Evaluation of methods to estimate the consequence costs of an oil spill

Transcription

1 SEVENTH FRAMEWORK PROGRAMME SST 2007 TREN 1 SST Maritime and logistics co-ordination platform SKEMA Coordination Action Sustainable Knowledge Platform for the European Maritime and Logistics Industry SKEMA Consolidation Study Evaluation of methods to estimate the consequence costs of an oil spill WP No2 SKEMA Consolidation Studies Task T3.2: Methods for assessing safety and security performance SKEMA Subject Index: SE3.2.2 Responsible partner: VTT Contributing partner: CETLE Planned submission date: Version 1-31/12/2008; Version Actual submission date: Ver 1 18/12/2008 Ver Distribution group: Consortium Dissemination level: PU (Public) Contract Number: Project Start Date: 16 th June 2008 End Date: 15 th May 2011 Co-ordinator: Athens University of Economics and Business

2 Document summary information Version Authors Description Date 1 T Nyman VTT Final draft 18/12/ T Nyman VTT Draft ver 2 3/6/ T Nyman VTT Final draft (updated) 30/6/2010 Checked by Task and WP Leader Checked by Peer Review Checked by Quality Manager Approved by Project Manager Quality Control Who Date SE 3.2 Methods for assessing safety and security performance Review of collision and grounding risk analysis methods which can utilize the SE3.2.1 historical AIS data and traffic patterns in seawaters SE3.2.2 Evaluation of methods to estimate the consequence costs of an oil spill SE3.2.3 Dynamic risk management methods ship risk indexes 2

3 Summary The objective of this Consolidation Study is for the purposes of the cost-benefit-analysis in the FSA process to bring out the different methods to estimate the consequence costs the oil spill has and different factors having an influence on the oil spill. Due to the huge scatter in the statistical oil spill cost values it is difficult to give exact monetary values which would be generally valid. In the guidelines of the IMO FSA process [25], there have not been any directions how the environmental risk should be evaluated. This was noticed in IMO and the Correspondence Group on Environmental Risk Evaluation Criteria was established. This report presents some preliminary findings of the work of the Correspondence Group as well as gives some data and methods which can be used in the cost benefit analysis of the FSA process. In the stdy, first the general statistics related to oil spills and their consequence costs are studied. Second, the work of the IMO Correspondence Group on Environmental Risk Evaluation Criteria is described. Third, the International Oil Pollution Compensation Fund (IOPCF) and the oil spill consequence cost compensation principles are described. Fourth, the different regression studies performed utilising the IOPCF oil spill statistics are reviewed. Fifth, examples of data and methods which can be used in different phases of the FSA process are presented. 3

4 Evaluation of methods to estimate the consequence costs of an oil spill Contents SUMMARY OBJECTIVES TARGET STAKEHOLDERS GLOSSARY TERMS APPROACH SPECIFIC ISSUES AND TOPICS TO BE ADDRESSED Introduction Oil spills and their consequences Preliminary results of the work carried by the Correspondence Group on Environmental Risk Evaluation Criteria IOPCF data [24] Different regression analyses performed based on IOPCF oil spill cost data Example process for assessment of consequence cost of an oil accident CONCLUSION

5 1 Objectives The objective of this Consolidation Study is for the purposes of the cost-benefit-analysis in the FSA process to bring out the different methods to estimate the consequence costs the oil spill has and different factors having an influence on the oil spill. Due to the huge scatter in the statistical oil spill cost values it is difficult to give exact monetary values which would be generally valid. In the guidelines of the IMO FSA process [25], there have not been any directions how the environmental risk should be evaluated. This was noticed in IMO and the Correspondence Group on Environmental Risk Evaluation Criteria was established. This report presents some preliminary findings of the work of the Correspondence Group as well as gives some data and methods which can be used in the cost benefit analysis of the FSA process. 2 Target stakeholders Maritime administrations Ship owners Port authorities Policy makers Maritime Operational Centers/Coast Guards/OPRC/SAR/VTM 3 Glossary terms Bunker oil: oil used by vessels as fuel oil Cargo oil: oil carried by tankers in cargo tanks Oil spill: An oil spill is the release of a liquid petroleum hydrocarbon into the environment due to human activity, and is a form of pollution. HELCOM: Helsinki Commission or The Baltic Marine Environment Protection Commission is the governing body of the Convention on the Protection of the Marine Environment of the Baltic Sea Area (Helsinki Convention). HELCOM works on protection of the marine environment of the Baltic Sea. Total spill costs components could be: clean-up costs, mitigating costs/preventing measures, property damage, economic losses (fisheries, tourism-related, etc.), environmental damage (reinstatement/restoration of the impaired environment, longterm environmental effects, reasonable legal costs). 5

