Scholarly article on topic 'Effects of More Stringent Sulphur Requirements for Sea Transports'

Effects of More Stringent Sulphur Requirements for Sea Transports Academic research paper on "Civil engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Transportation Research Procedia
OECD Field of science
Keywords
{"sea transports" / "European sulphur emission control area (SECA)" / "impact analysis" / "freight transport model" / "modal split" / "route choice"}

Abstract of research paper on Civil engineering, author of scientific article — Inge Vierth, Rune Karlsson, Anna Mellin

Abstract In 2008 the International Maritime Organization (IMO) decided on more stringent requirements from 2015 for airborne emissions of sulphur dioxide from sea transports in the sulphur emission control areas (SECA). The European SECA comprises the Baltic Sea, the North Sea and the English Channel. The paper contains an overview of the European studies that have been carried out to investigate the impacts of IMO's more stringent sulphur requirements. All studies were carried out after IMO's decision in 2008 (which means that the decision was taken based on other reasons). The studies focus on different aspects but all of them estimate how IMO's stricter requirements will affect the sea transport costs. The Swedish impact studies are described in particular: in the 2009 study the national transport model Samgods was used and in 2013 both the Samgods model and the agent-based simulation model Tapas. Impacts on the choice of transport chains, routes and ports are calculated. The results indicate that shippers to some extent can reduce the increase in transport cost by transferring flows from the Swedish east coast to the Swedish south and west coast, the Norwegian coast and the land-based route via Denmark. Modal back shifts from sea to rail and road occur. These shifts are modest, especially if higher prices for diesel and higher rail track fees are assumed on top of more stringent sulphur requirements in the SECA. One important question is to what extent the increases in costs that are due to more stringent requirements can be compensated for by improved efficiency of the transports, such as the exploitation of economies of scale.

Academic research paper on topic "Effects of More Stringent Sulphur Requirements for Sea Transports"

Available online at www.sciencedirect.com

ScienceDirect

Transportation Research

Procedía

ELSEVIER

Transportation Research Procedía 8 (2015) 125- 135

www.elsevier.com/locate/procedia

European Transport Conference 2014 - from Sept-29 to Oct-1, 2014

Effects of more stringent sulphur requirements for sea transports

Inge Viertha,b'*9 Rune Karlssonab, Anna Mellinabc

aSwedish National Road and Transport Research Institute (VTI), Teknikringen 10, 102 15 Stockholm, Sweden bCentre for Transport Studies (CTS), Teknikringen 10, 102 15 Stockholm, Sweden c Swedish Agency for Marine and Water Management, Gullbergs strandgata 15, 411 04 Göteborg, Sweden

Abstract

In 2008 the International Maritime Organization (IMO) decided on more stringent requirements from 2015 for airborne emissions of sulphur dioxide from sea transports in the sulphur emission control areas (SECA). The European SECA comprises the Baltic Sea, the North Sea and the English Channel. The paper contains an overview of the European studies that have been carried out to investigate the impacts of IMO's more stringent sulphur requirements. All studies were carried out after IMO's decision in 2008 (which means that the decision was taken based on other reasons). The studies focus on different aspects but all of them estimate how IMO's stricter requirements will affect the sea transport costs. The Swedish impact studies are described in particular: in the 2009 study the national transport model Samgods was used and in 2013 both the Samgods model and the agent-based simulation model Tapas. Impacts on the choice of transport chains, routes and ports are calculated. The results indicate that shippers to some extent can reduce the increase in transport cost by transferring flows from the Swedish east coast to the Swedish south and west coast, the Norwegian coast and the land-based route via Denmark. Modal back shifts from sea to rail and road occur. These shifts are modest, especially if higher prices for diesel and higher rail track fees are assumed on top of more stringent sulphur requirements in the SECA. One important question is to what extent the increases in costs that are due to more stringent requirements can be compensated for by improved efficiency of the transports, such as the exploitation of economies of scale.

© 2015The Authors.PublishedbyElsevierB.V. Thisis an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Selection and peer-review under responsibility of Association for European Transport

Keywords: sea transports; European sulphur emission control area (SECA); impact analysis; freight transport model; modal split; route choice

* Corresponding author: E-mail address: inge.vierth@vti.se

2352-1465 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Selection and peer-review under responsibility of Association for European Transport doi:10.1016/j.trpro.2015.06.048

1. Introduction

In 2008 the International Maritime Organization (IMO) decided on more stringent requirements for airborne emissions of sulphur dioxide from sea transports in the sulphur emission control areas (SECA). The European part of SECA comprises the Baltic Sea, the North Sea and the English Channel. The SECA along the coast of North America and Puerto Rico is not addressed in this paper. IMO's decision implies a larger difference between the requirements inside and outside the SECA. (See Table 1.)

