Economic and environmental assessments to support the decision-making process in the offshore wind farm decommissioning projects

The


Introduction
Due to the climate emergency, the global offshore wind energy industry has witnessed a large expansion during the past two decades.Various countries across the world have set their roadmaps to expand their offshore wind energy resources in the coming decades.The UK is the global leader country in terms of operational wind energy capacity with about 10.40 GW reported in 2020 [1,2], equivalent to 30% of global capacity.The UK government has recently announced an ambitious plan to boost its offshore wind energy capacity to 27.5 GW and 40 GW by 2026 and 2030, respectively [3].The European Union countries with a total capacity of 14.6 GW in 2020 [1] are also planning to expand their offshore wind infrastructure further in the coming decades and achieve a total capacity of 460 GW by 2050 [4][5][6].
Offshore wind farm (OWF) assets have also been developed technologically during the past decades which has significantly reduced wind energy production costs by up to 75% [7].Currently, OWFs consist typically of large 7-9 MW wind turbines (WTs) that are installed in relatively shorter times than ever before [7].Scotland is the home of the world's first commercial floating OWF, Hywind [8].Floating OWFs can be commissioned in deeper water depths and longer distances from the shore which enhance the energy production capacity [9].
The operational lifetime of an OWF is expected to be between 20 and 25 years [10,11].However, due to the harsh weather conditions and site-specific characteristic features, there are a lot of uncertainties about their operational lifetime [12].The OWFs can be repowered through a set of amendments in their designs to extend their operational lifetime [13,14].However, due to the high repair or upgrade costs, repowering of OWFs is typically not an ideal option from economic and technical viewpoints [15].This leaves decommissioning as the only practical option for the end of the lifetime of OWFs, in which most of the offshore assets are dismantled/removed and a set of activities need to be performed to return the seabed to its original state [16].The current experience of the wind energy sector in decommissioning is limited, as only five small OWFs have been already dismantled worldwide [17].In addition, as most previously decommissioned OWFs were in shallow waters with smaller assets in size and capacity, any previous experience is not fully applicable to the new OWF decommissioning projects [17].Moreover, OWF decommissioning includes a range of offshore operations performed by expensive vessels/equipment with leasing rates highly sensitive to the market situation and technology availability.The advances in decommissioning technology, vessels, equipment, and recycling are also in the primary stage and significant developments are expected to take place in the coming years [18][19][20].
The current global appetite for expansion of renewable energies highlights the fact that many OWFs will enter the decommissioning stage in the future which might potentially create serious environmental and economic challenges to different countries [18,[21][22][23][24].The previous experience of oil and gas and coal sectors in the US clearly shows the extent of the decommissioning risk to the environment and different stakeholders, in which a massive number of sites and infrastructure were abandoned by bankrupt companies [25,26].There are similar experiences in the offshore wind sector across the world which clearly show that abandoned OWF assets can cause serious environmental challenges [26].The OWF decommissioning challenge is also related to the United Nations' sustainable development goals that should be addressed by policymakers and industry [27].Hence, there is an urgent need for decommissioning regulations to protect the environment and taxpayers against the unwanted consequences of OWF assets at the end of their lifetime [28].
In the UK, the Energy Act 2004 gives the power to the secretary of state for the Department for Business, Energy and Industrial Strategy and Scottish ministers to request an appropriate form of financial security from OWF developers/owners with respect to their decommissioning obligations defined based on an agreed decommissioning programme (DP).According to the guidance recently published by the Scottish government [29], OWF owners/developers must provide the DP when they seek approval for the installation stage, which means that no installation operation will be allowed to take place without an already approved DP.In the prepared DP, the OWF owners/developers should predict the detailed decommissioning costs, techniques, and approaches [29].Hence, the government should be able to check and confirm the predicted decommissioning costs by the OWF owner/developer to protect the taxpayers in the event the owner/developer defaults on their obligations.This shows how accurate cost modelling approaches play a -Only recycling emissions calculated for decommissioning stage -Emissions were considered as per tonnes of material Martinez and Iglesias [44] Cost Top-down Decommissioning costs are estimated as percentage of installation costs Judge et al. [45] Cost Top-down Decommissioning costs parameters are assumed as per energy production capacity Johnston et al. [46] Cost Top-down The costs related to each lifecycle stage is expressed as a given percentage of overall project cost Reimers et al. [47] Emissions Top-down The decommissioning emissions are defined as a given percentage of overall emissions.Kumar et al. [48] Cost Top-down Decommissioning cost is estimated as per energy production capacity Myhr et al. [49] Cost Top-down Decommissioning costs are estimated as percentage of installation costs Machiridza and Bhattacharya [50] Cost Top-down The decommissioning cost is estimated as given percentage of the overall levelized cost of energy Milne et al. [20] Cost Bottom-up -The overall durations for different operations considered.-Detailed duration analysis for different components of each operation, e.g., for cutting duration, was not performed.
(continued on next page) S. Jalili et al. crucial role in protecting the environment, taxpayers, and local communities against the decommissioning risks.
The OWF decommissioning is an emerging field with limited data and ongoing technological developments which make the economic and environmental assessments difficult tasks [20,30].The main aim of the economic and environmental assessments of OWF decommissioning projects is to predict the costs and emissions produced by removal operations.There are two main approaches for the economic and environmental assessments, top-down and bottom-up.The top-down approach assumes a given percentage of overall cost and emissions to estimate the cost and emissions of a given decommissioning operation.However, as the size, layout, and site-specific features are not the same for all OWFs, the results from the top-down approach are not expected to be accurate and reliable [31].On the other hand, the bottom-up approach considers detailed duration and vessel/equipment parameters to predict the costs and emissions, which could potentially provide realistic results.However, the bottom-up approaches highly rely on available data and information.
Table 1 summarised the methodologies and main assumptions of different studies in the literature for the economic and environmental analyses of OWF decommissioning.From Table 1, most of the studies were focused on the development/employment of top-down approaches, while few studies were focused on the development of bottomup approaches.Although some studies in the literature have focused on the bottom-up cost formulations, the emission analysis based on the bottom-up approach has not been addressed for the operational aspects of decommissioning projects, as can be seen in Table 1.One of the works in the field is done by Kaiser and Snyder [32] where the authors developed bottom-up cost formulations for the OWF projects in the US by considering detailed duration and cost parameters [32,33].However, any cost formulations use a set of data and assumptions that are specific to the project location.Moreover, the study [32] does not provide the environmental assessment based on a bottom-up approach for the offshore decommissioning operations.This study aims to develop a new bottom-up approach for the economic and environmental assessments of OWF decommissioning projects in the North Sea region that could benefit both industry and policymakers with holistic figures on emissions produced by the offshore removal operations.The approach is developed based on the available data and experience and tries to provide appropriate assumptions to predict the costs and emissions caused by the different decommissioning operations.The study investigates two OWF case studies, the Lincs and Gunfleet Sands, in the UK continental shelf to validate the performance of the proposed approach.A sensitivity analysis of the overall decommissioning cost to variations in different parameters is also performed to identify the key parameters affecting the costs.The sensitivity analysis provides insight to policymakers regarding the cost reduction strategies in OWF decommissioning.
The study is organised as follows.The proposed economic assessment approach is presented in Section 2. The environmental assessment of decommissioning projects is investigated in Section 3 in which the detailed emission calculations for different operations are explained.In Section 4, the available data that can potentially affect the cost and emissions predictions are discussed.Section 5 investigates the performance of the proposed approach using two OWF case studies and discusses the decommissioning cost sensitivity analysis.Finally, Section 6 provides the concluding remarks.