6 CATS criterion: (cost to avert one tonne of oil spilled) is a function of: costs, assurance factor reflecting society's willingness to pay, and possibly other factors (for example, intangible environmental impacts, etc.). CATS for an RCO is the cost of the RCO divided by the tonnes of oil spill averted by the RCO. 4 Approach The study is based on literature search and on the material collected during performed FSA projects. 5 Specific issues and topics to be addressed 5.1 Introduction In collision and grounding accidents the possibility of oil outflow is significant. Depending on the ship type the spilled oil may be bunker or cargo oil. In addition the spilled oil may be crude oil or heavy or light distillates. The size of the oil spill depends on the ship size and structure and damage mechanics. The scatter in the reported damage costs obtained from ship accidents having oil spill is huge. Especially the evaluation of the costs of the environmental damages is difficult. There is mutual understanding that the following technical factors have an essential influence on the cost of oil spills [1]: Type of oil, Physical, biological and economical characteristics of the spill location, Weather and sea conditions, Amount spilled and rate of spillage, and Time of the year, and effectiveness of clean-up. The interactions between factors explained in the following chapters are complex, which make cost predictions based on simple parameters very unreliable (Figure 3). 6

7 In the following, first the general statistics related to oil spills and their consequence costs are studied. Second, the work of the IMO Correspondence Group on Environmental Risk Evaluation Criteria is described. Third, the International Oil Pollution Compensation Fund (IOPCF) and the oil spill consequence cost compensation principles are described. Fourth, the different regression studies performed utilising the IOPCF oil spill statistics are reviewed. Fifth, examples of data and methods which can be used in different phases of the FSA process is presented. Figure 1. Factors determining per unit oil spill clean-up costs (Etkin, 1999). 5.2 Oil spills and their consequences In the last decade, the number of oil spills and the total quantity of oil spilled in the seas have declined which can be seen in Figure 2, which is based on the data provided by the International Tanker Owners Pollution Federation (ITOPF). ITOPF has maintained a database of more than 10,000 oil spills from tankers, combined carriers and barges and shows the number of spills per year of 7 tonnes or more for the period Most spills are small (7 700 tonnes). The same downward trend is apparent in the total annual quantity of oil spilled during the last decade [21]. 7

8 Figure 2 Annual quantity of oil spilled and number of spills [21]. Reduction of oil pollution is one of the stated goals of new regulations, including the implementation of double hulls for tanker vessels. The management of safety at sea is based on a set of accepted rules that are, in general, agreed upon through the International Maritime Organization (IMO). Many of these regulations aim to reduce environmental risk and, more precisely, the risk that relates with accidental oil spillage. It can also be argued that much of the maritime safety policy worldwide has been developed in the aftermath of serious accidents (such as Exxon Valdez, Erika and Prestige in case of oil pollution). A big chapter that has only recently opened concerns environmental risk evaluation criteria. At the 55th session of Marine Environment Protection Committee (MEPC) that took place in 2006, the IMO decided to act on the subject of environmental criteria. At the 56th session of MEPC (July 2007) a correspondence group (CG), coordinated by Greece, was tasked to look into all related matters, with a view to establishing environmental risk evaluation criteria within Formal Safety Assessment (FSA). It can be seen from Table 1 that the per unit cleanup costs vary by region being highest for Asia 33,300 USD/tonne, followed by the United States (average around 24,000 USD/tonne). The figures are in 2006 US dollars. Also in the Europe the alternation is 8

9 significant, varying from around 78 to 23,100 USD (2004 values) (Table 2) [14]. In Figure 3 the unit clean-up costs of oil spill are presented as a function of the spill size. From the figure it can be seen that the bigger the oil spill is the lower the unit costs are [20]. Table 1 Average cleanup costs per tonne oil spilled [12]. Region Cost per tonne spilled (USD per tonne) Share of global oil tanker traffic in region (%) Middle East 1,300 7 South America 3, Africa 3, Oceania 6,900 2 Europe 13, North America 24, Asia 33, Weighted global average 15, US$ / ton ,700 1,700-3,400 Spill size [tons] ,400-34, > 34,000 Figure 3 Per-unit marine oil spill clean-up costs by spill size for non-us Spill, in 1999 US$. [20]. 9