Ship owners can apply different strategies to meet IMO's stricter requirements in the SECA from 2015. Three possibilities are: 1) to use 0.1 % marine gas oil that is more expensive than the currently used heavy fuel oil, 2) to switch to cleaner energy sources such as liquid natural gas or methanol, or 3) to use scrubbers to reduce the sulphur emissions. In the short term, all strategies lead to higher costs for sea transports. Shippers have the option of adapting transport chains that include sea transport by changing route and/or mode or, in the longer run, by moving production facilities. The shift from sea to land is sometimes referred to as modal backshift because it goes in the opposite direction to the goal in the EU's White Book (COM (2011)) to shift long distance road transports to more environmental friendly transports. It is, however, not obvious how developments in the market for marine fuels will influence diesel prices in general, nor is it clear how fuel taxes, infrastructure fees and regulations for the land-based modes will develop. Possible effects of IMO's more stringent sulphur requirements for sulphur emissions in the SECA have been studied in the countries that are most affected.

Table 1- IMO's highest permitted sulphur content in marine fuels inside and outside SECA

Within SECA Outside SECA

Today 1.5 weight % 4.5 weight %

From 2015 0.1 weight %

From 2020/2025 0.5 weight %

Other related topics are the plans for stronger requirements for nitrogen oxides for new vessels, and different economic measures related to carbon dioxide emissions are being discussed to internalize the external costs caused by sea transports. These are, however, not addressed in this paper.

This paper focuses on the calculation of transport impacts on freight transports in two Swedish government commissions. Key questions studied are to what extent sea transports are redistributed from the SECA to the seas outside of SECA and to what extent modal shifts from sea to rail and road can be expected. The Swedish rail share of around 22 % is about twice the average rail share of the 27 EU member states when the ton-kilometer for all modes are included. On the other hand, Sweden has no inland waterway transports. In addition, the effect on the choice of transport chains, routes and ports is calculated.

The structure of the paper is as follows. Section 2 looks at studies that have been commissioned to assess the impacts of IMO's stricter requirements in the SECA. Section 3 addresses the Swedish impact analyses of 2009 and 2013. The national freight transport model Samgods that has been used in both analyses is presented. Results from the Samgods analyses of 2009 and 2013 and two case studies that use the activity-based model Tapas are discussed. In section 4 some conclusions are drawn.

2. Studies performed

National and international impact studies have been carried out by transport ministries and public agencies, by associations of ship owners, and in research projects. The main focus has been on the impact in the respective country or region and conditions differ due to the country's location, volume and structure of transport demand, modal split, etc.

The Finnish Ministry of Transport and Communication commissioned the Centre for Maritime Studies at Turku University to study the impact of IMO's stricter sulphur requirements on the maritime industry and the manufacturing industry (Finnish Ministry of Transport and Communication (2009)). The health effects of using cleaner fuel were also estimated. Potential impacts in the form of modal shifts were not quantified. Four years later, the Finnish

Transport Safety Agency engaged Elomatic Marine Engineering to evaluate the cost burden resulting from IMO's sulphur directive with respect to the sea traffic allocated to Finland alone within the Baltic region (Finnish Transport Safety Agency (2013)). This study is based on statistics covering vessel traffic and port calls in Finland in 2011. The impacts on all maritime traffic in the Baltic were also estimated.

The Swedish government has commissioned two investigations: Swedish Maritime Administration (2009) and Trafikanalys (2013). The impact analyses performed within these government investigations are described below.

The UK Maritime and Coastguard Agency commissioned a study that analyses how the implementation of the IMO sulphur directive would reduce emissions, expressed in tons and monetary terms, and increase the shipping industry's costs (ENTEC (2009)). Three scenarios are considered: 1) 90 % of vessels switch fuel and 10 % use scrubbers, 2) all vessels switch fuel, and 3) all vessels use scrubbers.

The European Commission engaged AEA, TNO, IVL and EMRC to quantify the societal costs and benefits associated with the new sulphur requirements in all member states (AEA et al. (2009)). No modal shift is assumed. The impacts of stronger requirements for nitrogen oxides (NOx) and potential extensions of the SECA are also studied.

The European research project SKEMA analyses the possibilities for modal shifts as a consequence of IMO's stricter sulphur requirements for a selection of routes. This is done with the help of the NECL model for ten competing routes operating on four corridors and the Tapas model for five competing routes along a single corridor (Kehoe (2010)). The impact of the implementation of road infrastructure fees, which are assumed to raise road transport costs by 2 %, was also tested and found to not sufficiently counteract the cost increases for sea transports due to the sulphur directive.

The European Community Ship owners' Association (ECSA) commissioned a study to analyse the impacts of the stricter sulphur requirements in the SECA (Notteboom et al. (2010)). The report has three sections. The first focuses on the expected impacts of the stricter sulphur requirements on transport costs and prices. The second part estimates the impacts on the modal split for selected routes using both a stated preference approach and comparative cost analysis. The third part estimates the impacts on the external costs in the routes. It is, for example, shown how the decrease in external costs due to the IMO directive can be compensated by an increase in external costs for land-based transports when modal shifts are taken into account.

