- Email: [email protected]
Rebuttal letter to the article entitled: “Spatial planning to estimate the offshore wind energy potential in coastal regions and islands. Practical case: The Canary Islands”
Energy 153 (2018) 12e16
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Rebuttal letter to the article entitled: “Spatial planning to estimate the offshore wind energy potential in coastal regions and islands. Practical case: The Canary Islands” lez-Díaz c Ricardo Guerrero-Lemus a, *, Ignacio de la Nuez b, Benjamín Gonza nchez S/N, 38206, S/C de Tenerife, Spain Departamento de Física, Universidad de La Laguna, Avenida Astrofísico Francisco Sa tica, Universidad de Las Palmas de Gran Canaria, Campus Universitario de Tafira, 35010, Las Palmas de nica y Automa Departamento de Ingeniería Electro Gran Canarias, Spain c nchez S/N, 38206, S/C de Tenerife, Spain Departamento de Ingeniería Industrial, Universidad de La Laguna, Avenida Astrofísico Francisco Sa a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Available online 6 April 2018
The objective of this rebuttal letter is to provide a critical analysis of the article entitled “Spatial planning to estimate the offshore wind energy potential in coastal regions and islands. Practical case: The Canary Islands” [Energy 143 (2018) 91e103], mainly in relation to its methodology, suitable marine areas and electricity production costs. The absence of basic considerations about the characteristics of the insular power grids, the composition of the electricity costs in the Canary Islands, and the lack of rigor in some assumptions related to visibility constraints, offshore costs, integration costs, the mixing of data from different time periods and the references used, provides unrealistic and useless results for a necessary debate about the potential of offshore wind energy in the Canary Islands. In this rebuttal letter we will also demonstrate that the potential offshore wind capacity calculated by the authors is much lower. Moreover, the assertion that the electricity cost from offshore wind calculated by the authors is lower than the current electricity cost is wrong and, in fact, the cost of electricity from offshore wind is higher in the time period when the analysis was made, and also at present. © 2018 Elsevier Ltd. All rights reserved.
Keywords: Offshore wind Canary islands Cost analysis Wind energy
1. Introduction Electricity from offshore wind energy is growing substantially around the world (14,086.3 MW in 2016 [1]), mainly in developed countries with marine territorial areas where a shallow continental platform permits the placement of such devices not interfering with other marine activities and avoiding substantial environmental concerns. In Canary Islands, the debate about the location of offshore wind farms in future is growing because of the limited surface available onshore. There exists a growing desire for achieving a 100% electricity production from renewable resources, but also a growing concern about the location of wind farms onshore and the environmental impact associated for islands economically based on touristic activities. Then, the option of moving the future wind farms to marine locations is being promoted by the regional
* Corresponding author. E-mail address: [email protected] (R. Guerrero-Lemus). https://doi.org/10.1016/j.energy.2018.03.091 0360-5442/© 2018 Elsevier Ltd. All rights reserved.
government [2]. The article entitled “Spatial planning to estimate the offshore wind energy potential in coastal regions and islands. Practical case: The Canary Islands” [3] attempts to answer the question about the potential of offshore wind energy in the Canary Islands and electricity cost associated. However, the absence of basic considerations about the characteristics of the insular power grids involved, the composition of the electricity costs, and lack of rigor in some assumptions related to visibility constraints, offshore costs, integration costs, the mixing of data from different time periods and the references used provides unrealistic and useless results and conclusions for a necessary debate about the potential of offshore wind energy in the Canary Islands. Also, the cited Energy Strategy of the Canary Islands is a draft paper still under preliminary discussion [2], and it is being revised after the allegations of different entities. Below we will explain in more detail our assertions by means of the same sections as the article rebutted.
R. Guerrero-Lemus et al. / Energy 153 (2018) 12e16
13
2. Suitable marine areas It is not understandable how the authors select a maximum bathymetry of 500 m for the study but, in parallel, the floating wind turbines selected for the cost analysis are restricted to a maximum 125 m water depth. Also, none of the previous analyses reported by the authors consider water depths larger than 200 m. The authors justify this selection as they expect this article to serve as a longterm energy planning instrument (until 2050 and beyond), expecting also that in future the floating offshore turbines will operate in up to 500 m depths. However, the wind turbine selected for the study is currently in service in one of the Canary Islands (Gran Canaria) and the article is not providing any insight about the long-term evolution of the offshore wind turbines technology, when the average turbine sizes for new projects globally are expected to increase from about 4 MW in 2016 to almost 8 MW in 2022 [4]. Also, visibility constraints limited to 1 km in the work are not understandable, as coastal areas are very sensitive for the Canary Islands economy. Indeed, 13.1 million international tourist visited the Canary Islands only in 2016 [5]. Thus, many of the most favorable coasts, where substations are close to the shore, are also urban areas mostly devoted to touristic activities and nautical sports. Moreover, the Canary Islands have a very high population density (289 inhabitants/km2 [6,7]), mostly placed in coastal areas, as these are more suitable for economic activities. Additionally, a substantial share of the island’s surface (40.45% [8]), covering most of the unpopulated coast, is environmentally protected. We consider that a good approximation for considering visibility constraints should be to apply some of the intern boundary restrictions that the authors found in the literature and exposed in Table 1 to coastal urban areas and coastal protected areas. It is also important to mention that Canary Islands territorial waters are strongly protected by the European Union in Red NATURA2000 [9,10] and it is expected that this protection will increase in the near future [11]. These protection areas were updated in 2011 [11] and have been not properly considered in this paper. For example, all the north of La Gomera is ZEPA (Special Protection Zone for Birds) is protected (Fig. 1) [9], contradicting the planning of offshore wind farms exposed in Fig. 6 and the spatial restrictions of Table 2 of the paper rebutted. Moreover, based in the INDEMARES project [11], the Spanish Government is planning to add two new protected areas in 2019 to the Red NATURA 2000 (Fig. 2) that, added to the previous protected areas, will prevent almost all the coastal areas surrounded by Fuerteventura and Lanzarote to be considered for offshore wind without environmental actions to prevent any affection to the
Fig. 2. New protected areas proposed by the Spanish Government to be added to the Canary Islands protected coasts in 2019 [11].
ecosystems located in these areas. Of course, these affections do not exclude offshore wind energy, but increases the cost of any offshore wind farm project to avoid any environmental impact maybe to the point of making it unprofitable. Then, the planned areas for offshore wind farms in the Canary Islands (Fig. 3) that were considered in this work for reaching the 57.23 GW offshore power that could be installed, and based in a previous work produced at Universidad de Las Palmas de Gran Canaria [12], are not properly considered and the cost of many of them may increase considering environmental actions to prevent affections to the ecosystems located in them. 3. Wind farm configuration and turbines placement It is important to note that the authors do not consider buffers
Fig. 3. Planned areas for the off-shore wind farms considered in the previous works [3,12].
