Wave power devices as artificial reefs Olivia Langhamer*, Dan Wilhelmsson** Swedish Centre for Renewable Electric Energy Conversion, Dept. of Engineering Sciences, Uppsala University, Box 534, S-751 21 Uppsala, Sweden Email:
[email protected] *Also: Dept. of Animal Ecology, UU, Norbyvägen 18D, S-752 36, Uppsala Sweden ** Dept. of Zoology, Stockholm University, S-106 91 Stockholm, Sweden
Abstract Offshore energy installations, such as wave power parks have large area demands. Little is known about their effects on the marine environment, and early studies may minimize environmental risks and enhance potential positive effects on the marine environment. The Islandsberg project is a wave power park located on the Swedish west coast under construction since 2005. Here, buoys acting as point absorbers are connected to generators anchored on concrete foundations on the seabed. Most likely, biofouling has the potential to be an engineering concern but at the same time, these structures can be considered as artificial reefs. In the present study the colonisation of foundations by invertebrates and fish was investigated. The influences of holes in and on the surface of the foundations on colonisation and species composition were examined and a succession in colonisation over time is shown. Structural heterogeneity may enhance the colonisation of macrobenthic organisms. This pilot study is the first contribution to what constitutes an important research topic based on different ideas of an effective and purposeful reef design integrating wave power devices and their development. Keywords: artificial reef, biodiversity, crabs, fish, macrobenthos, renewable energy, wave power
Introduction Energy demands are rising worldwide and a significant expansion on new large-scale renewable energy generation sources is expected (Shaw et al. 2002). There is a high potential for offshore renewables and the technical development is leading towards increased effectiveness and acceptance. However, concerns are raised regarding environmental impacts in coastal waters of offshore energy installations, including for example noise, electromagnetic fields, habitat alterations, and changed hydrological conditions (Gill 2005, Petersen & Malm 2006). Thus, in addition to effective energy generation, there is a need to include environmental considerations during the planning, construction and operation phases of offshore energy development. Solid structures placed on the seabed to support, or as a part of, the offshore energy units should be regarded as artificial reefs (ARs). Construction and deployment of ARs, often specially designed, is a concept used worldwide for fisheries management/enhancement,
and coastal protection (Pickering et al. 1998, Jensen 2002). In addition, ARs are often used for preservation and rehabilitation of marine habitats (Pickering et al. 1998, Jensen 2002). Many man-made submerged structures do not have a primary function as ARs but will inevitably be colonised by organisms. These structures may attract marine organisms, either through adult migration from neighbouring natural reefs or by planktonic larvae settling on the added hard substrate. They may increase local biodiversity, species abundance and biomass (Johnson et al. 1994; Lozano-Alvarez et al. 1994; Gosling & Sutherland 2004). Whether the local biomass enhancement is a result of mere aggregation from the surroundings or a true increase in biomass vary with species and a number environmental factors and, in most cases, site specific input data only allows for general assumptions (Bohnsack et al. 1991, Baine 2001). Still, there are conflicting results whether the local increase in biomass is a result of mere aggregation from the surroundings or a true increase in biomass (Bohnsack 1989; Bohnsack et al. 1997). Bohnsack`s “production hypothesis” (1989) states that an added production by ARs implies enhancements of either food availability, feeding efficiency, shelter from predation or enhanced conditions for larval settlement. Most man-made constructions in the marine environment consist of materials such as treated wood, metal, glass, tires, rigid plastic, concrete, or fibreglass and do not attempt to mimic nature. How well artificial reefs do mimic natural habitats, particularly in their role in maintaining natural ecological assemblages, is a cause for concern and discussions in a world of increasing urbanization (Savard et al. 2000). ARs may host communities different to those on natural reefs, and may also alter the biological diversity in nearby areas (Connell & Glasby 1999; Connell 2001). When hard substrate is added to former soft bottom great changes can occur due to the introduction of the new kind of habitat that can lead to new trophic opportunities and changes on local food web interactions (Bohnsack & Sutherland 1985). Finally, it is important to determine if the new structures are beneficial to existing local species or facilitate the establishment and spread of non-native species, as noted on other urban structures, such as oil rigs and breakwaters (e.g. Bulleri & Airoldi 2005, Page et al. 2006). The extension of colonisation is dependant on a great number of parameters, e.g. abiotics (temperature, salinity, hydrodynamics, solar radiation, heterogeneity of substrate, water depth) and biotics of the surrounding marine environment (predation, competition, parasitism, larval availability, and larval behaviour). Sessile organisms on the introduced hard substrate may serve to
1