The German Ship owners' Association and the Association of German Seaport Operators engaged the Institute of Shipping and Economics and Logistics (ISL) to carry out an impact study (ISL, 2010). ISL estimate the expected mode shifts from sea to land due to IMO's stricter sulphur requirements in the SECA using a logit model. Different corridors to/from Germany are studied.

In 2010, the ship owners' associations from Belgium, Finland, Germany, The Netherlands, Sweden, and the UK together ordered a review of six previously performed impact studies (ENTEC (2010)). Due to the time scales of the work it was not possible to include the ISL study. Four of the six studies, (Finnish Ministry of Transport and Communication (2009), Swedish Martitime Administration (2009), Notteboom et al. (2010), Kehoe et al. (2010), consider the modal split effect with a somewhat different focus and the conclusion is that the implementation of the IMO sulphur directive will lead to some shifts from sea to road and rail. Depending on the route, sea transport volumes are estimated to decrease by 3 to 50 %. The extent to which this takes place depends on a number of route-specific factors such as level of competition and availability of alternative route as well as the projected fuel costs (ENTEC (2010)). Holmgren et al (2014) compile the impact studies that have been published up to 2013 and find that 'only a few studies have made a clear modal shift focus visible'.

3. Swedish impact analyses

It would be interesting to compare the quantitative outcomes and methods applied in all the consequence studies listed above. But the results are not comparable since different aspects are analyzed in the studies and not all reports contain numerical results. Different adaptions to IMO's more stringent requirements for airborne emissions of sulphur dioxide from sea transports in the SECA and different increases for the sea transport costs are assumed. Furthermore, the impacts are calculated for specific corridors in some cases and for countries or regions in other cases. Therefore we focus on the impact studies that have been performed within the Swedish government investigations of 2009 and 2013. These studies were carried out with help of the national freight transport model Samgods (2009 and 2013) and the activity-based Tapas model (2013).

3.1. Samgods

3.1.1. Model

The national freight transport model system Samgods consists of three parts:

• PC-matrices that describe the transport demand between production and consumption in, to, from and through Sweden for 34 commodities in tons, (see Edwards et al. (2008)). The PC-matrices are assumed to be constant, i.e. do not decrease due to higher transport costs. This assumption is normally used in freight models and is usually considered to be realistic in the short term. All model runs in the Samgods studies 2009 and 2013 refer to the base year 2006.

• A logistics model that minimizes the shippers' annual logistics costs (transport costs, warehouse and order costs) and transfers the commodity specific PC-flows to vehicle type specific origin-destination flows (OD-flows). The OD-flows can go directly or via one or more terminals from P to C. The model comprises 33 vehicle types: five road, eight rail, 19 sea and one air (see de Jong et al. (2008) and Vierth et al. (2009 (b)). For sea transports, different types of vessels (container, ro-ro vessels, other) and ferries (road, rail) are included. Different average speeds are assumed for different vehicle/vessel types, i.e. for container vessels 30-39 km/h, for ro-ro vessels 30 km/h, for other vessels 22-30 km/h. The economies of scale aspect is handled using different vehicle sizes. This aspect is especially important for vessels that differ significantly in size (and therefore costs). The cargo-carrying of the vessels varies from 1 000 to 250 000 tons. Ferry capacity is normally shared by different firms and economies of scale are assumed to be less relevant than for vessels. Transport costs comprise underway costs and transfer costs. The kilometer costs in turn comprise time-based costs and distance-based costs as well as infrastructure fees. When it comes to sea transports, time charter rates for vessel and personnel are assumed as time-based costs and fuel costs as kilometer costs. Fuel costs stand for a little less than half of the total underway costs for sea transports which corresponds to the shares in other studies, e.g. Lindstad et al. (2013). The transport costs have been updated from 2008 to 2012 and the flows through single ports have been calibrated by assuming more realistic transfer times and costs in the ports. In 2009, a version of the logistics model from 2008 has been used, and in 2013, a version from September 2012.

• An infrastructure network.

Within the deterministic logistics model, firms' annual logistics costs are minimized, taking into account the tradeoff between transport costs and warehouse costs. Also taken into account is the fact that transport costs per unit can be reduced by using larger vehicles when transporting goods from one or several shippers. Transport costs per vehicle-kilometer are used as input; load factors and costs per ton-kilometer are computed. The choice between predefined container and non-container transport chains is modelled. It is assumed that transport companies pass cost changes to shippers. Infrastructure restrictions in the form of maximum depth for vessels and maximum weight for trucks and trains are taken into account. Rail capacity restrictions in terms of number of trains per track can be taken into account manually by comparing calculated volumes to the maximum capacity. Capacity problems in ports are assumed to be negligible.