10 Table 2 Average per unit marine oil spill cleanup costs in various European countries [14]. Cost for oil spill cleanup Country US$ / gallon US$ / ton Denmark , Estonia , Finland , France , Germany , Greece , Ireland , Italy , Latvia , Lithuania Netherlands , Norway , Spain Sweden , UK , Yugoslavia , Average , Yugoslavia , Average , Preliminary results of the work carried by the Correspondence Group on Environmental Risk Evaluation Criteria At its fifty-ninth session, the Marine Environment Protection Committee recalled that MEPC 56 had noted that the one matter that needed consideration within the context of the Formal Safety Assessment Guidelines relevant to its work was the draft Environmental Risk Evaluation Criteria. In this connection, the need was recognized to carry out a more in-depth analysis of the proposed environmental risk evaluation criteria 10

11 for the purpose of the Formal Safety Assessment (FSA) before inclusion of such criteria in the IMO FSA Guidelines (MSC/Circ.1023-MEPC/Circ.392, as consolidated in MSC 83/INF.2). MEPC 59 noted that more time was needed to discuss the issues which seemed more complex than originally thought and so as to pave the way for the work of the working group at MEPC 60, also agreed to re-establish the correspondence group to prepare a basic document under the coordination of Professor Harilaos N. Psaraftis (Greece), with the following Terms of Reference: Using documents MEPC 59/17, MEPC 59/17/1 and MEPC 59/INF.21 as a basis, as well as taking into account the comments received at MEPC 59, the correspondence group was instructed to: 1. recommend in Step 4 of the FSA an appropriate volume-dependent CATS global threshold scale or function for ascertaining if a specific Risk Control Option (RCO) is cost-effective, including its integration within the FSA methodology; 2. recommend a way of combining environmental and safety criteria for those RCOs that effect both environmental and fatality risks; 3. conclude on an appropriate risk matrix or index for environmental criteria; 4. recommend an appropriate ALARP region and F-N diagram, including an appropriate value for the slope of the F-N curve; 5. address the issue of collection and reporting of relevant data; 6. prepare draft terms of reference for a working group at MEPC 60; and 7. submit a written report to MEPC 60. According to the report of the working group on environmental risk evaluation criteria [23] the Group noted that research work independently conducted by Japan (MEPC 59/17/1), Norway (MEPC 59/INF.21) and Greece (MEPC 60/17, annex 2) led to very similar non-linear functions of total spill cost versus spill weight, obtained by regression, as shown in Table 3 and Table 4 and Figure 4 below. Each of the above mentioned countries have performed statistical analyses of the oil spill cost data collected by the International Oil Pollution Compensation Fund (IOPCF). 11

12 Table 3 Non-linear functions of total spill costs (obtained by regression) [23]. Table 4 Total unit spill cost (total spill cost / spill weight) [23]. Figure 4 Total unit spill cost (US$/tonne) vs log(v) for V tonne [23]. Among the three non-linear regressions performed, the one proposed by Greece was considered most conservative (resulting in larger cost figures per tonne of oil spilled 12

13 except for spills less than 4 tonnes). Thus, this regression formula was proposed as a basis. When utilising the environmental oil spill cost evaluation method described in this chapter it should be noted the DISCLAIMER printed on the front page of the report: As at its date of issue, this document, in whole or in part, is subject to consideration by the IMO organ to which it has been submitted. Accordingly, its contents are subject to approval and amendment of a substantive and drafting nature, which may be agreed after that date IOPCF data [24] All of the three regression analyses discussed in chapter were based on the data of the International Oil Pollution Compensation Fund (IOPCF). The International Oil Pollution Compensation Funds (IOPC Funds) are three intergovernmental organisations (the 1971 Fund, the 1992 Fund and the Supplementary Fund) which provide compensation for oil pollution damage resulting from spills of persistent oil from tankers. The International Oil Pollution Compensation Funds (IOPC Funds) are part of an international regime of liability and compensation for oil pollution damage caused by oil spills from tankers. Under the regime the owner of a tanker is liable to pay compensation up to a certain limit for oil pollution damage following an escape of persistent oil from his ship. If that amount does not cover all the admissible claims, further compensation is available from the 1992 Fund if the damage occurs in a State which is a Member of that Fund. Additional compensation may also be available from the Supplementary Fund if the State is a Member of that Fund as well. There are at present three IOPC Funds: the 1971 Fund, the 1992 Fund and the Supplementary Fund. These three intergovernmental organisations were established at different times (1978,1996 and 2005 respectively), have different maximum amounts of compensation and have different Member States. The membership of the 1992 Fund is increasing. The Supplementary Fund was established to supplement the compensation available under the 1992 Civil Liability and Fund Conventions with an additional third tier of compensation. Membership of the Supplementary Fund is optional and any State 13

