Spatial variability of wave energy resources around the Canary Islands

Coastal Processes 111 1 5 Spatial variability of wave energy resources around the Canary Islands H. Chiri, M. Pacheco & G. Rodríguez De partam ento de Fí sic a, Univers idad de las Palma s de Gran C anaria , Spain Abstract Spatial variability of wave energy resource around the coastal waters of the Canary Archipelago is assessed by using a long-term data set derived by means of hindcasting techniques. Results revea( the existence of large differences in the energetic content available in different zones of the archipelago, mainly during spring and autumn. Areas with a higher wave power leve( are the north edge of Lanzarote, western side of Lanzarote and Fuerteventura, north and northwest in La Palma and El Hierro, as well as the north coast of Tenerife. The available energy potential slightly decreases in the north side of Gran Canaria and La Gomera. In general, only north and west edges of the archipelago during winter and autumn show practica( interest for the use of existing wave energy conversion devices. Keywo rd s: wave ener gy, spatial variabi/i ty, wave hindcastin g, Cana1 y /stand s. 1 Introduction The search for renewable energy sources offering clean altematives to scanty, expensive, and environmentally problematic fossil fuels represents a vital challenge for humankind. The research in this field ha s been during a long time almost exclusively focused on the development of solar and wind sources. However, solar energy is stored in the sea as waves, currents, and heat. That i s, energy in the ocean is much more concentrated than in th e direct solar radiation and in the wind. Small islands and archipelagos in the world are almost totally dependent on fossil fuels to meet their energetic needs. Nevertheless, in general, islands have a unique potential for renewable energy (i.e ., Jensen [ 1 ]). That is , a competitive • WIT Transactions on Ecology and th e Environme nt , Vol 169, www .witpress.com, ISSN 1743-354 1 (on-line) doi: 10.2495/ CP I 30021 2013 WIT Press 1 6 Coastal Processes 111 economic situation for renewable energy technologies, good renewable energy resources, positive attitude towards renewable energy. Furthermore, there is a need to demonstrate feasibility of renewable energy in a large-scale, integrated and organised form. Therefore islands are very important and interesting for the promotion of renewable energy world-wide and almost all of the mature renewable energy technologies, mainly solar and wind, have been utilised for electricity production in these areas. Wind-generated gravity waves in the sea surface constitute the highest energy density of all renewable energy resources (i.e., Clement et al. [2]). This fact has led scientists and engineers to develop a wide variety of devices to hamess this resource by means of its conversion into electrical energy. An up-to-date review of the state-of-the-art in this challenging area is provided by Cruz (3] . Nowadays, this technology is not still at an advanced stage of development. However, it has attracted increasing interest of researchers during the past few decades as a potential energy resource to small islands. ln particular, since the pioneering work of Mollison (4], Canary archipelago, as well as Azores and Madeira, have been considered as an interesting zone to exploit wave energy as a feasible altemative to conventional energy resources. Thus, several authors have conducted studies to assess the potential wave energy available for sorne specific islands (i . e. , Rodríguez et al . [5], Iglesias and Carballo (6 , 7] , Sierra (8]). The main goal ofthe present study is to extend the assessment ofthe potential availability of wave energy to all the islands of the Canary archipelago, with special emphasis on the spatial variability ofthis resource. The paper is structured as follows. The study area and the data base used in the study are presented in section 2. General notions on the methodology to estimate the wave energy flux are brietly outlined in section 3. Next, in section 4, the discussion of results derived from the analysis of the data set is presented. Conclusions drawn from results are summarised in section 5. 