3.1.2. Impact study 2009

The 2009 study, Vierth et al. (2009 (a)), focuses on the question to what extent higher prices for marine fuel will lead to modal shifts. Three scenarios assume that 0.1 % marine gas oil is used in the SECA instead of the heavier fuel. The transport cost increases vary between the different vessels types depending on the current level of sulphur in the fuel used. One explanation for the differences is the introduction of environmentally differentiated fairway dues in Swedish waters in the late 1990s. In Scenario 1 the fuel costs inside the SECA are estimated to rise by 12 % for ferries, 72 % for ro-ro vessels, 81 % for container vessels and 36 % for other vessels. Everything else is assumed to be the same. Scenario 2 and Scenario 3 assume higher crude oil prices inside and outside the SECA in addition to the higher fuel costs due to IMO's stricter requirements. The assumed price of € 502 per ton for 0.1 % marine gas oil is in the same range as the prices in other studies (see Holmgren et al. (2014)). The freight flows in the base scenario that describes the actual situation and the three investigation scenarios are compared.

Concerning Scenario 1, the difference map for the sea transports (see Fig. 1) indicates shifts from the Swedish east coast to the Swedish south and west coast to reduce the kilometers in the SECA. It is also computed to be advantageous to use the port of Narvik in the north of Norway outside the SECA.

Fig. 1. Calculated difference between Scenario 1 and base scenario for sea transports in tons on infrastructure network in the 2009 study (red colour shows increases and black decreases) N = Narvik, G = Gothenburg, O = Oresund bridge, S = Stockholm. The violet line marks the border

between the SECA and the Atlantic.

It is also calculated that there will be transfers from ports in northern Sweden to ports in central and southern Sweden, which lead to shorter sea transport distances in the SECA (often by ferry) and longer land-based hinterland transports. This is consistent with the conclusion in Notteboom et al. (2010) that the intermodal corridor competition between the Nordic countries and Benelux/Germany 'threatens certain ports with loss of market share since intermodal alternatives with shorter sea legs may appear more competitive'. The greatest single effect is the transfer from routes via the port of Gothenburg to land-based routes via the Oresund Bridge. In total, the results indicate that the redistribution from the SECA to the seas outside the SECA is rather limited. See Table 2. This is due to the limited possibilities that exist for avoiding the SECA for transports starting or ending in Sweden and also to the fact that some commodities are not suitable for land-based modes.

The calculations presented in Table 2 indicate that there will be a marginal increase in the transport performance on land and a reduction in sea transport performance of about one billion ton-kilometer, compared with the current 40 billion ton-kilometer within the Swedish sea territory. The sea ton-kilometer outside the Swedish territory are calculated to decrease by about 0.5 billion. Modal shifts are calculated to take place mainly to road in Sweden and to rail outside Sweden.

Table 2. Calculated difference between Scenario 1 and base scenario in Samgods study 2009.

Billion ton-km and % Billion ton-km

Inside Sweden Outside Sweden

Road +0.0 (+0.50%) 0.00

Rail +0.07 (+0.33%) +0.03

Sea -0.95 (-2.50%) -0.55

The 2009 study does not include elasticities, but it can be concluded that the demand for ton-kilometer by sea is quite inelastic. In Scenario 1 fuel cost increases of 12 - 81 % in the SECA are calculated to lead to 2.5 % less tonkilometer by sea in Sweden. The decrease for sea transports is calculated to be 10 % in Scenario 2 and 15 % in Scenario 3.

3.1.3. Impact study 2013

The 2013 study, Vierth et al. (2013), was a follow up of the 2009 study. The implementation year for IMO's sulphur directive had come closer and it was confirmed that the directive will be implemented as planned 1 January 2015. Several stakeholders discussed adjustments of the IMO directive and a postponement. Different technical solutions to meet the stricter sulphur requirements were and are developed. The Swedish rail track fees were increased in 2013 and further increases are planned. The Transport Administration discusses the possibility of using sea transports to relieve the rail infrastructure (Trafikverket (2011)).

In total 18 scenarios were proposed and developed by the agency Transport analysis (Trafikanalys (2013)). As in 2009, it is assumed that 0.1 % marine fuel oil replaces heavier fuel in the SECA so that the distance based costs increase by 0-23 % (low), 23-54 % (medium) or 40-76 % (high). In contrast to the 2009 study, the importance of higher diesel prices that lead to higher distance-based road transport costs inside and outside Sweden is also studied. The diesel price is assumed to 1) be constant, 2) increase by € 0.04 per litre or 3) increase by € 0.09 per litre based on, SWECO (2012). These cases correspond to the following changes in road transport costs: 1) constant, 2) increase by 3.6 %, and 3) increase by 7.2 %. The scenarios are tested assuming constant rail track fees (A) and increased rail track fees (B) in Sweden. (Rail track charges are assumed to be in the magnitude of 5 % of the rail transport costs.) The increase in track fees in Sweden varies between 38-51 % depending on the type of train and the rail transport costs in Sweden are assumed to increase by 2.5-3 %. See Table 3. The increase of the diesel price is to some extent dependent on the stricter sulphur requirements for marine fuel, while the increases of the rail track fees are independent.