14 which is a Member of the 1992 Fund may join. The membership of the Supplementary Fund is expected to increase fairly quickly. Due to a number of denunciations of the 1971 Fund Convention, this Convention ceased to be in force on 24 May 2002 and the 1971 Fund therefore no longer has any Member States. The 1971 Fund will continue to deal with a number of incidents which occurred in 1971 Fund Member States before that date. The three organisations have a joint Secretariat, based in London. The IOPC Funds are financed by levies on certain types of oil carried by sea. The levies are paid by entities which receive oil after sea transport, and normally not by States. Anyone who has suffered pollution damage in a Member State may make a claim against the IOPC Funds for compensation. Information on the types of claims which are admissible can be found in the Claims Manuals. The shipowner is normally entitled to limit his liability under the 1992 Civil Liability Convention. The limits were increased by some 50.37% on 1 November 2003 as follows. The increased limits apply to incidents occurring on or after that date: a) for a ship not exceeding units of gross tonnage, Special Drawing Rights (SDR) (US$6.6 million); b) for a ship with a tonnage between and units of tonnage, SDR (US$6.6 million) plus 631 SDR (US$923) for each additional unit of tonnage; and c) for a ship of units of tonnage or over, SDR (US$131 million). If it is proved that the pollution damage resulted from the shipowner's personal act or omission, committed with the intent to cause such damage, or recklessly and with knowledge that such damage would probably result, the shipowner is deprived of the right to limit his liability. The Working Group was informed that claims in respect of pollution damage fall under one of the following broad categories: 1. mitigating/preventive measures (including clean-up); 2. damage to third party property due to oiling; 14

15 3. economic loss including fishery and tourism-related losses; and 4. reinstatement/restoration of the impaired environment Different regression analyses performed based on IOPCF oil spill cost data Introduction In the following the results of the regression analyses performed by Kontovas et al., Yamada, and Skjong et al are presented Kontovas et al.[21] The paper reports on recent analysis of oil spill cost data assembled by the International Oil Pollution Compensation Fund (IOPCF). Regression analyses of clean-up costs (based on data of 84 incidents) and total costs (based on data of 91 incidents) were carried out, after taking care to convert to current prices and remove outliers. The following equations were obtained: Cleanup cost = 44,435 V (1) Total cost = 51,432 V (2) And the equations for unit costs were obtained dividing the above formulas by V: Unit cleanup cost = 44,435 V (3) Unit total cost = 51,432 V (4) The marginal costs can be obtained by differentiating regression formulas (1) and (2) with respect to V as follows: Marginal clean-up cost = 28,616 V (5) 15

16 Marginal total cost = 37,442 V (6) The marginal costs are interpreted as the additional costs if one more tonne of oil is spilled. In the first place, the results of this analysis have been useful in the context of the ongoing discussion within the International Maritime Organization (IMO) on environmental risk evaluation criteria. Furthermore, these results can be useful in estimating the benefit of regulations that deal with the protection of marine environment and oil pollution prevention Yamada [27] The purpose of the paper was to consider a practical way to estimate the cost of oil spills from ships within the framework of establishing environmental risk evaluation criteria in International Maritime Organization (IMO). Regression analysis between the cost of oil spills and the weight of oil spilled (oil spill weight) was carried out using historical oil spill data of 99 incidents from tankers reported by International Oil Pollution Compensation (IOPC) Funds. A nonlinear regression formula between the cost of oil spills and the oil spill weight is estimated from the historical data: C = 38,735 W 0.66 Where C denotes total costs including cleanup, indemnification, fishery-related, property damage, tourism-related, loss of income, preventive studies, other and environmental damage costs actually paid by IOPCF. W denotes the oil spill weight. A critical value of cost to avert one tonne of spilled oil (CATS cr ) is obtained based on the CATS value on 60,000 US$/ton. CATS cr = 25,441 W -0:34 16