Economic assessment
The OWF decommissioning stages can be described based on a work breakdown structure.According to Milne et al. [20], the work breakdown structure for the OWF decommissioning includes the project management, project preparation, offshore preparation, WT removal, offshore substation (OS) removal, meteorological mast (MM) removal, cable removal, seabed clearance and restoration, recycling and waste management, and monitoring.The focus of this study is on the removal operations of OWF decommissioning projects, including WT, OS, MM, and cable removals as well as seabed clearance and restoration.In this study, it is assumed that the removal of WT topsides and their foundations will be performed through two different campaigns.In the following subsections, the cost formulations for each removal operation will be presented based on a bottom-up approach.

WT topside removal
The WT topside includes the blades, nacelle, and tower section.The different components of WT are usually lifted by a jack-up vessel (JUV) and placed on a barge vessel (BV) pulled by tug boats (TBs) for transportation to the shore.With these assumptions, the removal cost of the WT topsides can be expressed in terms of the mobilisation and day rates of mentioned vessels as follows: where, C WT represents the removal cost of the WT topsides, C JUV m and C BV m are the mobilisation rates of JUV and BV, respectively, α is the number of BVs, C JUV D , C BV D , and C TB D are the day rates of the JUV, BV, and TB, respectively, β is the umber of TBs, and t JUV WT is the total removal duration of WT topsides using JUV in hours.In this study, all the cost units are in pounds.The removal duration of WT topsides in the OWF, represented by t WT , depends on the removal method, number of WTs, lifting durations, and vessel parameters.There are several WT removal methods defined based on reverse order of installation with different numbers of lifts and durations [33].Due to the nature of the investigated case studies, it is assumed that the blades will be removed in three separate crane operations.Then, the nacelle with attached rotor and tower section will be lifted and placed on the BV in two separate lift operations.With this assumption, the total duration of WT topside removal can be calculated by the following formula: In the above equation, γ> 1 represents the parameter to consider the weather delays, n t represents the number of WTs in the OWF, t JUV pos , t JUV up , and t JUV down are the positioning, jacking-up, and jacking-down durations of JUV, respectively, t B is the dismantling duration of each blade, t N represents the removal duration of the nacelle, and t T indicates the lifting duration of the tower section.It should be noted that all duration parameters in Equation ( 2) are in hours.

WT foundation removal
Foundation removal is one of the expensive operations in OWF decommissioning projects.It involves underwater pumping and cutting operations.It is also necessary to employ a remotely operated vehicle (ROV) to support the subsea operations.The foundation removal operation consists of preparation and lifting stages.In the preparation stage, the mud inside the foundation needs to be pumped out and the section of the monopile foundation is severed by using the abrasive water jet cutting technique.Thereafter, the foundation is lifted by JUV and placed on a BV.Depending on the project strategy, the types of employed vessels for the foundation removal can vary.It is quite common to employ the JUV to perform both the preparation and cutting process (for example, see Lincs DP [53]).However, due to the high day rate of JUVs, it would be better to minimise the waiting time of JUV during the preparation stage.As Kaiser and Snyder [33] argue, the foundation preparation stage can be done by an offshore support vessel (OSV) which is much cheaper for lease compared to JUVs.In this study, it is assumed that the foundation preparation stage is performed by an OSV.Then, the JUV arrives at the site to lift and place the foundations on the BV.Hence, the applied vessels for the foundation removal process would be OSV, JUV, BV, and TB.
Considering the aforementioned points, the foundation removal cost can be formulated in terms of the vessel\equipment costs as: In the above equation, C F is the total cost of foundation removal, C ROV represents the work duration of the JUV for foundation removal, and the definitions for the rest of the parameters are similar to those explained in Section 2.1.The work duration of the OSV is calculated based on the time required for the pumping and cutting processes as follows: where, n F represents the number of foundations in the OWF, t OSV pos is the positioning duration of the OSV, t p is the time required to pump the mud inside the foundation, t c is the time required for cutting the foundation section below the seabed, and t OSV move is the time required by the OSV to move to the next foundation location.The cutting duration t c can be obtained based on the cutting rate per the foundation diameter, represented by ν cut in hr/m, as follows: t c = ν cut D. The pumping duration t P depends on the mud volume inside the foundation and can be calculated by the following equation: £28.4 Note: The costs are converted from US dollars to British pounds @ 1$ = 0.71£.