attract fish and crustaceans by providing additional shelter and food (Johnson et al. 1994). The design of a reef to attract specific organisms, such as economically valuable species, is of great interest among coastal managers and entrepreneurs. Studies have shown an increase in species abundance with increasing structural volumes and complexities of artificial reefs (Potts & Hulbert 1994, Spieler et al. 2001). Further, more complex substrata may provide a broader selection of different micro-habitats, which may allow colonisation by higher species numbers (Menge 1976). Wave and tidal power is still on the developing stage with only some projects having reached demonstration status, and a few closing in on commercialisation. Moreover, there are several techniques employed today, most devices including structures on the surface (Oxley 2006). In Sweden, a test park for wave power has been built up from 2005 at the west coast north of Gothenburg. One wave power unit consists of a point absorber buoy on the surface driving a piston in the generator (Gustafsson et al. 2005). The linear generator is placed on the seafloor on a concrete foundation directly generating electric energy. Installation of such wave power parks requires a substantial area. For instance, a 10 MW installation of point absorbers (1000 ex. 10 kW/units, which is the unit size for calmer waters) claims an area of approx. one km2. Fishing will be limited or prohibited in wave- and tidal park areas. The parks are hence comparable to marine protected areas, where enhancements in local fish species richness, size and biomass has been shown (Halpern 2003, Sundberg & Langhamer 2005). Little is known about the influence of offshore energy installations on the development and distribution of macrobenthos (Sundberg & Langhamer 2005). This limits the possibilities to develop models of predicted impacts, and to identify options for design and management of different structures. In addition there are only a few examples of artificial reef studies at these depths in cold temperate waters (Baine 2001, Jensen 2002). The present works is a first attempt to study the macrobenthic communities on secondary ARs, constituting the foundations for wave energy devices, which is an energy conversion likely to grow (Henfridsson et al. 2007). In particular, we aimed to analyse the succession of colonisation of macrobenthic communities on wave power foundations provide indications on the influence on colonisation of surface complexity and inclination.
Materials and Methods Study area The test park Islandsberg is situated at the Swedish west coast in an archipelago about 100km north of Gothenburg, and about 2km offshore. The test park is situated between a northern marker (58º 11' 850N 11º 22' 460 E) and a southern marker (58º 11' 630N 11º 22' 460E). Fully built out it will contain approximately 40 buoys: ten with generators and an additional 30 for environmental studies, it will cover about 40.000m2 (Sundberg & Langhamer 2005). The first experimental setup consisted of 5 wave power devices without generators for biological studies i.e. cylindrical buoys
with a diameter of 1.8m and a height of ca. 0.8m, moored with a low-tensile line to 10t concrete foundations with a diameter of 2.5m each. There was one additional bigger buoy with 3m in diameter moored to a concrete foundation with a weight of 40t and an area of 16m2. The dimension of the foundations was adapted for extreme wave conditions and lifting capacities of buoys. The devices were put in place during spring 2005 with a distance between 100m and 300m from each other, all on a 25m deep seabed consisting of firm shell gravel with some clay. A natural reef is situated close to the southern tip of the test park. Experimental design Three of the five foundations were designed with holes: FI, FII and FIV. Two of the foundations; FIII and FV were designed without holes (Tab.1). Field sampling was carried out between the 9th th and 11 August 2005 and between the 18th and the 20th of July 2006 by scuba diving. The big foundation was only investigated in 2006. A total of 124 macrofauna samples were taken by photographing 7.0cm * 10.6cm areas of the horizontal and vertical surfaces of the foundations. Sessile organisms were identified down to family or species if possible. All organisms that could not be identified to species or genus were classified to larger taxonomic groups. Foundation FI
FII FIII FIV FV
Location West
Northeast Northwest Southeast Southwest
Size 40t (4*4*1m) 10t (3m *1m) 10t (3m *1m) 10t (3m*1m) 10t (3m *1m)
Sample Holes size 25 32*7cm, 16*15 cm, 8*30 cm 27 10*15cm 26
0
24
10*10cm
22
0
Table 1. Experimental design of the five foundations; FI, FII, FIII, FIV, FV, location, foundation size, sample size, and holes. The epibenthic coverage of each species on the foundations was calculated in Image Tool 3™ and statistical analyses were performed in Statistica™ v6 and Primer™ v6 (Clark & Gorley 2006). The data were assessed for normality, forth-root transformed to produce normally distributed data for multivariate analyses, and were used in the transformed state throughout. Similarities in species epibenthic composition between different foundations with and without holes, between horizontal and vertical surfaces and between years were calculated in ANOSIM from Bray-Curtis similarity matrices based on transformed abundance data. The Rvalue shows how big the differences between different treatments are and should be above 0.2 (Clark & Gorley 2006). Post hoc multiple comparison analyses were used to determine significant pairs of means when more than two groups were analysed.. The data were visually assessed with non-metric multidimensional scaling (nMDS) ordinations constructed from the Bray-Curtis matrices. The stress-value indicates how reliable the different positions in the MDS plot are and should be below 0.2 to show reliable positions in the graph.