17 CATS cr obtained by the present study is compared with that obtained by previous work. This study shows that the cost of oil spills estimated by the present regression formula is in fairly good agreement with the mean value obtained from historical data while the CATS cr gives relatively larger costs and shows the upper bound of the cost of oil spills Skjong et al. [12] In the SAFEDOR project the concept of Cost of averting a spill is used and expressed as an equation: Costs of averting a spill < F Costs of an occurred spill (1) F is an Assurance parameter that is F > 1 and it represents that spending money on preventing oil spills is always preferable to costs related to occurred spill. The value of F depends on decision makers and how the cost is allocated. The SAFEDOR report recommends as regards to the environmental risk following: Reference is made to the cost effectiveness of averting pollution. The decision parameter proposed is called CATS (Cost of Averting one Tonne of oil Spilled). Other pollutants that are considered having other damaging effect to the environment than oil (crude oil) are scaled to the equivalents effect by deriving conversion factors. The CATS value is presented in Table 5. Table 5 CATS value for environmental risk 17

18 5.2.4 Example process for assessment of consequence cost of an oil accident General When performing a FSA study for a certain sea area the local conditions should be taken carefully into account. The following estimations have to be made when calculating the consequence costs of an oil spill: oil spill probability in accident size of oil spill effectiveness of oil combating operations at sea and their costs oil spreading and width of the contaminated coast area costs of oil combating operations at sea and shore cleanup costs costs of environmental damage costs of damage caused to sea-dependent means of livelihood In the following, some methods and data needed for performing the above mentioned assessments are presented. In order to avoid duplication of costs when using different calculation methods, statistical cost values or other sources of information it has to be made certain which cost items are included in the given figures Probability of an oil spill in an accident Usually the oil accident has been seen as a result of an oil tanker cargo spill. However, a bigger threat than cargo spills are oil spills from bunker tanks of all vessels. Contrary to tankers transporting heavy oil as cargo, bunker tanks are not required to have a double hull structure as protection. Still in 1995 according to the questionnaire performed by the Finnish Maritime Administration roughly 90% of merchant vessels used heavy fuel oil as bunker fuel [3]. 18

19 In a detailed study performed by Safetec UK Ltd [2] the objectives were to collect data on the environmental sensitivity of the UK coastline and surrounding waters as well as carrying out a risk assessment using the most up-to-date ship routeing information to estimate the pollution risks in UK waters. In the report the average probabilities of cargo and bunker spills in different casualty types were estimated. The results are collected in Table 6. Table 6 Cargo and bunker spill probability [2]. Spill type Spill probability (Spills per casualty) Collisions Groundings Cargo spill Bunker spill Estimation of spill size In case of an accident of such severity that it results in outflow of cargo, a number of cargo tanks will leak. A portion of the tank content will then escape into the sea, the amount depending on the type of damage, the vessel s loading condition, the properties of the cargo and whether the vessel is a single or double hull ship. Some of the collision and grounding risk assessment methods also estimate the amount of spilled oil based on the estimation of the structural damage of the ship, but usually the estimates of the average spill sizes are sufficient. The average cargo spill size is estimated in a study published by HELCOM [4]. The estimate of the spill in ship to ship collisions is 1/200 of the total cargo for double hull and 1/40 of the total cargo for single hull tankers. In groundings, the estimate of the average size of the cargo spill is 1/130 of the total cargo for double bottom and 1/24 of the total cargo for single bottom tankers. The amounts are for cargoes which are lighter than water and insoluble. The fuel oil tanks usually locate in the engine room where a double hull arrangement is not required. In the case of bunker spill, all oil is assumed to outflow from tanks penetrated in collision and 50% in grounding assuming that the bunker tanks are 98% full [5]. 19

20 Effectiveness of oil combating According to a Finnish oil combating specialist the costs of shore cleaning operations are ten times more expensive than collecting the oil at sea and hundred times more expensive than pumping the oil from the damaged vessel [6]. Many factors determine the success of oil combating operations at sea: e.g. the location of the oil spill, water depth in the spill area, season, weather conditions, number of available oil combating vessels and their collecting capability, type of oil etc. As an example is the current Finnish oil combating vessels. They are designed to collect heavy oils but their possibilities to collect light oil products at sea are limited. Fortunately, the light oil spills usually evaporate into air with time. The vessels are capable of operating in sea conditions with significant wave heights of 1 1,5 metres and their collecting capabilities vary between m3/24h [7]. In the Gulf of Finland, the probability of suitable conditions is 80%, in the Northern Baltic 60% Oil drifting In spite of the oil combating operations, part of the spilled oil is usually drifting ashore. Based on the experimental tests in [11] the drifting speed of oil in water is about 2 3% of the wind speed. For the cost estimation, it is essential to get probabilistic estimates of the width of the spill and how fast the oil will reach the shore. Methods to estimate the oil drifting has been developed which take into account the statistics of the wind and current conditions in the sea area considered. In [10], the state-of-the-art in oil spill modelling is summarized, focusing primarily on the years from 1990 to the present. All models seek to describe the key physical and chemical processes that transport and weather the oil on the shore and in the sea. Current insights into the mechanisms of these processes and the availability of algorithms for describing and predicting process rates are discussed. Advances are noted in the areas of advection, spreading, evaporation, dispersion, emulsification, and interactions with ice and shorelines. Knowledge of the relationship between oil properties, and oil weathering and fate, and the development of models for the evaluation of oil spill response strategies are summarized. Specific models are used as examples where appropriate. Future directions in these and other areas are indicated. 20

