c n.c
B-base 0.001 0.01% n.c n.c
B-A 0.003 0.02% n.c n.c
SEA2-SEA1 0.010 0.08% n.c n.c
SEA3-SEA2 0.001 0.01% n.c n.c
SEA3-SEA1 n.c n.c 0.017 0.13%
SEA4-SEA1 n.c n.c 0.021 0.17%
SEA5-SEA3 n.c n.c 0.000 0.00%
SEA5-SEA4 n.c n.c -0.004 -0.03%
ASEA1-base 0.002 0.02% n.c n.c
BSEA1-base 0.007 0.06% n.c n.c

152

Table 28: Difference between 2015 and 2020 per effected area
SECA
NS, B NS, B
NS, B, MED,
BS
NS, B NS, B
Base-
line
SEA1 SEA3 ASEA1 BSEA1
S-deposition
abs -1300 -1955 -2546 -1887 -1689
% -2% -4% -5% -3% -3%
N-deposition
abs -6140 -8297 -8266 -8278 -8237
% -3% -5% -5% -5% -5%
Acidification
abs -61 -141 -141 -136 -126
% -4% -11% -11% -11% -10%
Eutrophication
abs -2519 -2929 -2948 -2917 -2894
% -11% -13% -13% -13% -13%
PM10
abs -177 -223 -231 -222 -219
% -6% -8% -9% -8% -8%
AOT40f
abs -277 -425 -420 -424 -424
% -7% -11% -11% -11% -11%

153

17 Appendix 5 - Health Benefit Diagrams

SEA1 Scenario - 2015


154


AS1 Scenario - 2015

155

BS1 Scenario - 2015


156



157


ASEA1 Scenario - 2020

158

BSEA1 Scenario - 2020

ASEA4 Scenario - 2020

159

BSEA4 Scenario 2020

160

18 Appendix 6 - Methodology for cost
calculations
Scenario description
The following parameters are specified in each scenario:
1. Fuel use [PJ]
2. Implementation of Tier I, II, III NO
X
control technologies [% of fuel]
3. Sulphur control:
* The sulphur content of fuel used, [mass % in fuel] or
* Implementation rate of new or retrofit scrubbers [% of vessels]
4. Costs for NO
X
and SO
2
control

Scenario background data specification
Scenarios: Baseline, SEA1, SEA2, SEA3, SEA4, SEA5 (see chapter 6 for more info)
Years: 2015 and 2020
Sea regions:
Baltic Sea
Black Sea
Mediterranean
North East Atlantic
North Sea & English Channel

Vessel types:
Ferry
Bulk carrier
Chemical tanker
Container ship
General Cargo
LG tanker
Oil tanker
RoRo cargo

Engine types:
Abbreviation Description
SSD-MG/DO Slow Speed engines using Marine Diesel or Gas Oil
MSD-MG/DO Medium Speed engines using Marine Diesel or Gas Oil
HSD-MG/DO High Speed engines using Marine Diesel or Gas Oil
GT-MG/DO Gas turbines using Marine Diesel or Gas Oil
ST-MG/DO Steam turbines using Marine Diesel or Gas Oil
SSD-RO Slow Speed engines using Residual Oil
MSD-RO Medium Speed engines using Residual Oil
HSD-RO High Speed engines using Residual Oil
GT-RO Gas turbines using Residual Oil
ST-RO Steam turbines using Residual Oil

161

Engine size classification:
Main engine
Small Medium Large
range [kW] 0 - 6 000 6 000 - 15 000 15 000 -
Average ME size [kW] 3 000 10 000 25 000
SSD 2 share [%] 48 58 55
Auxiliary engine size
Small Medium Large
range [kW] 0 - 1 000 1 000 - 2 000 2 000 -
Average AE size [kW] 560 1 480 3 800

The engine size classification follows the engine size classification as specified in the ENTEC 2005
reports.

The distribution of the engine sizes relative to the total fuel use is derived from ENTEC 2005 and is
expressed as % of total fuel use.