Fig. 1. ZEPA (Special Protection Zone for Birds) areas in the Canary Islands [9].
14
R. Guerrero-Lemus et al. / Energy 153 (2018) 12e16
between neighboring wind farms, usually estimated at about 2e20 km [13,14]. It should be necessary to limit and relate the different marine areas where the offshore wind turbines will be placed to the coastal areas that will be supplied by electricity and the substations currently placed inland. The turbine considered for the article is the G128-5.0 MW supplied by Gamesa (currently Siemens-Gamesa) [15], characterized by a hub height of 80 m and rotor diameter of 128 m. The authors choose a downwind distance of 12 diameters and a crosswind distance of 4 diameters (12D x 4D). If we avoid any consideration about buffers between neighboring wind farms and consider a rectangular cell placing a turbine in each corner, a marine surface density of (5 MW/[4 12 0.1282km2]) 6.36 MW/km2 is obtained. For example, Hywind 2 (Scotland) has 5 turbines (6 MW each) and occupies 15 km2 [16], reaching 2 MW/km2. However, in Table 9 of the article the authors reach an average array density of turbines in the Canary Islands of 14.5 MW/km2, result notably high by their own input data and comments. If we consider the 6.36 MW/km2, the expected total power installed would be about 25,100 MW instead of the 57,225 MW reported by the authors. Maybe some discrepancies can be due to the fact that many towers are placed in the limits of the area defined for offshore wind farms [12]. However, this procedure will also be wrong as the turbines should be placed at a sensible distance from the limits to avoid disturbances form other activities placed outside future new offshore wind farms placed at larger depths. Cooperative approaches between neighboring offshore activities are being proposed for minimizing future disturbances and maximizing profits [17]. Moreover, the Spanish legislation [18] considers that the surface affected by a marine concession for offshore wind farms cannot be arbitrary but must be defined in rectangles with side lengths multiple of tens of a second in latitude and longitude (~309.18 m) [19]. This legal constrain substantially reduces the power density in the marine areas defined by the authors. 4. Electricity production from wind energy The authors expose that the strong seasonal behavior of wind in Canary Islands (much stronger in summer than in winter) leads to increase in storage needs. Evaluating these storage needs is fundamental for estimating the electricity cost based in wind energy, as the current experience in one of the Canary Islands power system (El Hierro) can show [20,21]. There is also a daily pattern in wind energy, stronger in daylight time and softer at night, as we have exposed in a recent paper [22]. However, the authors avoid introducing storage costs in their calculations for obtaining the potential offshore wind capacity in the Canary Islands. Moreover, in Fig. 7 of the article the monthly variation of the wind resource is depicted, but the represented values are from only one year (1998). This data set is not representative for the aim of the study if the authors expect that this work will serve as a long-term planning instrument (until 2050 and beyond). In order to accomplish this statement, a description of the wind speeds and profiles evolution until 2050 is required as it has been performed by other authors in a similar study [23]. Thus, the wind evolution in Canary Island have been previously reported [24], assuming different scenarios according to the RCP4.5 and RCP8.5 greenhouse gas scenarios. In both scenarios, the results reveal a midterm and longterm remarkable decrease of the wind speed in the selected areas for the wind power farms. 5. Electricity production costs from wind energy The authors consider CAPEX (3750 EUR/kW for bottom-fixed turbines and 4600 EUR/kW for floating turbines) and yearly OPEX
values (112.5 EUR/kW for bottom-fixed turbines and 138 EUR/kW for floating turbines) without exposing any methodology for obtaining such values. They also consider that the decommissioning costs are negative due to the price of steel scrap and, in a very simplistic exercise, decide not to add these costs or benefits to the calculations. However, the same CAPEX and OPEX values are used in another work [12] produced by the second author and comentored by the first author, and it is based on the data exposed in Ref. [25] for the Jacket and WindFloat prototypes. Nevertheless, the characteristics of the reference scenario used in Ref. [25] for calculating the CAPEX and OPEX of the Jaket and WindFloat prototypes is completely different than the one used by the authors in terms of years of operation (20 yr.), number of turbines (100), water depth (30 m for bottom-fixed and 200 m for floating), year of commissioning (2018), installed capacity (500 MW), water depth for floating concepts (200 m), water depth for bottom-fixed concepts (30 m), distance to port and grid connection (200 km), average wind speed at hub height (10 m/s), and homogeneous medium clay as soil conditions. We propose to obtain much more rigorous data from the System Advisor Model (SAM) developed by NREL [26], as in other works [12]. Thus, as the authors consider the turbine model G128e5.0 MW (Gamesa), it is an appropriate option to consider the NREL 5 MW Offshore Reference offered by SAM in the turbine’s library (similarly to other authors [25]) and the offshore balance-ofsystem model [27] for this purpose. Then, we use the CAPEX value defined by defect for a 120 m depth floating turbine in a 160 MW wind farm but reducing the distance to landfall to 5 km. It results 5755.71 USD/kW, substantially above the value considered in this paper. Moreover, as the capacity installed in some Canary Islands is small, it is important to consider the economies of scale that are involved for the different insular power systems if no interconnections between them are proposed (as it is the case because of technical difficulties, as there exists a seabed deeper than 3000 m in most areas between islands). Then, if we install a 10 MW floating wind farm (near equivalent to the current capacity installed in El Hierro) the CAPEX rises to 18,439.61 USD/kW. Also for bottom-fixed turbines at 40 m maximum water depth the CAPEX value for a 160 MW wind farm is 4664.56 USD/kW, and for a 10 MW wind farm is 16,239.17 USD/kW. Increasing the maximum water depth from 120 m to 500 m means a 11.5% increase in BOS costs. A graph showing the maxima power peak reached in the different Canary Islands power grids and the economies of scale between the total installed cost and number of turbines considered for the wind farm using the SAM software is exposed in Fig. 4. The offshore wind turbine considered is semisubmersible, placed at 120 m maximum water depth, drag embedment, 3 mooring lines, a distance of 5 km to landfall and a distance of 90 km from the inshore assembly area to site. Moreover, the electricity generation of offshore wind is usually estimated 10e15% lower than the energy calculation based on power curves from wind turbines, due to electrical losses in the transformers and cabling, and the wind turbine downtime for schedule maintenance or technical failure [28]. However, the authors have decided to calculate the electricity production from wind energy without considering any losses because they consider that the potential offshore wind energy is so high in comparison to demand, that “a 10% generation reduction practically does not affect the analysis of results”. This is an unfortunate assertion for a scientific paper, and no power utility is expected to ignore a 10e15% generation reduction. There are also important inconsistencies about time periods used when the authors consider electricity prices from 2014, electricity demand values from 2015, installed conventional and renewable capacity and energy produced from 2016, CAPEX and
R. Guerrero-Lemus et al. / Energy 153 (2018) 12e16
15
25000
20000
La Palma
5000
La Gomera
10000
Gran Canaria Tenerife
Fuerteventura Lanzarote
15000
El Hierro
Total installed cost per kW (USD)
30000
0 10
100
number of 5 MW wind turbines (related to maximum power peak per island) Fig. 4. Curve showing the relation between the total installed cost per kW and the number of 5 MW semisubmersible turbines placed at 120 m maximum water depth, drag embedment and 3 mooring lines, in an offshore wind farm using the SAM software [26], and compared to the maxima power peaks reached in the different power grids of the Canary Islands.