21 Seatrack Web [8] is a web-based, operational oil drift forecast system applied for the Baltic Sea. The system has been developed in co-operation by several institutes around the Baltic Sea. SMHI (Sveriges Meteorologiska och Hydrologiska Institut) in Sweden is responsible for maintaining and running Seatrack Web in its server. The Seatrack Web graphical user interface has been available for Baltic wide use since Seatrack Web contains information from several numerical models as well as some meteorological observations. HIROMB (High Resolution Operational Model for the Baltic Sea) is the name of the hydrodynamic model component in the Seatrack Web system. HIROMB calculates daily forecasts of the Baltic Sea temperature, salinity, currents, ice situation and water levels. It receives also information of river inflows calculated with a river runoff model covering the entire Baltic drainage basin. Atmospheric forcing fields for HIROMB and drift calculation are provided from the Swedish HIRLAM atmospheric model. In paper [9] the complex of numerical models is presented simulating the wind waves, marine circulation and oil spill dynamics in the coastal zone with very winding coastal line, many islands and inlets and steep banks. Wind, cloudiness and air temperature are given from routine weather forecasting model HIRLAM (together with high resolution models ETA and ETB). Sea currents and water temperature are calculated by 3D nonhydrostatic circulation model FRESCO (Finnish-Russian-EStonian Cooperation) together with two-equation (k- ) turbulence model. The net current speed caused by wind waves (Stokes drift) is calculated using the narrow-directionality approximation wind wave model. Oil Spill Modelling System OSMS and backward models are applied for offline oil spill simulations using time-averaged flow, sea temperature, and eddy viscosity fields as input data. A set of OSMS models have been developed and incorporated in computer system SPILLMOD which can be used in risk assessment due to oil spill accidents and simulate the different strategies of oil spill countermeasures in marine environment Total response costs Oil spill cleanup response costs depend on a variety of factors, most notably, location, oil type, spill size, and cleanup strategy, making it difficult to develop a universal per- 21

22 unit cost factor. The study developed by Etkin [13] analyzes marine oil spill cleanup costs on the basis of country, proximity to shoreline, spill size, oil type, degree of shoreline oiling, and cleanup methodology to determine how each of these factors impacts per-unit cleanup costs. The results show that oil spill responses in different countries and regions of the world vary considerably in their costs most likely due to differences in cultural values, socio-economic factors, and labour costs. Location, oil type, and spill size also factor heavily in determining cleanup costs. Near shore spills and in-port spills are 4-5 times as expensive to clean up as offshore spills. Responses to spills of heavy fuels are more than ten times as expensive as spill responses for lighter crudes and diesel fuels. Spill responses for spills under 30 tonnes are more than ten times as expensive, on a per-unit basis, as for spills of 300 tonnes. The cleanup cost estimation modelling technique can be applied to marine spills of different types. The model is developed from updated cost data collected from case studies of over 300 spills in 40 nations. The model takes into account oil type, location, spill size, cleanup methodology and shoreline oiling to deduce a per-unit cleanup cost figure. The estimation is based on the following formulae: Cei = Cui Ai = estimated total response cost for scenario i, C ui = C li t i o i m i s i = response cost per unit for scenario i; C li = r i l i C n = cost per unit spilled for scenario i, Symbol Explanation C n General cost per unit spilled in nation, n, t i Oil type modifier factor for scenario i, o i m i Shoreline oiling modifier factor for scenario i Cleanup methodology modifier factor for scenario i s i Spill size modifier factor for scenario i, r i Regional location modifier for scenario i, l i Local location modifier for scenario i, A i Special spill amount for scenario i, l oi Length of oiled shoreline In the article tables for determining the parameters in the equation are presented taking into account the oil type, spill size, spill location and cleanup method. 22