2 Study area and data set The Canary !stands are an archipelago of seven major volcanic islands in the Atlantic Ocean, located about 100 Km offthe African coast (27.5° N - 29.5º N, 13º W-18 .5º W) (Fig. 1 ). Gran Canaria and Tenerife represent the two most populated islands, adding together more than 80% of total inhabitants. The rest of the residents are mainly concentrated in Fuerteventura, Lanzarote and La Palma, while the minor islands of El Hierro and La Gomera are barely populated. A quick look at Figure 1 reveals the complex geometry ofthe islands coast, the proximity between them, and to the African continent. Another important factor to consider is the altitude of the islands. The altitude in five of the seven islands is over l 500m, with 3700m in Tenerife, while Lanzarote and Fuerteventura are considerably tlat. The estimation of the wave energy potential around Canary lslands has been carried out by using time series of characteristic wave parameters obtained by means of a hindcasting approach. The W AM numerical model provides the WIT Transact io ns on Ecology and th e Envi ro nm e nt , Vol 169, www. wi tpress.com, ISSN 1 743 -354 1 (on-line) 2013 WIT Press Coastal Processes 111 1 7 29 . s~~--~--~--~--~~-~--~~-~--~---~__,o-_.___, ___ ~-~ 29 -8 28.5 :§ :'i 27 .5 ,_ .. - 19 Figure 1: -• -- ! --: -- ·t ·---··--·· -18 - 17 - 16 - 15 -1 4 -1J Long itu de (º) Map ofthe Canary lslands and location ofthe hindcast points. directional spectrum at each one of the points marked in Fig. 1. From these, characteristic parameters to evaluate wave power resource can be derived. The WAM model is a 3'd generation spectral wave model which solves the spectral wave action balance equation without any a priori assumption on the wave spectrum shape by using a finite-difference scheme. The wave field is described by the two-dimensional wave action density spectrum, N(w,{J), where w is the angular wave frequency and () is the wave direction. The model uses wave action density spectrum because action density is conserved in the presence of currents. The spectral action balance equation reads aN + a(CgxN) + a(CgyN) + a(CwN) + a(C9N) = ~ (1) at ax ay aw ae w The first terrn on the left-hand side represents the local rate of change of wave action density in time ; the second and third terrns stand for the propagation of wave action over geographical space, with propagation velocities C g. •· C KY> C ra C8 in the geographical and spectral space, respectively. The terrn at the right hand side is the source function, including the effects of generation, dissipation, nonlinear interactions, and bottom friction dissipation. Details about WAM can be found in Komen et al. [9]. The WAM model was used to provide the 44-year hindcast wave climate database ( 1958-2001) from the European HIPOCAS project (see Guedes Soares [ 1 O]). The model was forced by the output of a high-resolution atmospheric model (REMO regional atmospheric model). The HIPOCAS database was generated by running the models on a grid covering the North Atlantic with a resolution of 0.5° x 0.5° far from the coast and 0.25° x 0.25° close to this. Temporal resolution is 3 hours. Details on the methodology used to obtain wave conditions at any grid dot are given in Pilar et al. [ 11]. Figure 1 indicates the position ofthe 65 points considered in the study, indicated as red dots. 3 Wave energy flux estimation The wave energy flux, power density, per meter ofwave crest is given by r2rr Joo P = pg Jo 2 rr C 9 (f, h)S(f, 8)df d(} • WIT Transactions on Ecology and the Environment, Vol 169, www. wi tpress.com, ISSN 1743-354 1 (on-line) 2013 WIT Press (1) Coastal Proces se s 111 1 9 constitute a third kind of area, with the lowest values of wave power, not exceeding 12 kW/ rn. Yalues of average wave power for each clirnatic season are shown in Figs 36. During the spring (Fig. 3) the areas with higher energy power available are north and west strips of the archipelago, particularly the north edge of Lanzarote and La Palma islands. The general distribution follows the sarne pattem than that of the annual average values, with very low values of power in the south and southeast of the rnain part of the islands (3-9 kW/rn), and interrnediate values in the rernaining areas. Figure 4 illustrates the average wave power during surnrner. lt reveals a significant reduction of the wave energy potential in the entire archipelago. Areas with larger energy availability remain the same, but now values are lower than 10 kW/ rn. Energy levels corresponding to autumn are depicted in Fig. 5. During this season, energy levels are similar to those estimated for the annual period. However, power values in the areas with a higher exposure degree to wave fields approaching from the NNW directional sector are comparatively higher. In the north and west of La Palma, as well as in the north of Lanzarote wave power level