Table 3
The available and assumed lower and upper bound values for the duration parameters related to the different OWF decommissioning operations.
where, V p is the volume of the mud inside the foundation in m 3 and Q p is the pumping rate in m 3 /hr.The foundations are usually cut from a given depth below the seabed.The total mud volume that should be pumped out of foundation can be calculated as follows: where, D F is the foundation diameter, d c is distance of the cutting line from the seabed, and parameter e represents the additional space needed to provide access of cutting tool to the cutline.In this study, it is assumed that the foundation will be cut from 1 m under the seabed (i.e., d c = 1m), based on Ref. [54].Moreover, the parameter e is taken as 1 m in this study.
As was mentioned earlier, the JUV will be employed to lift the foundation and place it on a BV deck space.The work duration of the JUV can be obtained by the following equation: where, t JUV L,F is lifting duration of the foundation by the JUV and the definition for the rest of the parameters are similar to those in the previous section.

OS and MM removal
The removal processes for the OS and MM consist of topside and foundation removal stages.The lifting operations in both stages are typically performed by the JUV.The dismantled components are transported to the shore by the BV supported by the appropriate number of TBs.The removal cost of the OS can be written in terms of the vessel/ equipment costs as: where, C OS represents the removal cost of OS, t JUV OS is the total removal duration of OS, and the definitions for the rest of the parameters are given in previous sections.Depending on the foundation type, the removal duration can be obtained by the following equations: where, n OS represents the number of OSs in the OWF, t c,top is the time required to cut and disconnect the topside of the OS, t L,top indicates the lifting duration of the OS topside, t c,p is the time required for cutting the jacket piles under the seabed, and t L,J is the time required to lift the jacket and place it on a BV.
• If the foundation of OS is a monopile structure where, t p is the mud pumping duration obtained from Equation ( 5) and t c is the foundation cutting duration which is assumed same as explained in Section 2.2.The cost calculation for the MM removal operation is similar to the formulations provided above for the OS removal, but with significantly shorter duration parameters.As the topside and foundation of MM are significantly smaller in size and lighter in weights, the duration parameters t c,top , t L,top , t p and t c are expected to be shorter than those for the OS removal operation.

Cable removal
Current decommissioning regulations allow the cables to be left in their situation if they are buried at an appropriate depth under the seabed.Thus, the assumption of leaving cables in their situation is common in the recent OWF decommissioning programmes.In this case, a full inspection and burial are required, especially for the cable ends disconnected from the WTs.It is worth mentioning that the regulations on subsea cables may change and they might not be allowed to be left in place in future.Therefore, this study assumes that the cables will be removed entirely from the seabed, and the removal costs and emissions will be calculated.

Table 4
The available data for the mobilisation and day rates of the different vessel/ equipment employed in the OWF decommissioning projects.

Mobilisation/ Demobilisation
Day rates a Based on the 2017 market.b Exchanges rate is applied: 1£ = 1.16€.c Assumed due to the lack of the data.

Table 5
Emission factors for different pollutants in kg/metric ton employed within the proposed environmental assessment approach to calculate the emissions caused by the vessels during the OWF decommissioning operations [65].The cable removal operation requires a cable laying vessel (CLV) with subsea inspections performed by an ROV.The cost of cable removal operation can be obtained as follows: The cable removal is expected to take place in a relatively shorter time than the installation.Kaiser and Snyder [33] suggest converting the installation durations into the equivalent removal durations by using an inflation factor as the following equations: In the above equations, L I and L E represent the lengths of inter-array and export cables, respectively, r I indicates the inter-array cable installation rate in km/day, r E is the installation rate for the export cables in km/day, IF I and IF E are inflation rates for the inter-array and export cables, respectively.

Seabed clearance and restoration
Following the completion of removal operations, a set of activities needs to take place to return the OWF site to its original state before the installation of assets.The holes resulting from the foundation removal need to be refilled and the scour protection around the foundations can be removed.As marine life typically forms on the scour protection over the lifetime of OWF, most of the OWF decommissioning projects have not been decided to remove the scour protection material on the seabed.This can be an ideal option from environmental and cost perspectives.However, this study assumes that the scour protection will be removed for assessment purposes.
The total cost of the seabed clearance and restoration activities can be simply written as: where, C SC is the total cost of the seabed clearance and restoration op-   erations, C SP represents the cost of the scour protection removal, and C RD is the cost of rock dumping activities performed to refill the foundation location in the OWF site.
For the scour protection removal operation, a derrick crane barge vessel (DCBV) is employed.The removed scour materials are transported to the shore by a BV pulled by TBs.An ROV is also required for inspection and support of subsea activities.With these assumptions, the cost of scour protection removal operation can be formulated as: in which: In Equations ( 15) and ( 16), C SP is the overall cost of scour protection removal operation, C DCBV are the volume of scour protection material around the i th WT and i th OS, respectively, V MM is the scour protection material volume around the foundation of MM, r ret indicates the removal rate of scour protection material, and the definitions for the other parameters are similar to those mentioned in the previous subsections.
The rock dumping cost can be calculated as follows: in which: where, C RD represents the cost of the rock dumping activity, C RDV m is the mobilisation cost of the RDV, C RDV D indicates the day rate of the RDV, t RDV RD is the total rock dumping operation using a RDV, and r RD is the rock dumping rate in locations per day.