23 Costs of damage caused by oil spill on sea-dependent means of livelihood Generally presented spill cost prediction models usually use the spilled quantity as the basis for total cost estimation. However, in reference [17] it has been presented based on Swedish experiences that the total damage and clean up costs correlate better with the total length of the polluted shoreline than with the total spill volume. Tourism and fishery sectors generally represent the sectors suffering the highest economic losses due to the spill. In order to estimate the possible losses within the tourism sector, regional statistics on the total consumption value minus production costs was compiled from official statistics. The figures were then distributed to county or municipal level by using regional or local accommodation statistics as weighting factor. The derived figures for each county or coastal municipality were then divided by the total coast length to form a local sensitivity index. The index multiplied by a relative damage rate and by the contaminated beach length given by the scenario case, then represents the estimated socioeconomic damage cost in the tourist sector of the community. A similar approach was used to formulate a socioeconomic sensitivity index for fishery. The damage costs to tourism were determined using the formula: Damage tourism Coastdamage m Sensitivity / m Damagerate % / 100 where: Damagetourism = damage costs of an oil spill to tourism in, Coastdamage = the length of polluted coastline in metres, Sensitivitytourism = the socioeconomic sensitivity index of tourism in /metre Damagerate = the pollution rate of the coastline tourism Oil Spill Cost Estimation Model (BOSCEM) The EPA Basic Oil Spill Cost Estimation Model (BOSCEM) was developed to provide the US Environmental Protection Agency (EPA) Oil Program with a methodology for estimating oil spill costs, including response costs and environmental and socioeconomic damages, for actual or hypothetical spills. The model is described in [18]. The model can quantify relative damage and cost for different spill types for regulatory impact evaluation, contingency planning, and assessing the value of spill 23

24 prevention and reduction measures. EPA BOSCEM incorporates spill-specific factors that influence costs spill amount; oil type; response methodology and effectiveness; impacted medium; location-specific socioeconomic value, freshwater vulnerability, habitat/wildlife sensitivity; and location type. Including these spill-specific factors to develop cost estimates provides greater accuracy in estimating oil spill costs than universal per-gallon figures used elsewhere. The model s basic structure allows for specification of response methodologies, including dispersants and in situ burning, which may have future applications in freshwater and inland settings. Response effectiveness can also be specified, allowing for analysis of potential benefits of response improvements. The reference [18] includes the formulas and tables of parameters which can be used in cost estimation for specific cases. The method is outlined in Figure 5. Figure 5 EPA BOSCEM basic interrelationships between oil spill base costs and modifiers. The circled numbers indicate steps of input by user [18]. 24

25 Other costs In addition to the costs mentioned above there are costs all of which are not specific to oil spills but which have to be taken into account when performing the cost-benefit analysis in a FSA study. Some of the costs are collected into Table 7. Table 7 Other accident costs. Cost item Explanation Costs of the loss of cargo Specific to oil spills. Depends on the oil type and spill size Vessel salvage and repair costs of On the average per accident (2008 price damaged vessel level) [16]: 1,1 M collisions 0,8 M groundings Personal injuries (2005 price level) [6] Loss of life Permanent injury Serious temporary injury Minor temporary injury Authority services / accident [6] 25

26 6 Conclusion The objective of this Consolidation Study is for the purposes of the cost-benefit-analysis in the FSA process to bring out the different methods to estimate the consequence costs the oil spill has and different factors having an influence on the oil spill. Due to the huge scatter in the statistical oil spill cost values it is difficult to give exact monetary values which would be generally valid. In the guidelines of the IMO FSA process [25], there have not been any directions how the environmental risk should be evaluated. This was noticed in IMO and the Correspondence Group on Environmental Risk Evaluation Criteria was established. This report presents some preliminary findings of the work of the Correspondence Group as well as gives some data and methods which can be used in the cost benefit analysis of the FSA process. In the stdy, first the general statistics related to oil spills and their consequence costs are studied. Second, the work of the IMO Correspondence Group on Environmental Risk Evaluation Criteria is described. Third, the International Oil Pollution Compensation Fund (IOPCF) and the oil spill consequence cost compensation principles are described. Fourth, the different regression studies performed utilising the IOPCF oil spill statistics are reviewed. Fifth, examples of data and methods which can be used in different phases of the FSA process are presented. 26