Table 3. Scenarios for increased kilometer costs for sea transports and road transports in the 2013 study (rail track charges are assumed to be constant (A) or increase by 38-51 % depending on train type in the existing infrastructure plan (B)

Increased fuel costs for road (inside and outside Sweden) Increased km-cost for sea transports in Low (0-23%) Medium (23-54%) SECA High (40-76%)

1) € 0.00 per litre (constant road-km costs) Low 1 Medium 1 High 1

2) € 0.04 per litre (3.6% increased road-km costs) Low 2 Medium 2 High 2

3) € 0.09 per litre (7.2% increased road-km costs) Low 3 Medium 3 High 3

Scenario Medium 1A with medium increased sea transport costs in the SECA constant rail track fees and can be compared with Scenario 1 in the 2009 study. Due to the updated transport costs and calibration of the port flows the effects on the flows in the network and the modal split are calculated to be somewhat smaller in the 2013 study (than in 2009). This is especially true for the shift to ports in the north of Norway.

Below we focus on Scenario Medium 2B which includes medium increased sea transport costs in the SECA, medium diesel price increases and higher rail track fees in Sweden. This scenario was chosen to be the main scenario (by Transport Analysis). The results in Fig. 2 indicate that the total freight transports by sea in the Baltic Sea will increase and that there will be a slight increase for the ferry transports. The sea transports in the Kattegat along the Swedish coast are reduced, while the flows west of Sjaelland towards and through the Kiel Canal increase. The calculated increases and reductions are in general assessed to be plausible although the shift from the route west of Denmark to the route east of Denmark can be exaggerated.

In Scenario Medium 2B tail flows are calculated to increase mainly towards the Oresund Bridge while the volumes to/from the port of Gothenburg decrease. However, if the rail track fees in Sweden are kept unchanged in Scenario Medium 2A the decrease of the rail flows to/from Gothenburg is not obtained. For road, the overall pattern indicates a redirection of volumes from the ports on the Swedish east coast to ports on the Swedish west coast and to the ferry lines in the south of the country.

Fig. 2. Calculated difference between Scenario Medium 2B with increasing rail track fees and base scenario for sea transports in tons on the infrastructure network in the 2013 study (red colour shows increases and black colour decreases). K = Kattegat, Sj = Sjaelland, C = Kiel Canal, G

= Gothenburg, S = Stockholm.

The calculations in Scenario Medium 2B with increasing rail track fees indicate increases for rail and decreases for sea as well as for road both inside and outside Sweden. The increases for rail are about twice as large when constant track fees are assumed in Scenario Medium 2A with constant rail track fees. One must have in mind the assumption that the rail transport costs only increase in Sweden while the sea transport costs increase in the SECA and the road transport costs increase all over the world. The figures in Table 4 exemplify that rail transports can be replaced by combined sea and road transports.

Table 4. Calculated difference between Scenario Medium 2B in Samgods study 2013.

Billion ton-km and % Billion ton-km

Inside Sweden Outside Sweden

Road -0.20 (-0.50%) -0.35

Rail +0.30 (+1.42%) +0.55

Sea -0.35 (-0.88%) -0.75

The results for rail should be interpreted as potential increases since the model does not take into account capacity restrictions and demand for rail (unlike road) is near capacity in several parts of the network. Scenario High 3A with high transport cost increases for sea and road and constant rail transport costs was used to simulate the maximum use of the rail network. As expected, the model indicates that the transport flows (in tons) increase more in the sparse parts of the rail network (3.9-6.1 %) than in the total network (3.7 %). According to the Transport Administration, an average increase of about 5 % should be manageable.

As in the 2009 study, the demand for sea transports is rather inelastic. In the Scenario Medium 2B with increasing rail track fees, an increase in sea transport costs of 22-54 % is estimated to reduce ton-kilometer in Sweden by about 2 %. This means that elasticities in the range -0.09 to -0.04 are obtained which is of the same magnitude as the elasticities calculated with the European Transtools model (NEA (2007)).

3.2. Tapas

3.2.1. Model

Tapas (Transport and Production Agent-based Simulator) is an agent-based simulation model for quantitative impact assessment of policy and infrastructure measures. It takes into account how companies and other actors in a transport chain are expected to act under different conditions. The basic idea is that production and transport activities appear as consequences of ordering and planning processes that take place in order to fulfill customers' product demand. The choices of the actors in transport chains (customers of products and transport services, production and transport planners etc.) and the interactions between these actors are modelled. Tapas incorporates the complexity of the choices with respect to consignment size, route and mode and takes into account timetabled as well as demand-driven transports. The aim is to represent the distributed decision-making structure as it appears in transport chains.