OPEX from 2009, 2010 and 2017, and average monthly wind speed from 1998. It is also very important to remark that the average electricity prices (175.61 EUR/MWh in 2014; 138.03 EUR/MWh in 2015; and 130.19 EUR/MWh in 2016) [29] also include some quite fixed and regulated costs to secure the service of conventional power plants, as the conventional power units can be programmed in advance to cover any forecasted demand of electricity. This quite fixed retribution is not considered for electricity produced from wind energy in the Spanish regulation because of the inherent intermittency and lack of dispatchability of wind energy, and should be substracted from the electricity prices considered by the authors. These
average monthly electricity cost capacity cost
175 150
EUR/MWh
125 100 75 50 25 0 2014
2015
2016
2017
2018
Year Fig. 5. Average monthly electricity costs and capacity costs related to the guarantee of supply from the conventional power units.
regulated costs are directly related to the capacity of each conventional power unit and average 35.46 EUR/MWh per month in the period Jan2015 e Sep2017 (Fig. 5) [30]. Then, the electricity cost that should be considered in 2015 is 102.57 EUR/MWh, which is the variable cost related to the production of electricity from the conventional power units, and results lower than any of the offshore marginal costs from wind energy obtained by the authors (Table 15 of the rebutted paper) for any power grid in the Canary Islands (the lowest marginal cost calculated by the authors was 131 EUR/MWh for depths < 50 m in Gran Canaria). Also, when the LCOE is calculated by the author, a lifetime of 30 years is introduced. This value is above the standards and expectations for this technology in the long term [31,32] due to the challenges still existing in this area (materials, corrosiveness in marine areas, mechanical fatigue, etc.). The integration costs of 30 EUR/MWh used by the authors have been obtained from a paper where the electricity from wind energy is integrated in extremely large and interconnected power grids (Germany), average electricity prices about 70 EUR/MWh and penetration rates between 0 and 40% [33]. However, in the small and insular power systems of the Canary Island the electricity price was about 138.03 EUR/MWh in the time period considered and current installed wind capacity is close to produce curtailment effects in the main power grids [22]. Also, the integration costs rise with the penetration of electricity from wind energy. In this regard, the same researchers consider that an increase in the penetration rate of one percentage point in electricity from wind energy increases the profile costs by 0.5 EUR/MWh, almost ten times more than the balancing costs [33]. Moreover, we consider that it should be not recommended to publish a paper about electricity cost analysis from offshore wind only based on CAPEX obtained from a submarine cable consultancy [34]. Also, direct information about investment costs of the offshore wind projects should be filtered, as many of them are not
16
R. Guerrero-Lemus et al. / Energy 153 (2018) 12e16
comparable because of specific incentives offered by the governments involved in the development of the projects. A prestigious international agency or national lab [31,32] with a neutral vision of the different energy sectors should be mandatory. Also, it is important to mention that the project cost for Windfloat2 (Portugal) and Fukushima (Japan) offshore wind farms used in the rebutted work [35,36], and Ref. 32 are not available. Nevertheless, the comments and suggestions of this paragraph are not very important because the authors decided the CAPEX values applied in the paper without exposing any methodology, as it has been mentioned above. 6. Conclusions This rebuttal letter has demonstrated that the lack of basic considerations about the characteristics of the insular power grids involved and the composition of the electricity costs, and lack of rigor in some assumptions related to visibility constraints, off-shore costs, integration costs, mixing data from different time periods and references used makes the results obtained of the article entitled “Spatial planning to estimate the offshore wind energy potential in coastal regions and islands. Practical case: The Canary Islands” unreal and useless for a necessary debate about the potential of offshore wind energy in the Canary Islands. We also demonstrate that the main result about potential offshore wind capacity is much lower, and the consideration that the electricity costs form offshore wind provided by the authors is lower than the current electricity cost is wrong and, in fact, the cost of electricity from offshore wind is higher in the time period the analysis was made and at present. A new research article about offshore wind potential in the Canary Islands and including all these considerations is currently being developed by the authors of this rebuttal letter. References [1] Data and Statistics - IRENA REsource n.d. http://resourceirena.irena.org/ gateway/dashboard/?topic¼4&subTopic¼17 (Accessed 12 December 2017). tica [2] Gobierno de Canarias. Economía presenta al sector la Estrategia Energe implantando hasta 2025-Lanzadera - Portal de que el Gobierno esta n Comunicacio 2017. http://www.gobcan.es/noticias/lanzadera/85019/ economia-presenta-sector-estrategia-energetica-gobierno-implantando-2025 (Accessed 12 December 2017). [3] Schallenberg-Rodríguez J, Montesdeoca NG. Spatial planning to estimate the offshore wind energy potential in coastal regions and islands. Practical case: the Canary Islands. Energy 2018;143:91e103. https://doi.org/10.1016/ J.ENERGY.2017.10.084. [4] IEA. Renewables 2017 n.d. https://www.iea.org/renewables/ (Accessed 7 November 2017). [5] Statista. The Statistics Portal. International tourist numbers canary islands 2001-2016 | Statistic n.d. https://www.statista.com/statistics/449189/yearlynumber-of-international-tourists-visiting-the-canary-islands/ (Accessed 20 November 2017). noma de Canarias. Superficie por [6] ISTAC. Estadísticas de la Comunidad Auto islas de Canarias n.d. http://www.gobiernodecanarias.org/istac/jaxi-istac/ tabla.do (Accessed 20 November 2017). n residente en Canarias [7] INE. Instituto Nacional de Estadística. Poblacio (Tabla9681) n.d. http://www.ine.es/jaxiT3/Datos.htm?t¼9681 (Accessed 20 November 2017). noma de Canarias. Espacios natu[8] ISTAC: Estadísticas de la Comunidad Auto rales protegidos n.d. http://www.gobiernodecanarias.org/istac/jaxi-istac/ tabla.do (Accessed 20 November 2017). n Territorial de Canarias (IDECanarias). GRAFCAN. Zonas [9] Sistema de Informacio n para las Aves (ZEPA) n.d. http://visor.grafcan.es/ de Espacial Proteccio visorweb/default.php?svc¼svcZEPA&lat¼28.3&lng¼-15.8&zoom¼8&lang¼es (Accessed 24 November 2017). n Territorial de Canarias (IDECanarias). GRAFCAN. [10] Sistema de Informacio n (ZEC) n.d. http://visor.grafcan.es/visorweb/ Zonas Especiales de Conservacio default.php?svc¼svcZEC&lat¼28.3&lng¼-15.8&zoom¼8&lang¼es (Accessed 24 November 2017).