27 References 1. White, I. C., Molloy, F. C Factors that determine the cost of oil spills. International Oil Spill Conference IOSC p May , Miami Beach, FL, USA. The International Tanker Owners Pollution Federation Limited. 2. Safetec UK. Identification of Marine Environmental High Risk Areas (MEHRAs) in the UK. Doc. No.: ST-8639-MI-1-Rev 01. December Mäkelä, K. et.al. Calculation system of the emissions of the Finnish water-borne traffic (In Finnish). MEERI VTT Yhdyskuntatekniikka. Research Report 535. (MOBILE report M2T9916-7). Espoo HELCOM. Study of the Risk for Accidents and the Related Environmental Hazards from the Transportation of Chemicals by Tankers in the Baltic Sea Area. Baltic Sea Environment Proceedings No. 34. Helsinki Commission Michel & Winslow Cargo Ship Bunker Tanks: Designing to Mitigate Oil Spillage. SNAME Joint California Sections Meeting Ramboll Finland Oy. Costs of vessel traffic accidents. (In Finnish) Finnish Maritime Administration publication 3/2008. Helsinki SYKE. Oil pollution preparedness on the open sea Final report of the working group (in Finnish). The Finnish Environment 41/2007. The Finnish Environment Institute. Helsinki Mark Reed,, a, Øistein Johansena, Per Johan Brandvika, Per Dalinga, Alun Lewisb, Robert Fioccob, Don Mackayc and Richard Prentkid. Oil Spill Modeling towards the Close of the 20th Century: Overview of the State of the Art. Spill Science & Technology Bulletin Volume 5, Issue 1, April 1999, Pages Maria Gästgifvars, Annakaisa Sarkanen, Mikael Frisk,Hannu Lauri, Kai Myrberg, Pekka Alenius, Oleg Andrejev,Ossi Mustonen, Heli Haapasaari ja Alexander Andrejev. Surface drifter experiments and drift trajectory forecast in the Gulf of Finland. (In Finnish). Report number: The Finnish Environment 720. The publication is available in the internet: Helsinki 27

28 12. Skjong, R., Vanem, E., Endresen, Ø., Risk Evaluation Criteria, SAFEDOR- D DNV-RiskEvaluationCriteria-rev Etkin, D. S. Worldwide Analysis of Marine Oil Spill Cleanup Cost Factors. Arctic and Marine Oil Spill Program Technical Seminar. June Etkin, D. S. Comparative methodologies for estimating on-water response costs for marine oil spills. International Oil Spill Conference IOSC 2005, p May , Miami Beach, FL, USA. Environmental Research Consulting Ramboll Finland Oy et al. Costs of vessel traffic accidents (In Finnish). Finnish Maritime Administration Publication 3/2008. Helsinki Swedish Club Main engine damage an update of the 1998 study. The Swedish Club Sveriges Ångfartygs Assurans Förening. The Swedish Club Highlights. September SSPA Socioeconomic impacts of major oil spills. (in Swedish). SSPA report August Etkin, D.S Modeling oil spill response and damage costs. Proc. 5th Biennial Freshwater Spills Symp Christos A. et al. Assessing environmental risk: Is a single figure realistic as an estimate for the cost of averting one tonne of spilled oil? Working Paper NTUA- MT Division of Ship Design and Maritime Transport. School of Naval Architecture and Marine Engineering. National Technical University of Athens. Athens February Etkin, D. S. Worldwide Analysis of Marine Oil Spill Cleanup Cost Factors. Arctic and Ma-rine Oil Spill Program Technical Seminar. June Kontovas, C.A., et al. An empirical analysis of IOPCF oil spill cost data. Marine Pollution Bulletin (2010), doi: /j.marpolbul IMO. Formal Safety Assessment. Report of the Correspondence Group on Environmental Risk Evaluation Criteria Submitted by Greece, on behalf of the Coordinator of the Correspondence Group. MEPC 60/ December IMO. Formal Safety Assessment. Report on the Working Group on Environmental Risk Evaluation Criteria within the contex of Formal Safety Assessment. MEPC 60/WP March

29 25. IMO. Consolidated text of the Guidelines for Formal Safety Assesment (FSA) for Use in the IMO Rule-Making Process. (MSC/Circ.1023-MEPC/Circ.392) Psarros, G., Skjong, R., Endersen, O., Vanem, E., A perspective on the development of Environmental Risk Acceptance Criteria related to oil spills, Annex to International Maritime Organization document MEPC 59/INF.21, submitted by Norway. 27. Yamada, Y., The cost of oil spills from tankers in relation to weight of spilled oil. Marine Technology 46 (4),