Tapas is built using a two-tier architecture, with a physical simulator and a decision-making simulator, which are connected in such a way that the production and transport activities, that appear in the physical simulator, are initiated by decisions taken in the decision-making simulator. The physical simulator models product types, transport infrastructure, vehicles, terminals, and production facilities. Production costs, reloading costs, inventory holding costs and transport costs, consisting of time based costs, distance based costs and infrastructure fees, are included and the actors are assumed to be cost minimizers. In order to satisfy a customer's product demand, the agents participate in an ordering process that involves selection of which transport and production resources and infrastructure to use, as well as planning of how to use resources and infrastructure.

3.2.2. Impact study 2013

Within the framework of the Tapas study (Ramstedt & Holmgren, 2013) case studies were carried out for shippers that use sea transports for exports from Sweden to mainland Europe and the UK. One firm that exports paper and another firm that exports steel supplied information about their existing sea transport solutions and potential alternatives when IMO's stricter sulphur requirements in the SECA would become compulsory. The Tapas study is based on information delivered by these two firms, other sources and own assumptions. The same price increases for marine fuel and diesel as in the 2013 Samgods study are assumed. The rail track fees in Sweden are assumed to increase by 50 % instead of the 38-51 % of the 2013 Samgods study. Below we relate as for the Samgods study to the scenario with medium cost increases for sea and road transports and increased rail track fees.

Alternative response strategies are studied: (1) direct rail transport, (2) use of large vessels and (3) intermodal rail/sea transport via the port of Gothenburg. The alternatives are compared to the base scenario (today's transport chains) and pairwise. Limited rail capacity is described as a problem but not taken into account in the calculations.

In the "paper case", the firm's transports currently go by sea from three ports at the Northern part of the Swedish east cost to ports in the Benelux area or in the UK, from where the paper is distributed by truck. Typically, roro-vessels (with 8,200 tons loading capacity) go two times per week and call several ports at the Swedish east coast during a route. Given that the same transport chain is used, the transport costs from the different ports are calculated to increase by 10 to 12 % due to the stricter sulphur requirements in the SECA. The costs for direct rail transports are calculated to be 10-20% lower than the costs for the sea transport chain in the base scenario, given that there are operators that are willing to perform these transports (and that there is sufficient rail capacity). The use of vessels of double size is calculated to reduce today's transport costs by 15 to 18 %. However, loading and unloading in the ports need to be more efficient if time-tables are to be kept. Investments in larger vessels are needed and there can be infrastructure restrictions for larger vessels in some ports. These aspects are not taken into account in the calculations. The results indicate that the long rail transport from the paper factories to Gothenburg (around 1 000 km) makes the intermodal rail/sea transport via Gothenburg only competitive when vessels of double size are used between Gothenburg and the Benelux area. This alternative is calculated to lead to 7 % lower costs than in the base scenario.

In the "steel case", the firm exports steel products from a factory in the middle of Sweden to Denmark. An intermodal transport chain is used: rail to an east coast port (24 % of the transport distance), sea to a port in Denmark using a LOLO-vessel with 2,000 tons loading capacity (65 % of the transport distance), road to the clients (11 % of the transport distance). Given that the same transport chain is used, the transport costs are calculated to increase by around 3 %. The cost increase is lower since the sea share and the distance travelled in the SECA is lower. Direct rail transports are calculated to lead to 22 % lower transport costs compared to the existing intermodal solution. This is

mainly due to the reduction of the transfers. The costs for intermodal rail/sea transports via the port of Gothenburg are computed to be 9 % lower than the costs for today's solution. This solution is judged to be attractive for transports to Germany and the Benelux countries as well. In contrast to the "paper case" another port is chosen but no additional transfers are needed. The use of vessels that have a four times higher loading capacity is calculated to have 11% lower transport costs than today's solution. However it is a challenge to fill the larger vessel.

In both case studies, the results indicate that some of the alternatives are already competitive (given that there is sufficient rail capacity) and become more competitive when the IMO sulphur directive is in place. See Table 5.

Table 5. Calculated costs differences between medium and base scenario in Tapas study 2013.

Cost reductions compared to base

Comment

"Paper case"

- (1) Direct rail transports

- (2) Larger vessel (2x)

- (3) Rail/sea via Gothenburg

- (2+3)

"Steel case"

- (1) Direct rail transports

- (2) Larger vessel (4x)

- (3) Rail/sea via Gothenburg

10-16% 15-18% No reductions 7%

22% 11%

Rail capacity restrictions not taken into account

Requires more consolidation

Rail capacity restrictions not taken into account

Requires more consolidation, rail capacity restrictions not taken into account

Rail capacity restrictions not taken into account

Requires more consolidation

Rail capacity restrictions not taken into account

3.2.3. Discussion

The Swedish impact studies that assessed the consequences of IMO's stricter sulphur requirements in 2009 (Samgods) and 2013 (Samgods and Tapas) have somewhat different focuses. In the 2009 Samgods study, effects of increased fuel costs for sea transports, due to stricter sulphur requirements in the SECA and higher prices for crude oil, are simulated, everything else equal. The studies performed in 2013 comprise cost increases for sea transports for three different levels due to stricter sulphur requirements as well as cost increases for road in general and rail in Sweden and a wide range of scenarios is studied. We think that case studies for specific sea transport chains that are influenced by the stricter sulphur requirements, such as those based on the Tapas model, can be a useful supplement to the aggregated Samgods analysis that looks at all Swedish freight transports.