[11] Ministerio de Agricultura y Pesca A y MAG de E. Proyecto LIFEþ INDEMARES n del medio Espacios Marinos Protegidos - Biodiversidad marina - Proteccio marino - Costas y Medio Marino - mapama.es n.d. http://www.mapama.gob. es/es/costas/temas/proteccion-medio-marino/biodiversidad-marina/espaciosmarinos-protegidos/red-natura-2000-ambito-marino/red-natura-2000declaracion-lugares.aspx (Accessed 24 November 2017). lica en Mar [12] García Montesdeoca N del P. Estudio del Potencial de Energía Eo Abierto (Off-Shore) en Canarias. Universidad de Las Plamas de Gran Canaria; 2016. €ller B. Offshore wind energy potential in China: under technical, [13] Hong L, Mo spatial and economic constraints. Energy 2011;36:4482e91. https://doi.org/ 10.1016/J.ENERGY.2011.03.071. [14] Frandsen S, Barthelmie R, Pryor S, Rathmann O, Larsen S, Højstrup J, et al. The necessary distance between large wind farms offshore - study. 2005. logo. Factsheet Gamesa G128e5.0 MW n.d.:54e5. [15] Cata [16] Hywind Scotland Pilot Park - 4C Offshore n.d. http://www.4coffshore.com/ windfarms/hywind-scotland-pilot-park-united-kingdom-uk76.html (Accessed 12 December 2017). lez J, Burgos Pay [17] Serrano Gonza an M, Riquelme Santos JM. Optimal design of neighbouring offshore wind farms: a co-evolutionary approach. Appl Energy 2018;209:140e52. https://doi.org/10.1016/J.APENERGY.2017.10.120. [18] (BOE) BO del E. Real Decreto 1028/2007, de 20 de julio, por el que se establece n de las solicitudes de el procedimiento administrativo para la tramitacio n de instalaciones de generacio n ele ctrica en el mar territorial. autorizacio 2007. [19] National Geospatial-Intelligence Agency. U.S. Government. Length of a degree of latitude and longitude n.d. https://msi.nga.mil/MSISiteContent/StaticFiles/ Calculators/degree.html (Accessed 10 December 2017). [20] Bueno C, Carta JA. Technicaleeconomic analysis of wind-powered pumped hydrostorage systems. Part II: model application to the island of El Hierro. Sol Energy 2005;78:396e405. https://doi.org/10.1016/J.SOLENER.2004.08.007. Wilhelmi JR. Power-frenchez-Ferna ndez JA, [21] Martínez-Lucas G, Sarasúa JI, Sa quency control of hydropower plants with long penstocks in isolated systems with wind generation. Renew Energy 2015;83:245e55. https://doi.org/ 10.1016/J.RENENE.2015.04.032. lez-Díaz B, Ríos G, Dib RN. Study of the new Spanish [22] Guerrero-Lemus R, Gonza legislation applied to an insular system that has achieved grid parity on PV and wind energy. Renew Sustain Energy Rev 2015;49. https://doi.org/ 10.1016/j.rser.2015.04.079. [23] Chancham C, Waeqsak J, Gagnon Y. Offshore wind resource assessment and wind power plant optimization in the Gulf of Thailand. Energy 2017;139: 706e31. https://doi.org/10.1016/J.ENERGY.2017.08.026. rez JC, Díaz JP. Expo sito, Francisco Javier. Future projections of [24] Gonz alez A, Pe wind resource in a mountainous archipelago, Canary Islands. Renew Energy 2017;104:120e8. https://doi.org/10.1016/J.RENENE.2016.12.021. [25] Myhr A, Bjerkseter C, Ågotnes A, Nygaard TA. Levelised cost of energy for offshore floating wind turbines in a life cycle perspective. Renew Energy 2014;66:714e28. https://doi.org/10.1016/J.RENENE.2014.01.017. [26] National Renewable Energy Laboratory. System advisor model (SAM) 2017.9.5 n.d. https://sam.nrel.gov/download (Accessed 20 November 2017). [27] Maness M, Maples B, Smith A. NREL offshore balance-of-system model. 2017. [28] Blanco MI. The economics of wind energy. Renew Sustain Energy Rev 2009;13:1372e82. https://doi.org/10.1016/J.RSER.2008.09.004. n del Operador del Sistema (ESIOS). Red Ele ctrica de [29] Sistema de Informacio ~ a (REE). Precio medio de la demanda de los SNP por sistema. Canarias. Espan n.d. https://www.esios.ree.es/es/analisis/573?vis¼1&start_date¼01-01-2010 T00%3A00&end_date¼31-12-2017T23%3A50&compare_start_date¼01-012009T00%3A00&groupby¼year&compare_indicators¼574&geoids¼8742 (Accessed 20 November 2017). [30] (CNMC) CN del M y la C. Liquidaciones Provisionales de las Actividades Regctrico. n.d. uladas del Sector Ele [31] Wiser R, Jenni K, Seel J, Baker E, Hand M, Lantz E, et al. Expert elicitation survey on future wind energy costs. Nat Energy 2016;1:16135. https:// doi.org/10.1038/nenergy.2016.135. [32] Wiser RH, Jenni K, Seel J, Baker E, Hand MM, Lantz E, et al. Forecasting wind energy costs and cost drivers: the views of the World’s leading Experts | electricity markets and policy group. LBNL-1005717. 2016 (Accessed 12 December 2017), https://emp.lbl.gov/publications/forecasting-wind-energycosts-and. [33] Hirth L, Ueckerdt F, Edenhofer O. Integration costs revisited e an economic framework for wind and solar variability. Renew Energy 2015;74:925e39. https://doi.org/10.1016/J.RENENE.2014.08.065. [34] 4C Offshore. Submarine Cable Consultancy. n.d (Accessed 20 November 2017), http://www.4coffshore.com/. [35] WindFloat Atlantic (WFA) Phase 2-4C Offshore n.d. http://www.4coffshore. com/windfarms/windfloat-atlantic-(wfa)-phase-2-portugal-pt04.html (accessed December 12, 2017). [36] Fukushima floating OffshoRe wind FARm demonstration project (Forward) phase 1-4C offshore n.d. http://www.4coffshore.com/windfarms/fukushimafloating-offshore-wind-farm-demonstration-project-(forward)-phase-1japan-jp06.html (accessed December 12, 2017).