Social costs
The social cost is an attempt to put a price on emissions.The social cost assessment can be beneficial for policymakers to understand whether the costs and benefits of a proposed policy in expanding the OWFs to curb climate change are justified.The social costs related to the emission of various pollutants can be calculated by multiplying the emission values by the social cost factors listed in Table 2 [55].In this study, the social costs will be calculated for the investigated case studies through the multiplying the social cost factors in Table 2 by the emission amounts calculated from Section 3.

Environmental assessment
The emissions produced by decommissioning activities mainly depend on the fuel consumption and emission rates of the vessels/ equipment involved in different operations.For each decommissioning activity, the overall emissions can be decomposed into two parts, including the emissions resulting from the crane operations and the emissions produced by the transportation activities of dismantled components.In this section, the formulations for the emission calculation in different OWF decommissioning activities are presented.
The total emission amount produced by decommissioning activities can be simply written as: where, E total represents the total emission amount, E WT , E F , E OS , E MM , E C , and E SC are the emissions produced by the WT removal, foundation removal, OS removal, MM removal, cable removal, and seabed clearance and restoration operations, respectively.The detailed formulations for each component of Equation ( 19) will be presented in the subsequent subsections.

WT removal emissions
The emissions produced by the WT removal operation E WT can be expressed in terms of the emissions generated by the crane and transport operations as follows: where, E O WT indicates the emissions produced by the lifting and positioning operations in WT removal operation and E tr WT represents the emissions caused by the transportation of dismantled WT components to

Table 9
The decommissioning strategies assumed in this study for the Gunfleet Sands case study to perform the economic and environmental assessments.

Asset
Installation techniques and equipment [68,69] Decommissioning assumptions adopted in this study

WTs
• A JUV employed for installation • Installation method: Tower + Nacelle + Blade + Blade + Blade • Reverse order of installation is considered for WT removal • A JUV was assumed for lifting operations and two BVs were assumed for transportation.TBs are also required.

Monopiles and transition pieces
The installation of the monopiles and transition pieces was performed by the crane barge vessel and JUV in deeper and shallower waters, respectively.
• Internal cutting for monopile removal is assumed • Abrasive water jet cutting tool will be used for cutting the monopile • The mud inside the monopile needs to be pumped up to 1 m below the cutting line • It is assumed that the foundation will be cut from 1 or 2 m below the seabed the shore.The emissions resulted from the crane operations E O WT are mainly related to the JUV, which can be expressed by the following equation: where, t JUV WT is the activity duration of JUV during the WT topside removal calculated from Equation ( 2), e r is the emission factor for a given pollutant in kg/metric tonne, and f JUV represents the fuel consumption rate of the JUV in tonne/hr.
The emissions of transportation activities depend on the project strategy.In this study, it is assumed that the dismantled components will be transported to the shore by using BVs pulled by TBs.Thus, the specifications of TBs should be considered in the transport emission calculations.The following equation expresses the transport emissions for the WT removal operation:

Table 10
The vessel/equipment leasing rates assumed for the minimum and maximum cost scenarios in the Gunfleet Sands case study.
where, β is the number of utilised TBs, f TB is the fuel consumption rate of TB in tonne/hr, t JUV WT is already known from Equation ( 2), and t tr WT represents the transport duration of WT components to the shore.The transport duration t tr WT depends on the deck capacity of the BV and removal strategy.Let us assume that the BV can carry the n CWT number of WT topside units in each transport cycle.With this assumption, the transport duration t tr WT can be calculated by the following equation: where, γ is the weather delay parameter, fix(.) is a function that rounds the input value to the nearest integer value, n t is the number of WTs, d port represents the distance between the port and OWF site, υ TB is the towing speed of the BVs in knots, t ol WT represents the off-loading duration of each WT unit at the port, and t s indicates the service time of the BV.

Foundation removal emissions
Similar to the previous subsection, the emissions produced by the foundation removal activities can be simply expressed in terms of the emissions resulting from the crane/cutting and transport activities as follows: where, E O F represents the emissions produced by the crane and cutting activities in foundation removal operation and E tr F indicates the emissions generated by the transport operation of foundation units to the shore.As was explained in subsection 2.2, the JUV and OSV are involved in foundation removal operations.With this assumption, the emissions produced by crane and cutting operations can be written as follows: where, f OSV and f JUV are the fuel consumption rates of the JUV and OSV in tonne/hr, respectively, t OSV F is the activity duration of OSV known from Equation (4), and t JUV F is the activity duration of JUV in the lifting operation of foundations obtained from Equation (7).
The emissions produced by the transport operation of dismantled foundations can be written as: In the above equation, t JUV F is known from Equation ( 7) and t tr F represents the transport duration of foundation units to the port.Let n CF be the number of foundation units transported by the BV in each transport cycle.Then, the transport duration can be calculated by the following formula:

Table 12
The emissions for the different decommissioning activities calculated from the environmental assessment approach for the minimum cost scenario in the Gunfleet Sands case study (tons).