When it comes to the modal split calculated in the Samgods studies, the overall impact on sea transport demand in ton-kilometer is calculated to be relatively low, which is in line with the literature. Here we see a need for in-depth studies for different segments, i.e. different commodities, routes, vessel types, routes, route length, sea leg length, etc. The importance of these differentiations is confirmed in the other impact studies. Examples are Finnish Ministry of Transport and Communication (2009) and ISL (2010) which conclude that container ships will be particularly affected. ISL (2010) and Notteboom et al. conclude that medium-long routes will be most affected.

In addition, we see a need to study the choice of shipment size, vessel type and size (exploitation of economies of scale) as well as the degree of consolidation. As far as we know there are no results on these issues except from the findings in the Tapas study so far. The North East Cargo Link II project estimates modal shifts to rail and road but identifies also potential for sea transports as higher costs lead to incentives for optimizing routes and load factors, Malmqvist & Alden (2013). The possibilities for reducing sea transport costs per unit by switching to chains/vessels where it is feasible to exploit economies of scale is modelled within the Samgods. However, it needs to be validated if the strength of the cost reductions is correct.

4. Conclusions

Several studies have been carried out to estimate the possible impact of IMO's more stringent requirements for sulphur emissions in the SECA from 2015. Interestingly, all impact studies were carried out after the IMO decision

in 2008 (which means that the decision was taken based on other reasons). The studies focus on somewhat different aspects but all of them estimate how the stricter sulphur requirements affect the sea transport costs. Some studies estimate to what extent the changed competition interface will lead to modal shifts. This is typically done for specific transport corridors.

Sweden is the only country that has used a transport model comprising an international infrastructure network to study the impacts of more stringent sulphur requirements. This model was applied to calculate the transport effects in government commissions in 2009 and 2013. It is assumed that heavy marine fuel is replaced by 0.1 % marine gas oil. Impacts on the choice of transport chains, routes and ports due to higher sea transport costs are calculated.

The results indicate, for example, that shippers to some extent can reduce the increase in logistics costs in the situation with increased sea transport costs' by transferring flows from ports on the Swedish east coast to ports on the south coast (mainly short-distance ferry) or west coast. The results indicate that the relocation from SECA to the seas outside of SECA is rather limited, which is due to the limited options transports starting or ending in Sweden have for avoiding the SECA and the fact that some commodities are not suitable for land-based modes. The results are judged to be plausible but need to be interpreted with caution at the detailed level.

Modal back shifts from sea to rail and road are expected, but the shifts are quite modest, especially in the scenarios where not only the isolated impact of the more stringent sulphur requirements is addressed, but where higher prices for marine fuel in general (Scenario 2 and 3 in the study 2009) or higher diesel prices as well as higher rail track fees are assumed (several scenarios in the study 2013). It is shown that results are sensitive to the level of cost increases that are assumed for road and rail.

In the main scenario in the 2013 study, which includes medium increased sea transport costs in the SECA, medium diesel price increases and higher rail track fees in Sweden, a 1 % decrease of the ton-kilometer by sea in Sweden is computed. The highest decrease (15 %) is calculated in the 2009 study when high increased sea transport costs in the SECA due to stricter sulphur requirements and 150 % higher prices for crude oil inside and outside the SECA are assumed, everything else equal.

In addition to the shifts from sea to rail and road we see a need to capture possible adjustments within the sea mode. One important question is to what extent the increases in costs that are due to more stringent requirements, can be compensated for by improved efficiency of the transports, such as exploitation of economies of scale underway or in the ports or the level of consolidation.

Acknowledgements

The authors would like to thank the Swedish Maritime Administration, the Transport Analysis agency and the Centre for Transport Studies for funding.

References

AEA TNO, IVL & EMRC. (2009). Cost Benefit Analysis to Support the Impact Assessment accompanying the revision of Directive 1999/32/EC

on the Sulphur Content of certain Liquid Fuels. AEA Technology. COM(2011)144 final. (2011, 3 28). WHITE PAPER, Roadmap to a Single European Transport Area - Towards a competitive and resource

efficient transport system. European Commission. de Jong, G., Ben-Akiva, M., & Baak, J. (2008). Method Report Logistics Model in the Swedish Freight Transport Model system. Significance. Edwards, H; Bates, J; Swahn, H. (2008). Base Matrices Report ( Final version 13 March 2008). SIKA.