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Rebuttal letter to the article entitled: “Spatial planning to estimate the offshore wind energy potential in coastal regions and islands. Practical case: The Canary Islands” lez-Díaz c Ricardo Guerrero-Lemus a, *, Ignacio de la Nuez b, Benjamín Gonza nchez S/N, 38206, S/C de Tenerife, Spain Departamento de Física, Universidad de La Laguna, Avenida Astrofísico Francisco Sa tica, Universidad de Las Palmas de Gran Canaria, Campus Universitario de Tafira, 35010, Las Palmas de nica y Automa Departamento de Ingeniería Electro Gran Canarias, Spain c nchez S/N, 38206, S/C de Tenerife, Spain Departamento de Ingeniería Industrial, Universidad de La Laguna, Avenida Astrofísico Francisco Sa a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Available online 6 April 2018
The objective of this rebuttal letter is to provide a critical analysis of the article entitled “Spatial planning to estimate the offshore wind energy potential in coastal regions and islands. Practical case: The Canary Islands” [Energy 143 (2018) 91e103], mainly in relation to its methodology, suitable marine areas and electricity production costs. The absence of basic considerations about the characteristics of the insular power grids, the composition of the electricity costs in the Canary Islands, and the lack of rigor in some assumptions related to visibility constraints, offshore costs, integration costs, the mixing of data from different time periods and the references used, provides unrealistic and useless results for a necessary debate about the potential of offshore wind energy in the Canary Islands. In this rebuttal letter we will also demonstrate that the potential offshore wind capacity calculated by the authors is much lower. Moreover, the assertion that the electricity cost from offshore wind calculated by the authors is lower than the current electricity cost is wrong and, in fact, the cost of electricity from offshore wind is higher in the time period when the analysis was made, and also at present. © 2018 Elsevier Ltd. All rights reserved.
Keywords: Offshore wind Canary islands Cost analysis Wind energy
1. Introduction Electricity from offshore wind energy is growing substantially around the world (14,086.3 MW in 2016 [1]), mainly in developed countries with marine territorial areas where a shallow continental platform permits the placement of such devices not interfering with other marine activities and avoiding substantial environmental concerns. In Canary Islands, the debate about the location of offshore wind farms in future is growing because of the limited surface available onshore. There exists a growing desire for achieving a 100% electricity production from renewable resources, but also a growing concern about the location of wind farms onshore and the environmental impact associated for islands economically based on touristic activities. Then, the option of moving the future wind farms to marine locations is being promoted by the regional
* Corresponding author. E-mail address: [email protected] (R. Guerrero-Lemus). https://doi.org/10.1016/j.energy.2018.03.091 0360-5442/© 2018 Elsevier Ltd. All rights reserved.
government [2]. The article entitled “Spatial planning to estimate the offshore wind energy potential in coastal regions and islands. Practical case: The Canary Islands” [3] attempts to answer the question about the potential of offshore wind energy in the Canary Islands and electricity cost associated. However, the absence of basic considerations about the characteristics of the insular power grids involved, the composition of the electricity costs, and lack of rigor in some assumptions related to visibility constraints, offshore costs, integration costs, the mixing of data from different time periods and the references used provides unrealistic and useless results and conclusions for a necessary debate about the potential of offshore wind energy in the Canary Islands. Also, the cited Energy Strategy of the Canary Islands is a draft paper still under preliminary discussion [2], and it is being revised after the allegations of different entities. Below we will explain in more detail our assertions by means of the same sections as the article rebutted.
R. Guerrero-Lemus et al. / Energy 153 (2018) 12e16
13
2. Suitable marine areas It is not understandable how the authors select a maximum bathymetry of 500 m for the study but, in parallel, the floating wind turbines selected for the cost analysis are restricted to a maximum 125 m water depth. Also, none of the previous analyses reported by the authors consider water depths larger than 200 m. The authors justify this selection as they expect this article to serve as a longterm energy planning instrument (until 2050 and beyond), expecting also that in future the floating offshore turbines will operate in up to 500 m depths. However, the wind turbine selected for the study is currently in service in one of the Canary Islands (Gran Canaria) and the article is not providing any insight about the long-term evolution of the offshore wind turbines technology, when the average turbine sizes for new projects globally are expected to increase from about 4 MW in 2016 to almost 8 MW in 2022 [4]. Also, visibility constraints limited to 1 km in the work are not understandable, as coastal areas are very sensitive for the Canary Islands economy. Indeed, 13.1 million international tourist visited the Canary Islands only in 2016 [5]. Thus, many of the most favorable coasts, where substations are close to the shore, are also urban areas mostly devoted to touristic activities and nautical sports. Moreover, the Canary Islands have a very high population density (289 inhabitants/km2 [6,7]), mostly placed in coastal areas, as these are more suitable for economic activities. Additionally, a substantial share of the island’s surface (40.45% [8]), covering most of the unpopulated coast, is environmentally protected. We consider that a good approximation for considering visibility constraints should be to apply some of the intern boundary restrictions that the authors found in the literature and exposed in Table 1 to coastal urban areas and coastal protected areas. It is also important to mention that Canary Islands territorial waters are strongly protected by the European Union in Red NATURA2000 [9,10] and it is expected that this protection will increase in the near future [11]. These protection areas were updated in 2011 [11] and have been not properly considered in this paper. For example, all the north of La Gomera is ZEPA (Special Protection Zone for Birds) is protected (Fig. 1) [9], contradicting the planning of offshore wind farms exposed in Fig. 6 and the spatial restrictions of Table 2 of the paper rebutted. Moreover, based in the INDEMARES project [11], the Spanish Government is planning to add two new protected areas in 2019 to the Red NATURA 2000 (Fig. 2) that, added to the previous protected areas, will prevent almost all the coastal areas surrounded by Fuerteventura and Lanzarote to be considered for offshore wind without environmental actions to prevent any affection to the
Fig. 2. New protected areas proposed by the Spanish Government to be added to the Canary Islands protected coasts in 2019 [11].
ecosystems located in these areas. Of course, these affections do not exclude offshore wind energy, but increases the cost of any offshore wind farm project to avoid any environmental impact maybe to the point of making it unprofitable. Then, the planned areas for offshore wind farms in the Canary Islands (Fig. 3) that were considered in this work for reaching the 57.23 GW offshore power that could be installed, and based in a previous work produced at Universidad de Las Palmas de Gran Canaria [12], are not properly considered and the cost of many of them may increase considering environmental actions to prevent affections to the ecosystems located in them. 3. Wind farm configuration and turbines placement It is important to note that the authors do not consider buffers
Fig. 3. Planned areas for the off-shore wind farms considered in the previous works [3,12].