reaches values between 24 and 27 kW/m. In contrast with the other periods of the year, during winter (Fig. 6) Canary lslands coasts receive a relatively height quantity of wave energy. In sheltered areas by shadowing effects, levels of energy follows been low, but in the other parts of the archipelago the available power experiences a considerable increase. This is rnaximum the northem and western edges ofthe archipelago, where wave power values increases between 75% and 80% with respect to the annual mean values. Zones with higher wave energy power are located norwest of La Palma, with a power leve( close to 40 kW/m, and to the north of Lanzarote, where wave energy power reaches values in the range 36-39 kW/ rn . Results derived from the study reveals an appreciable spatial variability in the wave energy availability in the archipelago. The reasons for a so large variability are diverse. At one hand, waves can reach the coasts of the islands frorn any direction. Nevertheless, for a given direction of approach, waves only irnpact on that islands exposed to wave fields corning frorn that sector and, of that islands, on ly the coasts oriented in the direction of wave travel. That is , sorne islands can act as a, partial or total, barrier against sorne wave conditions for other islands. Furthermore, any island shelters their own coasts located at the opposite side of wave approaching. These effects of self-blockage and rnutual-blockage depend to a large extent on the geornetry and dirnensions of each island. On the other hand, characteristics of wave fields traveling toward the islands depend substantially ofthe directional sector frorn which waves approach. 5 Concluding remarks Analysis of wave energy potential around Canary lslands reveals that there are large differences in the energetic content available in different zones of the iÍ WIT Transac ti ons on Ecology and the Environme nl , Vol 169, www.wilpress.com, ISSN 1 743 -35 41 (on-line) 2013 WIT Press • ~ ~ ~ =i §. ~ ~ ~ ""' "' o !'! =· " o o " 3 "' • o -" ~ gi zo -o -.JOO ... '< Y' "' w" "' Q. :: So ~" o m "" .!.. < ~ · a · ~::i 3 ~ ~ < 2. °' -'° !El "' o ~ =i ~ 2goN 28°N 27ºN 19"W 18"W '!} o u o '!ti o ., lll Palm3 ' ¡;;, · · · o o o o o (> ,. o ,., e '( ' - ~ u " e LA n Gom e ra o ~ { o o o (> El Hierro o (> o Figure 2: 17"W o o o (> Tener· ife o (> o o o o l C.'l na ri a 16"W o o "" V e o o o 15"W "'' " Annual spatial di st ri bu tion ofwave energy resource. 14"W o o o o e uerteverrtura k.Whn 1 3 - 6 6 - g o g - 12 o 12 - 15 o 15 - 18 o 18 - 21 o 21 - 24 l<l l omel!!rs o 15 ll !ll N J :~ fil N o (") o ~ ~ "'O o (") ('O "' "' ('O "' • ~ ~ ~ -i ~ ::;i : : Col ~ ~ ~ ~ g g· :; ~ • o Vi ~ "' o zo - o .....JO:: .... '< '-:'" w = "' o. ::: :;. -" o rn = = .!.. < ~ - a · ~ = ~ ~ < !:<. . :t:: '" o , _, :;: :::; [ 29"N 28ºN 27"N 19°W 18oW 17"W ,. , A V V o ~ o o o (> u Paln ~ . ., (> o (> o :JY 'i - ~ ~ ' : .. 1.· _: . o o o (> ~ •• · ., ~ . (> ' t' . .,. ,. · , ~ ., ·' '· o " o o -,, ,.,._ ... .. Gomera., Tenerife . º ~ · · / º º 'El Hierro o (> o o " o <> o 16"W o o o o o --~ :~ ~ ~ C' -~ o o o o o o 15"W " " -' o Figure 3: Spatial distribution of wave energy resource during spring. 14"W o o o o k\ :V im N 3 - 6 ~ \ 6 - 9 o 912 o 12 - 15 o 15 - 18 1 n o ~ o 18 - 21 "' §. "O l<ll ometl!rs ~1 .., o o 16 ll 8'.l () w w "' "' "' "' "' - N • ~ ~ ~ ::; . -1 ~ · ~ ~ g; "' o ~ =· 8 g 3 "' • o -" ~ gi zo - o ~22 t~ "' o. ::: :;. ~" o tT1 I. ~ :;· :;· " o ~o 3 " ·ª 2: "' · '° o N o ~ -1 l 29"N 28ºN 27°N 19'W o o La Paln' o ~ ~ ( IHi:..ro o 18"W 17"W o (> o (> o t.ft Gomera Tenetife Cana1i a 16"W 15"W {~ o 14"W o o o o uertevetltura k\ \' /m <3 3 - 6 6 - 9 o 912 f..1 Dmttl'U N J\ 015.xl 60 90 Figure 4: Spatial distribution ofwave energy resource during summe r. N N (") o ~ §. '"O a n o V> V> o V> .. ~ ~ ~ ::¡ ~ · ~ "'d:::: (11 g; ~ !l n o· o " .. 3 ~ v; ~ VI(') za - o ~r!.2 l¡J :,, w" V> Q. ~ =- ~" o m "" .!... < =· :::; · " o -" 3 " ·ª ~ _$ "' o w ~ ::¡ l 19"W 29°N ,,., o '-/ o ... -::i~ ~ - - · i ~ La Pa ln •1 < .;· , ; (> , o o '() o 0 4, o El H ierro o (> o 27°N 18"W 17"W 16"W 15"W r" ._, o o o o o o o º ir o o ~ ~ -,~;.. ~ i -r - ~ o o o <> ' ·,~ ""° (> o o o o ;,.;; 4)8p·_ -- ~~ • 'S ~.t J.~~ ~ ... ~ o t.;¡ ~) ' ... ,, ;' ~ ;:,~ . 'l?f!:. ~, t o o Go rn ern Te ner ffe o o (> o o ( > ~ ~ Gran o o o Canar ia o o Figure 5: Spatial distribution ofwave energy resource during autumn. o o o o r, o -- ~ :_ · : ~..., e 1 kW/m o 3-6 6-9 o 9 - 12 1 o 12 -15 o 15 -18 o 18 - 21 o 21 -24 o 24 -27 14 1omsers o 15 ll llO 14"W N j ~ QO (") o el [ '"'O a (') <> Vl Vl <> Vl N w