Table 13
The emissions for the different decommissioning activities calculated from the environmental assessment approach for the maximum cost scenario in the Gunfleet Sands case study (tons).

OS and MM removal emissions
As explained in Section 2.3, the OS and MM removal operations are similar with different duration parameters.In this subsection, the emission formulation for the OS removal operation will be discussed and similar equations can be used for the MM removal operation.The emissions for the OS removal operation, represented by E OS , can be split up into two parts as follows: where, E o OS is the emissions produced by the crane operations and E tr OS is the emissions caused by the transportation of OS components.The emissions produced by the crane operations can be obtained by the following equation: in which t JUV OS is the activity duration of the JUV in OS removal operation obtained from Equation ( 9) or (10), depending on the OS foundation type.
The transport emissions can be calculated by considering the TB fuel consumption as follows: where, t tr OS is the duration required to transport the dismantled parts of the OS which is calculated by the following equation: where, t ol,F OS and t ol,T OS are the offloading duration of the foundation and topside of the OS.In Equation (30), it is assumed that the TB will be in active mode during both crane and transport operations.

Cable removal emissions
For the cable removal operation, the emissions can be expressed in terms of the fuel consumption of CLV as: where, f CLV is the fuel consumption of the CLV in tonne/hour, t CLV I represents the time required for the removal of inter-array cables known from Equation ( 12), and t CLV

E
indicates the removal duration of the export cables obtained from Equation (13).

Emissions for the seabed clearance and restoration
As was discussed earlier in Section 2.5, the seabed clearance and restoration include the scour protection removal and rock dumping operations.Hence, the emission for these activities can be expressed as: where, E SC is the emissions produced by the seabed clearance and  restoration activities, E SP represents the emissions caused by the scour protection removal, and E RD is the emissions resulting from the rock dumping operation.The emissions produced by the scour protection removal E SP can be written in terms of the emissions caused by operational and transport operations as: where, E O SP is the emissions produced by the scour protection removal operation and E tr SP indicates the emissions caused by the transportation of removed materials.The emissions resulted from the scour protection removal operation E O SP can be calculated as follows: where, f DCBV represents the fuel consumption of the DCBV in tonnes/hr and t DCBV SP indicates the total removal duration of the scour protections calculated from Equation (16).As the BV is used for the transport of removed materials, the emission E tr SP can be written as: where, f TB represents the fuel consumption rate of the TB in tonnes/hr.In the above equation, it is assumed that the TB will be in an active mode during the whole operation.
The emissions caused by the rock dumping activity E RD can be ob-

Table 16
Overall information on different assets in the Lincs OWF [53].a Approximate value assumed in this study.

Table 17
Technical specifications of monopile and jacket structures in the Lincs OWF [53].

Monopiles for WTs
Outer shaft diameter

Table 18
Comparison of the assumptions for the vessels/equipment considered by the Lincs DP [53] and this study.tained by the following equation: where, f RDV represents the fuel consumption of the RDV in tonnes/hr and t RDV RD is the total rock dumping duration obtained from Equation ( 18).

Parameters
The lack of available information is one of the key barriers to the development of accurate cost and emission models.The accuracy of economic and environmental assessments approach presented in this study depends on a variety of parameters, such as duration, vessel/ equipment leasing, fuel consumption, and emission parameters, which

Table 19
The duration and vessels/equipment lease rate parameter values employed within the economic and environmental assessments approach for the Lincs case study.

Table 20
The removal costs and durations for different decommissioning activities in the Lincs case study obtained from the proposed economic assessment approach and Linc Limited DP [53].a Lincs DP [53] predicted £7.2 m for foundation removal plus £298 k for the cutting activities.
can significantly affect the cost and emission estimations.The duration and leasing parameters depend primarily on the geographical location of the OWF, utilised technology, availability of vessels/equipment, weather conditions, project planning, market conditions, etc.In this study, the experience and information gathered from available studies and technical reports are employed to provide the best possible estimations.Table 3 presents the available ranges for the different duration parameters related to each decommissioning activity.This table reflects the fact that the assumptions for duration parameters are subjected to significant uncertainties due to weather conditions.In addition, Table 4 lists possible ranges for the leasing parameters of vessels/equipment based on different sources.It is observable that the available experience offers wide intervals for the leasing costs.The leasing costs depend on the contract duration, supply and demand, the market situation in oil and gas industry, etc.As can be seen from Tables 3 and 4, appropriate values are assumed in this study for the parameters with no historically available values.
The emission formulations depend on the emission factor and fuel consumption rates.The emission factor varies depending on the type of pollutant.Table 5 lists the emissions factors for different pollutants.The fuel consumption rates depend on the vessel type as well as the activity mode.In this study, an average value of fuel consumption is assumed for each vessel as listed in Table 6.

Case studies
In this section, the cost and environmental assessments of two OWF decommissioning case studies in the UK are investigated.Both OWFs consist of WTs with individual capacities of 3.6 MW.The Gunfleet Sands OWF is the first case study, which is used to show how the uncertainties in the duration and leasing parameters can cause dramatic changes in the cost and emission estimations.In the Lincs OWF, the cost and emissions are more realistic, and the results are verified by the cost estimation available from the source reports.The overall intention of this section is to provide the cost and emission estimations for the mentioned OWFs based on their real site-specific information.In the investigated case studies, the constant social cost, emission, and fuel consumption rates listed in Table 2, Table 5, and Table 6 are used.