Fig. 1. ZEPA (Special Protection Zone for Birds) areas in the Canary Islands [9].
14
R. Guerrero-Lemus et al. / Energy 153 (2018) 12e16
between neighboring wind farms, usually estimated at about 2e20 km [13,14]. It should be necessary to limit and relate the different marine areas where the offshore wind turbines will be placed to the coastal areas that will be supplied by electricity and the substations currently placed inland. The turbine considered for the article is the G128-5.0 MW supplied by Gamesa (currently Siemens-Gamesa) [15], characterized by a hub height of 80 m and rotor diameter of 128 m. The authors choose a downwind distance of 12 diameters and a crosswind distance of 4 diameters (12D x 4D). If we avoid any consideration about buffers between neighboring wind farms and consider a rectangular cell placing a turbine in each corner, a marine surface density of (5 MW/[4 12 0.1282km2]) 6.36 MW/km2 is obtained. For example, Hywind 2 (Scotland) has 5 turbines (6 MW each) and occupies 15 km2 [16], reaching 2 MW/km2. However, in Table 9 of the article the authors reach an average array density of turbines in the Canary Islands of 14.5 MW/km2, result notably high by their own input data and comments. If we consider the 6.36 MW/km2, the expected total power installed would be about 25,100 MW instead of the 57,225 MW reported by the authors. Maybe some discrepancies can be due to the fact that many towers are placed in the limits of the area defined for offshore wind farms [12]. However, this procedure will also be wrong as the turbines should be placed at a sensible distance from the limits to avoid disturbances form other activities placed outside future new offshore wind farms placed at larger depths. Cooperative approaches between neighboring offshore activities are being proposed for minimizing future disturbances and maximizing profits [17]. Moreover, the Spanish legislation [18] considers that the surface affected by a marine concession for offshore wind farms cannot be arbitrary but must be defined in rectangles with side lengths multiple of tens of a second in latitude and longitude (~309.18 m) [19]. This legal constrain substantially reduces the power density in the marine areas defined by the authors. 4. Electricity production from wind energy The authors expose that the strong seasonal behavior of wind in Canary Islands (much stronger in summer than in winter) leads to increase in storage needs. Evaluating these storage needs is fundamental for estimating the electricity cost based in wind energy, as the current experience in one of the Canary Islands power system (El Hierro) can show [20,21]. There is also a daily pattern in wind energy, stronger in daylight time and softer at night, as we have exposed in a recent paper [22]. However, the authors avoid introducing storage costs in their calculations for obtaining the potential offshore wind capacity in the Canary Islands. Moreover, in Fig. 7 of the article the monthly variation of the wind resource is depicted, but the represented values are from only one year (1998). This data set is not representative for the aim of the study if the authors expect that this work will serve as a long-term planning instrument (until 2050 and beyond). In order to accomplish this statement, a description of the wind speeds and profiles evolution until 2050 is required as it has been performed by other authors in a similar study [23]. Thus, the wind evolution in Canary Island have been previously reported [24], assuming different scenarios according to the RCP4.5 and RCP8.5 greenhouse gas scenarios. In both scenarios, the results reveal a midterm and longterm remarkable decrease of the wind speed in the selected areas for the wind power farms. 5. Electricity production costs from wind energy The authors consider CAPEX (3750 EUR/kW for bottom-fixed turbines and 4600 EUR/kW for floating turbines) and yearly OPEX
values (112.5 EUR/kW for bottom-fixed turbines and 138 EUR/kW for floating turbines) without exposing any methodology for obtaining such values. They also consider that the decommissioning costs are negative due to the price of steel scrap and, in a very simplistic exercise, decide not to add these costs or benefits to the calculations. However, the same CAPEX and OPEX values are used in another work [12] produced by the second author and comentored by the first author, and it is based on the data exposed in Ref. [25] for the Jacket and WindFloat prototypes. Nevertheless, the characteristics of the reference scenario used in Ref. [25] for calculating the CAPEX and OPEX of the Jaket and WindFloat prototypes is completely different than the one used by the authors in terms of years of operation (20 yr.), number of turbines (100), water depth (30 m for bottom-fixed and 200 m for floating), year of commissioning (2018), installed capacity (500 MW), water depth for floating concepts (200 m), water depth for bottom-fixed concepts (30 m), distance to port and grid connection (200 km), average wind speed at hub height (10 m/s), and homogeneous medium clay as soil conditions. We propose to obtain much more rigorous data from the System Advisor Model (SAM) developed by NREL [26], as in other works [12]. Thus, as the authors consider the turbine model G128e5.0 MW (Gamesa), it is an appropriate option to consider the NREL 5 MW Offshore Reference offered by SAM in the turbine’s library (similarly to other authors [25]) and the offshore balance-ofsystem model [27] for this purpose. Then, we use the CAPEX value defined by defect for a 120 m depth floating turbine in a 160 MW wind farm but reducing the distance to landfall to 5 km. It results 5755.71 USD/kW, substantially above the value considered in this paper. Moreover, as the capacity installed in some Canary Islands is small, it is important to consider the economies of scale that are involved for the different insular power systems if no interconnections between them are proposed (as it is the case because of technical difficulties, as there exists a seabed deeper than 3000 m in most areas between islands). Then, if we install a 10 MW floating wind farm (near equivalent to the current capacity installed in El Hierro) the CAPEX rises to 18,439.61 USD/kW. Also for bottom-fixed turbines at 40 m maximum water depth the CAPEX value for a 160 MW wind farm is 4664.56 USD/kW, and for a 10 MW wind farm is 16,239.17 USD/kW. Increasing the maximum water depth from 120 m to 500 m means a 11.5% increase in BOS costs. A graph showing the maxima power peak reached in the different Canary Islands power grids and the economies of scale between the total installed cost and number of turbines considered for the wind farm using the SAM software is exposed in Fig. 4. The offshore wind turbine considered is semisubmersible, placed at 120 m maximum water depth, drag embedment, 3 mooring lines, a distance of 5 km to landfall and a distance of 90 km from the inshore assembly area to site. Moreover, the electricity generation of offshore wind is usually estimated 10e15% lower than the energy calculation based on power curves from wind turbines, due to electrical losses in the transformers and cabling, and the wind turbine downtime for schedule maintenance or technical failure [28]. However, the authors have decided to calculate the electricity production from wind energy without considering any losses because they consider that the potential offshore wind energy is so high in comparison to demand, that “a 10% generation reduction practically does not affect the analysis of results”. This is an unfortunate assertion for a scientific paper, and no power utility is expected to ignore a 10e15% generation reduction. There are also important inconsistencies about time periods used when the authors consider electricity prices from 2014, electricity demand values from 2015, installed conventional and renewable capacity and energy produced from 2016, CAPEX and
R. Guerrero-Lemus et al. / Energy 153 (2018) 12e16
15
25000
20000
La Palma
5000
La Gomera
10000
Gran Canaria Tenerife
Fuerteventura Lanzarote
15000
El Hierro
Total installed cost per kW (USD)
30000
0 10
100
number of 5 MW wind turbines (related to maximum power peak per island) Fig. 4. Curve showing the relation between the total installed cost per kW and the number of 5 MW semisubmersible turbines placed at 120 m maximum water depth, drag embedment and 3 mooring lines, in an offshore wind farm using the SAM software [26], and compared to the maxima power peaks reached in the different power grids of the Canary Islands.