Share of fuel used in respective engine size category:
Small Medium Large
25% 41% 34%

Scenario specific parameters
Fuels considered and sulphur content
The following fuels are available for use in the scenarios. The use of each fuel is scenario-specific
and depends on whether the sea region is considered as a Sulphur Emission Control Area (SECA) in
the scenario. When scrubbers are used instead of low-sulphur fuel oils, a sulphur content of maximum
2.94 % is used for the region.

Fuel type Sulphur content
Marine Diesel (MD) 0.5 %
MD 0.1 %
Heavy Fuel Oils (HF) /
Residual Oils (RO)
2.94 %
HF 1.5 %
HF 0.5 %
HF 0.1 %

Implementation of Tier I, II, III NO
x
control technologies
The implementation of Tier control technologies correspond to the vessel vintages specified in
Extremis. In the cost estimates it is assumed that the age distribution will remain the same for the
years 2005, 2015 and 2020.

For 2015 and all sea regions, the following implementation rates of Tier technologies are obtained:
2015 Tier III Tier II Tier I 30 % Tier I
No NO
x
-
control
Ferry 0% 16% 28% 17% 39%
Bulk carrier 0% 20% 37% 13% 30%
Chemical tanker 0% 29% 44% 8% 19%
Container ship 0% 32% 46% 7% 16%
General Cargo 0% 7% 26% 20% 47%
LG tanker 0% 29% 44% 8% 19%
Oil tanker 0% 29% 44% 8% 19%
RoRo cargo 0% 16% 28% 17% 39%

162

For 2020 in sea regions considered as NO
X
-ECAs, the following Tier technology implementation rates
are obtained:
2020 NO
X
-ECA Tier III Tier II Tier I 30% Tier I
No NO
X
-
control
Ferry 16% 16% 20% 15% 34%
Bulk carrier 20% 22% 27% 9% 22%
Container ship 32% 29% 25% 4% 10%
General Cargo 7% 15% 22% 17% 39%
LG tanker 29% 23% 33% 5% 11%
Oil tanker 29% 23% 33% 5% 11%
RoRo cargo 16% 16% 20% 15% 34%

For 2020, in sea regions considered as outside NO
X
-ECA, the following Tier technology
implementation rates are obtained:
2020 outside
NO
X
-ECA
Tier III Tier II Tier I 30% Tier I
No NO
X
-
control
Ferry 0% 32% 20% 15% 34%
Bulk carrier 0% 42% 27% 9% 22%
Container ship 0% 61% 25% 4% 10%
General Cargo 0% 22% 22% 17% 39%
LG tanker 0% 52% 33% 5% 11%
Oil tanker 0% 52% 33% 5% 11%
RoRo cargo 0% 32% 20% 15% 34%

Sulphur control - scrubbers
In the calculations where SO
2
scrubbers are used instead of low-sulphur fuels, the following
implementation rates are obtained from considering the age distribution in Extremis The scrubber
implementation rates takes into consideration whether retrofit is needed or if new installations are
sufficient. The 'Retrofit' categories include the vessels constructed before 2011, while the 'New'
category includes the vessels constructed after 2011.

Scrubbers -new
and retrofits
2015 2015 2020 2020
Retrofit New Retrofit New
Ferry 84% 16% 68% 32%
Bulk carrier 80% 20% 58% 42%
Chemical tanker 71% 29% 48% 52%
Container ship 68% 32% 39% 61%
General Cargo 93% 7% 78% 22%
LG tanker 71% 29% 48% 52%
Oil tanker 71% 29% 48% 52%
RoRo cargo 84% 16% 68% 32%
163

The Gemini Building
Fermi Avenue
Harwell International Business Centre
Didcot
Oxfordshire
OX11 0QR

Tel: 0845 345 3302
Fax: 0870 190 6138

E-mail: info@aeat.co,uk

www.aea.co.uk

CBA of S PDF

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

CBA of S PDF

Transféré par

Droits d'auteur :

Formats disponibles

Cost Benefit Analysis to Support the Impact Assessment

accompanying the revision of Directive 1999/32/EC on the

Vous aimerez peut-être aussi