Gunfleet sands OWF
The Gunfleet Sands OWF is located 8.5 km off the southeast coast of Clacton-on-Sea, Essex, UK.The installation process of this OWF took

Table 21
The emissions for different decommissioning activities obtained from the proposed environmental assessment approach for the Lincs case study (tons).Fig. 6.The CO 2 emission percentage break-down distribution obtained from the proposed environmental assessment approach for different decommissioning activities in the Lincs case study.

Table 22
The social costs of different decommissioning activities caused by the different pollutants in the Lincs case study.Fig. 7.The total removal cost percentage break-down distribution obtained from the proposed economic assessment approach for different decommissioning activities and pollutant in the Lincs case study.

Table 23
The categorisation of different parameters employed within the proposed economic assessment approach for decommissioning cost sensitivity analysis.demonstration purposes.The initial design lifetime of this OWF was considered to be 20 years [67].
The general information of the Gunfleet Sands OWF assets is presented in Table 7.The foundation type of the WTs is a steel monopile structure with the specifications listed in Table 8.Although the initial environmental assessment report of this OWF published in 2007 [68] has set few decommissioning objectives, appropriate assumptions need to be made for different decommissioning activities, as presented in Table 9.The assumptions in Table 9 were adopted by considering the available limited information from the installation phase in Refs.[68,69].
In this case study, the decommissioning costs and emissions are calculated for the minimum and maximum cost scenarios to show how the uncertainties in available data can affect the results.In the minimum cost scenario, the shortest durations and cheapest vessel/equipment leasing rates from Tables 3 and 4 are assumed, while the longest duration and most expensive vessel/equipment leasing rates are selected from Tables 3 and 4 for the maximum costs scenario.In both cases, a 20% delay in operational times is considered due to weather conditions (i.e., γ = 1.20).Table 10 lists the minimum and maximum leasing rates assumed for different vessels/equipment in this case study.The values in Table 10 are selected based on the previous experience presented in Table 4.The durations and costs calculated for each decommissioning activity are presented in Table 11.The overall observation from Table 11 suggests that the costs and operational durations are significantly sensitive to the variations in the available data.The average WT removal duration from 1.225 days/turbine in the minimum scenario increases to 2.15 days/turbine in the maximum scenario, showing about 75% changes in terms of the duration.However, the change in the cost of WT removal operation is more dramatic, increasing from £9.1 m to £31.6 m, which shows more than a 300% increase in the cost value.A similar conclusion can be made for the other activities.It is worth mentioning that the change in the cable removal cost value is surprisingly large which highlights the level of uncertainty of available data for this activity.Fig. 2 illustratively compares the minimum and maximum costs for each activity.
As shown in the emission formulations presented in Section 3, the emission amounts can be affected by the uncertainties in duration parameters.To investigate the extent of emissions' sensitivity to the uncertainties in duration parameters, the detailed emissions of different pollutants produced by decommissioning activities for the two scenarios are presented in Tables 12 and 13, respectively.The results provide the transport, operational, and overall emissions.From Tables 12 and 13, it can be observed that the overall CO 2 emission increases from 17,912 tonnes in the minimum scenario to 37,919 tonnes in the maximum scenario, about a 111% change in emission amounts.Although the differences in emission amounts obtained from the two scenarios are remarkable, the effects of uncertainties in initial data on the emissions are not as great as their impact on the cost values.Tables 14 and 15 present the social costs caused by the different pollutants in minimum and maximum cost scenarios, respectively.These tables show that the social costs are about £2.76 m and £5.78 m for the minimum and maximum scenarios, respectively.The changes in the social costs due to uncertainties in duration parameters are also more than 100%.It can also be seen that NOx is a major contributor to the social cost values.The removal and social costs are combined and the percentage break-down distributions for different cost items are presented in Figs. 3 and 4. From these figures, the social costs account for about 10% and 5% of total removal costs in the minimum and maximum scenarios, respectively.Once again, these figures show that the cable removal costs are significantly different in the two scenarios, showing the impact of high uncertainties in the cable removal rate parameters.