OPEX from 2009, 2010 and 2017, and average monthly wind speed from 1998. It is also very important to remark that the average electricity prices (175.61 EUR/MWh in 2014; 138.03 EUR/MWh in 2015; and 130.19 EUR/MWh in 2016) [29] also include some quite fixed and regulated costs to secure the service of conventional power plants, as the conventional power units can be programmed in advance to cover any forecasted demand of electricity. This quite fixed retribution is not considered for electricity produced from wind energy in the Spanish regulation because of the inherent intermittency and lack of dispatchability of wind energy, and should be substracted from the electricity prices considered by the authors. These
average monthly electricity cost capacity cost
175 150
EUR/MWh
125 100 75 50 25 0 2014
2015
2016
2017
2018
Year Fig. 5. Average monthly electricity costs and capacity costs related to the guarantee of supply from the conventional power units.
regulated costs are directly related to the capacity of each conventional power unit and average 35.46 EUR/MWh per month in the period Jan2015 e Sep2017 (Fig. 5) [30]. Then, the electricity cost that should be considered in 2015 is 102.57 EUR/MWh, which is the variable cost related to the production of electricity from the conventional power units, and results lower than any of the offshore marginal costs from wind energy obtained by the authors (Table 15 of the rebutted paper) for any power grid in the Canary Islands (the lowest marginal cost calculated by the authors was 131 EUR/MWh for depths < 50 m in Gran Canaria). Also, when the LCOE is calculated by the author, a lifetime of 30 years is introduced. This value is above the standards and expectations for this technology in the long term [31,32] due to the challenges still existing in this area (materials, corrosiveness in marine areas, mechanical fatigue, etc.). The integration costs of 30 EUR/MWh used by the authors have been obtained from a paper where the electricity from wind energy is integrated in extremely large and interconnected power grids (Germany), average electricity prices about 70 EUR/MWh and penetration rates between 0 and 40% [33]. However, in the small and insular power systems of the Canary Island the electricity price was about 138.03 EUR/MWh in the time period considered and current installed wind capacity is close to produce curtailment effects in the main power grids [22]. Also, the integration costs rise with the penetration of electricity from wind energy. In this regard, the same researchers consider that an increase in the penetration rate of one percentage point in electricity from wind energy increases the profile costs by 0.5 EUR/MWh, almost ten times more than the balancing costs [33]. Moreover, we consider that it should be not recommended to publish a paper about electricity cost analysis from offshore wind only based on CAPEX obtained from a submarine cable consultancy [34]. Also, direct information about investment costs of the offshore wind projects should be filtered, as many of them are not
16
R. Guerrero-Lemus et al. / Energy 153 (2018) 12e16
comparable because of specific incentives offered by the governments involved in the development of the projects. A prestigious international agency or national lab [31,32] with a neutral vision of the different energy sectors should be mandatory. Also, it is important to mention that the project cost for Windfloat2 (Portugal) and Fukushima (Japan) offshore wind farms used in the rebutted work [35,36], and Ref. 32 are not available. Nevertheless, the comments and suggestions of this paragraph are not very important because the authors decided the CAPEX values applied in the paper without exposing any methodology, as it has been mentioned above. 6. Conclusions This rebuttal letter has demonstrated that the lack of basic considerations about the characteristics of the insular power grids involved and the composition of the electricity costs, and lack of rigor in some assumptions related to visibility constraints, off-shore costs, integration costs, mixing data from different time periods and references used makes the results obtained of the article entitled “Spatial planning to estimate the offshore wind energy potential in coastal regions and islands. Practical case: The Canary Islands” unreal and useless for a necessary debate about the potential of offshore wind energy in the Canary Islands. We also demonstrate that the main result about potential offshore wind capacity is much lower, and the consideration that the electricity costs form offshore wind provided by the authors is lower than the current electricity cost is wrong and, in fact, the cost of electricity from offshore wind is higher in the time period the analysis was made and at present. A new research article about offshore wind potential in the Canary Islands and including all these considerations is currently being developed by the authors of this rebuttal letter. References [1] Data and Statistics - IRENA REsource n.d. http://resourceirena.irena.org/ gateway/dashboard/?topic¼4&subTopic¼17 (Accessed 12 December 2017). tica [2] Gobierno de Canarias. Economía presenta al sector la Estrategia Energe implantando hasta 2025-Lanzadera - Portal de que el Gobierno esta n Comunicacio 2017. http://www.gobcan.es/noticias/lanzadera/85019/ economia-presenta-sector-estrategia-energetica-gobierno-implantando-2025 (Accessed 12 December 2017). [3] Schallenberg-Rodríguez J, Montesdeoca NG. Spatial planning to estimate the offshore wind energy potential in coastal regions and islands. Practical case: the Canary Islands. Energy 2018;143:91e103. https://doi.org/10.1016/ J.ENERGY.2017.10.084. [4] IEA. Renewables 2017 n.d. https://www.iea.org/renewables/ (Accessed 7 November 2017). [5] Statista. The Statistics Portal. International tourist numbers canary islands 2001-2016 | Statistic n.d. https://www.statista.com/statistics/449189/yearlynumber-of-international-tourists-visiting-the-canary-islands/ (Accessed 20 November 2017). noma de Canarias. Superficie por [6] ISTAC. Estadísticas de la Comunidad Auto islas de Canarias n.d. http://www.gobiernodecanarias.org/istac/jaxi-istac/ tabla.do (Accessed 20 November 2017). n residente en Canarias [7] INE. Instituto Nacional de Estadística. Poblacio (Tabla9681) n.d. http://www.ine.es/jaxiT3/Datos.htm?t¼9681 (Accessed 20 November 2017). noma de Canarias. Espacios natu[8] ISTAC: Estadísticas de la Comunidad Auto rales protegidos n.d. http://www.gobiernodecanarias.org/istac/jaxi-istac/ tabla.do (Accessed 20 November 2017). n Territorial de Canarias (IDECanarias). GRAFCAN. Zonas [9] Sistema de Informacio n para las Aves (ZEPA) n.d. http://visor.grafcan.es/ de Espacial Proteccio visorweb/default.php?svc¼svcZEPA&lat¼28.3&lng¼-15.8&zoom¼8&lang¼es (Accessed 24 November 2017). n Territorial de Canarias (IDECanarias). GRAFCAN. [10] Sistema de Informacio n (ZEC) n.d. http://visor.grafcan.es/visorweb/ Zonas Especiales de Conservacio default.php?svc¼svcZEC&lat¼28.3&lng¼-15.8&zoom¼8&lang¼es (Accessed 24 November 2017).