EMSA. (2010). The 0,1% Sulphur in Fuel Requirement as from 1 January 2015 in SECAs: An Assessment of Available Impact Studies and

Alternative Means of Compliance. EMSA (European Maritime Safety Agency). ENTEC. (2009). Impact Assessment for the Revised Annex VI of MARPOL, Commissioned by the UK Maritime and Coastguard Agency. London.

ENTEC. (2010). Study To Review Assessments Undertaken Of The Revised MARPOL Annex VI Regulations (Final report). Finnish Ministry of Transport and Communication. (2009). Sulphur Content in Ships Bunker Fuel in 2012. A Study on the Impacts of the New Regulation on Transport Costs by the Centre for Maritime Studies, University of Turku. Helsinki: Finnish Ministry of Transport and Communication (MTC).

Finnish Transport Safety Agency. (2013). Merenkulun uusien ympäristömääräysten aiheuttamien kustannusten kartoittaminen. Helsinki: http://www.trafi.fi/filebank/a/1384763246/c056f2edd518847970e2117c1e982034/13636-Trafin_julkaisuja_24-2013_-_Merenkulun_uusien_ymparistomaaraysten_kustannukset.pdf. Holmgren,J; Nikopoulou, Z; Ramstedt, L; Woxenius, J. (2014). Modelling modal choice effects of regulation on low-sulphur marine fuels in

Nothern Europe . Transportation Research Part D, 62-73. IMO. (2014, 8 25). Prevention of Air Pollution from Ships. Retrieved from http://www.imo.org/blast/mainframe.asp?topic_id=233. ISL. (2010 ). Reducing the sulphur content of shipping fuels further to 0.1% in the North Sea and Baltic Sea in 2015: Consequnences for shipping

in this area. . Bremen : Institute of Shipping and Economics and Logistics . Jiang, L; Kronbak, J; Christensen, L P. (2014). The costs and benefits of sulphur reduction measures: Sukphur scrubbers versus marine gas oil.

Transportation Research Part D, 19-17. Kehoe et al. (2010). Impact Study of the future requirements of Annex VI of the MARPOL Convention on Short Sea Shipping, Task 2. SKEMA.

Lindstad, H; Asbjornslett, B E; Jullumstro, E. (2013). Assessment of profit, cost and emissions by varying speed as a function of sea condistions

ans freight market e. Transportation Research Part D 19 (2), 5-15. Malmqvist, G; Alden, B. (2013). Sulphur Regulation in the Baltic Sea: Scenarios for the Midnorduc Region: Threats and Opportunisties. Municipality of Sundsvall.

Mellin, A., Vierth, I., & Karlssson, R. (2013). Uppdaterad analys av transporteffekter av IMO:s skärpta emissionskrav - modellberäkningar pa

uppdrag av Sjöfartsverket. VTI (VTI notat 17-2013). NEA. (2007). TRANSTOOLS Modal split model - Revisions for Transtolls Version 1.3. Rijswijk: NEA.

Notteboom et al. (2010). Analysis of the Consequences of Low Sulphur Fuel Requirements, Commissioned by European Community

Shipowners' Associations (ESCA)t. ESCA. Ramstedt, L., & Holmgren, J. (2013). Agentbaserad analys av svaveldirektivet - tva fallstudier. Vectura & BTH.

Schinas, O., & Stefanakos, C. N. (2014). Selecting technologies towrds compliance with MARPOL Annex VI: The perspective of operators.

Transportation Researcg Part D, 28-40 (Vol. 28). SWECO. (2012). Effekter av svaveldirektivet - En rapport till Svenskt Näringsliv. Stockholm:

http://www.transportgruppen.se/Documents/Publik_F%C3%B6rbunden/Sveriges_Hamnar/Rapporter/Effekter%20av%20svaveldirektivet%2 0Sweco%20augusti%202012.pdf.

Swedish Maritime Administration. (2009). Konsekvenser av IMO:s nya regler för svavelhalt i marint bränsle. Norrköping: Sjöfartsverket (Rapport 0601-08-03406).

Trafikanalys. (2013). Konsekvenserna av skärpta krav för svavelhalten i marint bränsle - slutredovisning. Stockholm: Trafikanalys (Rapport 2013:10).

Trafikverket. (2011). Järnvägens behov av ökad kapacitet - förslag pa lösningar för aren 2012 - 2021. Borlänge: Trafikverket (TRV 2011/17304, September 2011).

Vierth, I., Lord, N., & Mellin, A. (2009 a). Transporteffekter av IMO:s skärpta emissionskrav - modellberäkningar pa uppdrag av Sjöfartsverket.

VTI (VTI-notat 15/2009) in Swedish (English summary). Vierth, I.; Lord, N.; McDaniel, J. (2009 b). Representation av det svenska godstransport- och logistiksystemet. VTI (VTI Notat N17A). Vierth, I., Mellin, A., & Karlsson, R. (2013). Analys av effekter av IMO:s skärpta svavelkrav - Modellberäkningar pa uppdrag av Trafikanalys. VTI (VTI Rapport 33/2013).