Lincs OWF
The location of the second investigated case study, the Lincs OWF, is shown in Fig. 5.This OWF is located 8 km off the coast at Skegness, Lincolnshire, UK.The Lincs includes 75 WTs with 3.6 MW capacities.The overall information on the assets in the Lincs OWF is provided in Table 16.In this OWF, the WTs and OS are supported by steel monopile and jacket structures, respectively.The technical specifications of the foundation structures are listed in Table 17.The DP [53] of the Lincs OWF was predicted 20 years as the operational lifetime.The main intention of this case study is to verify the cost estimation formulations by comparing the results to those predicted in the Linc Limited DP [53].
Although the Lincs DP [53] assumes that the subsea cables and scour protection will be left in their situ, this study assesses the costs and emissions for the complete removal of mentioned assets.The DP [53] provided a set of assumptions on the employed vessels and equipment.It recommends using a single JUV supported by a BV for the WT and foundation removal activities.With this assumption, the JUV will be required to keep waiting during the transportation of dismantled units to the shore, which increases the leasing duration of the JUV.In this study, it is assumed that two BVs will be employed for transportation, one on-site and one in transit.The DP [53] also assumes that the 9 WT and 10 foundation units will be transported by BV in each transport cycle.No information on the ROV activities was mentioned in the DP [53].In this study, the ROV costs are also considered in the cost estimations.The assumptions are compared to those described in the Lincs DP [53] in Table 18.The assumed duration and cost parameters for the Lincs case Fig. 11.Sensitivity of total removal cost to the vessel/equipment leasing rates obtained from the proposed economic assessment approach.
S. Jalili et al. study are listed in Table 19, which are selected partly based on the available information from the Lincs DP [53] and partly based on the previous experience and available data.
The durations and costs obtained in the current study are compared to those reported by the Lincs Limited DP [53] in Table 20, in which 20% of weather delays are considered.The DP [53] predicted the WT and foundation removal costs.The WTs removal cost based on the formulations presented in this study is expected to be about £13.9 m, while the DP [53] predicted this to be about £12.2 m, a 14% difference.The calculated average WT removal duration is about 1.23 days/unit which is shorter than 2.16 days/unit reported by the DP [53].The cost of foundation removal obtained by this study and the DP [53] are about £8.8 m and £7.50 m, respectively, resulting in a 17% difference.The differences in the cost and duration values may be caused by the different removal assumptions and strategies.For example, the DP [53] assumed a single BV for the transportation which can cause unwanted delays in JUV crane operations during both WT and foundation removal operations.Considering two BVs in this study, one on-site and one in transit, prevents such delays to occur in the project schedule.The differences in the cost values may also be caused by the vessel/equipment rates.In the DP [53], the vessel/equipment rates were not reported.In addition, this study considered the ROV and TBs to support different operations, while it does not seem to be considered in the DP [53].It should also be noted that the costs reported by the DP [53] are calculated based on the vessel/equipment rates in 2009, which could be another reason for the predicted cost differences.However, the differences in the cost values are relatively low.The costs for the full cable and scour protection removals are estimated to be about £7.9 m and £6.21 m, respectively.These results reveal that the removals of cables and scour protection materials are significantly expensive activities.Should the cables and scour protection materials leave in their situ, about £14.11 m or equivalently £180 k per WT would be saved in the project costs.The overall cost of removal operations in the Lincs case study is estimated to be about £38.3 m.
The emissions produced by the different decommissioning activities in the Lincs case study are listed in Table 21.Major part of the emissions is produced by the WT and foundation removal operations with about 11,000 and 9700 tonnes of CO 2 emissions, respectively.The emissions caused by the transport activities account for about 46% of total produced emissions in the project, which highlights the fact that the transport strategies play an important role in the environmental impact of OWF decommissioning projects.The decommissioning activities in the Lincs case study are expected to produce about 581, 88, 16, and 30,000 tonnes of NO x , SO X , PM, and CO 2 emissions, respectively.Fig. 6 shows the CO 2 percentage breakdown distribution, which shows that the WT and foundation removals produce about 36% and 32% of total CO 2 emission in this case study.Moreover, Table 22 lists the social costs caused by the different pollutants for the Lincs case study, which shows an overall social cost of £4.6 m.Fig. 7 presents the total cost breakdown distributions for this case study.From Fig. 7, the social costs account for about 11% of overall costs, which shows the necessity of considering the social cost in the economic assessment of OWF decommissioning project.

Cost sensitivity analyses
As was discussed in the previous sections, the costs of different decommissioning activities depend on a set of duration and leasing parameters.The main aim of this section is to see how these parameters can affect the overall cost values.To this end, a cost sensitivity analysis of different duration and leasing parameters is performed for the Lincs case study.The different parameter categorises that could affect the cost values of OWF decommissioning activities are listed in Table 23.The overall assumption of the sensitivity analyses in this section is that the changes in the values of different parameters are in the interval of [− 90%, 200%].
The results of sensitivity analysis for the vessel duration parameters are illustrated in Fig. 8.As can be seen, positioning parameters of JUV and DCBV as well as anchor retrieval of DCBV have the most significant impacts on the overall costs.The reason behind this observation is related to the high leasing rate of these vessels.In contrast, it can be seen from Fig. 8 that the changes in the movement parameter of OSV have no significant impact on the cost values.
The variations in the overall cost values due to changes in the removal durations and rates are illustrated in Fig. 9.Among the different removal parameters, the removal duration of the blade t B and tower t T , cable removal rates (i.e., r I and r E ) and scour protection removal rate r ret have significant impact on the overall cost values.It should be mentioned that the total removal cost is a decreasing nonlinear function of parameters r I , r E , r rd , and r ret involved in the cable removal, site clearance and restoration activities (see sections 2.4 and 2.5), while it is an increasing function for other removal parameters.The parameters involved in the cutting operations are also important parameters which can affect the overall costs as illustrated in Fig. 10.Fig. 10 reveals that a 90% increase in cutting speed can reduce the overall cost by about 4%.
The vessel leasing rates are also important parameters that should be properly estimated to predict realistic decommissioning costs.To see how the vessel/equipment costs can make changes in overall cost estimations, Fig. 11 demonstrates the sensitivity of the overall cost to the leasing rates.Fig. 11 reveals that the day rate of the JUV has the most remarkable impact on the overall costs.It suggests that the 100% and 200% increases in JUV day rates can result in about 37% and 75% changes in the overall cost values, respectively.The day rate of the BV is also an important parameter.The 100% and 200% changes in BV day rates can cause about 12% and 24% increases in the cost values, respectively, which are still remarkable changes.Similar conclusions can be made for the leasing rates of other equipment/vessels, but with relatively smaller impacts.