[11] Ministerio de Agricultura y Pesca A y MAG de E. Proyecto LIFEþ INDEMARES n del medio Espacios Marinos Protegidos - Biodiversidad marina - Proteccio marino - Costas y Medio Marino - mapama.es n.d. http://www.mapama.gob. es/es/costas/temas/proteccion-medio-marino/biodiversidad-marina/espaciosmarinos-protegidos/red-natura-2000-ambito-marino/red-natura-2000declaracion-lugares.aspx (Accessed 24 November 2017). lica en Mar [12] García Montesdeoca N del P. Estudio del Potencial de Energía Eo Abierto (Off-Shore) en Canarias. Universidad de Las Plamas de Gran Canaria; 2016. €ller B. Offshore wind energy potential in China: under technical, [13] Hong L, Mo spatial and economic constraints. Energy 2011;36:4482e91. https://doi.org/ 10.1016/J.ENERGY.2011.03.071. [14] Frandsen S, Barthelmie R, Pryor S, Rathmann O, Larsen S, Højstrup J, et al. The necessary distance between large wind farms offshore - study. 2005. logo. Factsheet Gamesa G128e5.0 MW n.d.:54e5. [15] Cata [16] Hywind Scotland Pilot Park - 4C Offshore n.d. http://www.4coffshore.com/ windfarms/hywind-scotland-pilot-park-united-kingdom-uk76.html (Accessed 12 December 2017). lez J, Burgos Pay [17] Serrano Gonza an M, Riquelme Santos JM. Optimal design of neighbouring offshore wind farms: a co-evolutionary approach. Appl Energy 2018;209:140e52. https://doi.org/10.1016/J.APENERGY.2017.10.120. [18] (BOE) BO del E. Real Decreto 1028/2007, de 20 de julio, por el que se establece n de las solicitudes de el procedimiento administrativo para la tramitacio n de instalaciones de generacio n ele ctrica en el mar territorial. autorizacio 2007. [19] National Geospatial-Intelligence Agency. U.S. Government. Length of a degree of latitude and longitude n.d. https://msi.nga.mil/MSISiteContent/StaticFiles/ Calculators/degree.html (Accessed 10 December 2017). [20] Bueno C, Carta JA. Technicaleeconomic analysis of wind-powered pumped hydrostorage systems. Part II: model application to the island of El Hierro. Sol Energy 2005;78:396e405. https://doi.org/10.1016/J.SOLENER.2004.08.007. Wilhelmi JR. Power-frenchez-Ferna ndez JA, [21] Martínez-Lucas G, Sarasúa JI, Sa quency control of hydropower plants with long penstocks in isolated systems with wind generation. Renew Energy 2015;83:245e55. https://doi.org/ 10.1016/J.RENENE.2015.04.032. lez-Díaz B, Ríos G, Dib RN. Study of the new Spanish [22] Guerrero-Lemus R, Gonza legislation applied to an insular system that has achieved grid parity on PV and wind energy. Renew Sustain Energy Rev 2015;49. https://doi.org/ 10.1016/j.rser.2015.04.079. [23] Chancham C, Waeqsak J, Gagnon Y. Offshore wind resource assessment and wind power plant optimization in the Gulf of Thailand. Energy 2017;139: 706e31. https://doi.org/10.1016/J.ENERGY.2017.08.026. rez JC, Díaz JP. Expo sito, Francisco Javier. Future projections of [24] Gonz alez A, Pe wind resource in a mountainous archipelago, Canary Islands. Renew Energy 2017;104:120e8. https://doi.org/10.1016/J.RENENE.2016.12.021. [25] Myhr A, Bjerkseter C, Ågotnes A, Nygaard TA. Levelised cost of energy for offshore floating wind turbines in a life cycle perspective. Renew Energy 2014;66:714e28. https://doi.org/10.1016/J.RENENE.2014.01.017. [26] National Renewable Energy Laboratory. System advisor model (SAM) 2017.9.5 n.d. https://sam.nrel.gov/download (Accessed 20 November 2017). [27] Maness M, Maples B, Smith A. NREL offshore balance-of-system model. 2017. [28] Blanco MI. The economics of wind energy. Renew Sustain Energy Rev 2009;13:1372e82. https://doi.org/10.1016/J.RSER.2008.09.004. n del Operador del Sistema (ESIOS). Red Ele ctrica de [29] Sistema de Informacio ~ a (REE). Precio medio de la demanda de los SNP por sistema. Canarias. Espan n.d. https://www.esios.ree.es/es/analisis/573?vis¼1&start_date¼01-01-2010 T00%3A00&end_date¼31-12-2017T23%3A50&compare_start_date¼01-012009T00%3A00&groupby¼year&compare_indicators¼574&geoids¼8742 (Accessed 20 November 2017). [30] (CNMC) CN del M y la C. Liquidaciones Provisionales de las Actividades Regctrico. n.d. uladas del Sector Ele [31] Wiser R, Jenni K, Seel J, Baker E, Hand M, Lantz E, et al. Expert elicitation survey on future wind energy costs. Nat Energy 2016;1:16135. https:// doi.org/10.1038/nenergy.2016.135. [32] Wiser RH, Jenni K, Seel J, Baker E, Hand MM, Lantz E, et al. Forecasting wind energy costs and cost drivers: the views of the World’s leading Experts | electricity markets and policy group. LBNL-1005717. 2016 (Accessed 12 December 2017), https://emp.lbl.gov/publications/forecasting-wind-energycosts-and. [33] Hirth L, Ueckerdt F, Edenhofer O. Integration costs revisited e an economic framework for wind and solar variability. Renew Energy 2015;74:925e39. https://doi.org/10.1016/J.RENENE.2014.08.065. [34] 4C Offshore. Submarine Cable Consultancy. n.d (Accessed 20 November 2017), http://www.4coffshore.com/. [35] WindFloat Atlantic (WFA) Phase 2-4C Offshore n.d. http://www.4coffshore. com/windfarms/windfloat-atlantic-(wfa)-phase-2-portugal-pt04.html (accessed December 12, 2017). [36] Fukushima floating OffshoRe wind FARm demonstration project (Forward) phase 1-4C offshore n.d. http://www.4coffshore.com/windfarms/fukushimafloating-offshore-wind-farm-demonstration-project-(forward)-phase-1japan-jp06.html (accessed December 12, 2017).