Concluding remarks
The main contribution of this study is to develop a new approach for the economic and environmental assessments of OWF decommissioning projects in the North Sea region based on a bottom-up approach.The detailed formulations are provided for the cost and emission calculations of different decommissioning operations.The proposed formulations include a set of duration and vessels/equipment leasing parameters which may affect the cost and emission estimations.The study gathered available experience and information from different sources to achieve the best possible cost and emission estimations.
To show the effectiveness of the approach, the economic and environmental assessments of two real-world OWF case studies in the North Sea region were investigated, Gunfleet Sands and Lincs OWFs.In the Lincs case study, the costs and emissions were estimated based on the best possible assumptions for the duration and cost parameters as well as decommissioning strategy.The preciseness of cost estimates for the Lincs case study were investigated through a comparison between the costs obtained by the proposed approach and those reported in the Lincs DP.The results suggested that the proposed approach can estimate the decommissioning costs with an error between 14% and 17% in the cost values.To show how the overall decommissioning cost values can be affected by changes in the different parameter values, a cost sensitivity analysis was performed for the different categories of parameters.
The overall conclusions made from this study can be listed as follows: • Climate change emergency accelerated the vast expansions of OWFs globally.However, the OWF decommissioning is a big challenge ahead with significant economic and environmental impacts.This study tried to assess these impacts which are in line with the UN sustainable development goals in affordable and clean energy as well as industry, innovation, and infrastructure.
• As it was shown in the Gunfleet Sands case study, the performance of the economic assessment model can be significantly affected by quality of the assumptions made for the employed technology, duration parameters, vessel/equipment mobilisation and day rates, transportation strategies, and fuel parameters.Realistic assumptions for these parameters are expected to reduce the error levels in the proposed approach.• The vessel/equipment leasing rates are subjected to their availability, contract duration, and market situation, while the duration parameters depend more on the technology developments and weather conditions.• This study shows that the social costs caused by the decommissioning projects are not negligible and they should be considered by the policymakers to understand whether the costs and benefits of a proposed policy to curb climate change are justified.The results suggested that the social costs of the projects can vary between 5% and 11%.• The study highlighted the importance of transport strategies in the environmental analysis of OWF decommissioning projects, accounting about 46% of total emissions.• The study reflected the fact that the full removal operations of subsea cables and scour protection are relatively expensive activities with large amounts of emissions, accounting 33% and 31% of overall cost and CO 2 emissions.For the Lincs case study, the full cable removal with seabed clearance and restoration will cost about £6.4 m and generate about 9380 tonnes of CO 2 emissions.The policymakers and regulators need to consider these impacts.• The results obtained from the cost sensitivity analysis benefit policymakers with an insight into cost reduction strategies in OWF decommissioning projects.The results revealed that the leasing rates and duration parameters of the JUV have a significant impact on the overall cost values.Shorter tower and blade removal durations could also significantly reduce the overall removal costs.Foundation cutting speed is also another important parameter which highlights the necessity of future developments in cutting tools.• The proposed approach is a bottom-up model developed based on the publicly available data which can be modified/enhanced for the new dataset.The approach provides the flexibility for the companies in industry to use their own internal data for the economic and environmental analyses.• The approach was developed based on the current practice, technology, and available data.As the technology significantly evolves over the time, the accuracy of the approach would need to be enhanced/improved by new data and technical assumptions.

m
and C ROV D represent the mobilisation cost and day rate of ROV, respectively, C OSV D indicates the day rate of the OSV, t OSV F the work duration of the OSV for foundation removal, t JUVF ) where, C C is the cable removal cost, C CLV m is the mobilisation cost of the CLV, C CLVi D and C CLVe D are the day rates of the CLV for the inter-array and export cables, respectively, t CLV I represents the removal duration of interarray cables by a CLV, and t CLV E is the removal duration of export cables using a CLV.

m
and C DCBV D are the mobilisation cost and day rate of the DCBV, respectively, t DCBV SP is the total removal duration of scour protection using a DCBV, t DCBV pos represents the positioning duration of the DCBV, t DCBV a represents the time required by the DCBV to retrieve its anchors, V WT i and V OS i

Fig. 2 .
Fig. 2. The removal cost comparisons between the minimum and maximum cost scenarios in the Gunfleet Sands case study for the different decommissioning activities.

Fig. 3 .
Fig. 3.The cost percentage break-down distribution for each decommissioning activity and pollutant in the Gunfleet Sand case study (minimum scenario).

Fig. 4 .
Fig. 4. The cost percentage break-down distribution for each decommissioning activity and pollutant in the Gunfleet Sand case study (maximum scenario).
tonnes for connecting the transition piece Jacket for OS Size 20 m × 26 m × 30 m Piles 4 leg piles with a diameter of 54" Seabed penetration 26 m Jacket weight 750-1000 tonnes Piles weight 580 tonnes

DDFig. 8 .
Fig. 8. Sensitivity of total removal cost to the vessel duration parameters obtained from the proposed economic assessment approach.

Fig. 9 .Fig. 10 .
Fig.9.Sensitivity of total removal cost to the removal durations and rates obtained from the proposed economic assessment approach.

Table 6 Fuel
[66]umption rate parameter values used within the proposed environmental assessment approach to calculate the fuel consumptions of different vessels employed during the OWF decommissioning operations[66].aAssumed in this report based on average fuel consumption of 10 tonnes/day.S.Jalili et al.
Specifications DescriptionGeneral Distance to shore 8.5 km from the south-east of Clacton-on-Sea,

Table 11
The costs and durations calculated from the economic assessment for different decommissioning activities in the Gunfleet Sands case study.

Table 14
The social costs caused by the different pollutants for the minimum cost scenario in the Gunfleet Sands case study.

Table 15
The social costs caused by the different pollutants for the maximum cost scenario in the Gunfleet Sands case study.