2
Similarity of percentage (SIMPER) on the transformed data were performed when the ANOSIM global R value was significant in order to identify taxa that contributed most to the dissimilarity between, and similarity within treatments and years. Total number of species (S), species richness (d) and Shannon Wiener diversity (H´(log10); Krebs 1989) of macrobenthos were tested to see differences between reef types over the years. In 2006, fish, crabs and lobsters associated with the five foundations were recorded through visual censuses on the structures, including the holes where present, and on the surrounding bottom within 1m distance. For three of the foundations (FII, FIII, FV), similarly sized controls were sampled on bare sand bottoms at 10m distance from foundations. The censuses were conducted before the samples of epibiota were taken to minimize disturbance by divers. Furthermore, on FI, a detailed survey on inhabitants of individual holes with different diameters was conducted. Two species of gobies (Pomatoschistus spp.) are common on the Swedish west coast; sand goby (P. minutus) and common goby (P. microps). It is difficult to distinguish between them in visual surveys. At the depths of 25m, most of the goby species are likely to be P. minutus (Jansson et al. 1985, Thorman and Widerholm 1986). Flatfishes were recorded as Pleuronectidae, and the two sculpins, sea scorpion (Taurulus bubalis) and the bullrout (Myoxocephalus scorpius), were recorded as cottids. In 2005, in terms of mobile fauna, only crabs were recorded on the four smaller foundations. Comparisons of fish abundance, species number (Wilcoxon´s Matched Pairs Test) and assemblage structures (ANOSIM- Analysis of similarities) were made between the foundations (FII, FIII, FV) and the bare sandy bottom. Shoals of juvenile fish were excluded in the data analyses to avoid skewness of data. Descriptive data was prepared for all foundations, including average abundance and species number of fish and shellfish. Further, for comparisons of occupancy of holes of different sizes on the large foundation, averages and frequency of occupants were calculated in percent.
Results A total of 28 species were identified in 124 photographic samples. The most frequent epibenthic organisms colonising the wave power foundations were sea squirts (Ascidiaceae), hydroids, serpulid worms (Pomatoceros triqueter) and barnacles (Balanus sp.). In the first year of sampling (2005) the foundations were only colonised by a few species, mostly calcareous serpulid worms and barnacles. The total coverage degree of macrobenthos in 2005 was 7.8% and increased to 76.9% in 2006, and community structure differed significantly between years (R= 0.6; p= 0.001). The nMDS ordination illustrates these relationships with the 2005 versus 2006 samples showing minimal overlap (Fig.1). SIMPER analysis described the communities as being on average 47.1% similar within and 82% dissimilar between years. Taxa that contributed to the dissimilarity between 2005 and 2006 were primarily hydroids, barnacles, ascidians and serpulids. Eight species contributed to 90% of the dissimilarities (Tab.2). Two of these taxa, hydroids and barnacles, accounted for 40% of the dissimilarity.
Taxa Hydroids Balanus sp. Corella parallelogramma Pomatoceros triqueter Ascidiella aspersa/scabra red algae Ascidiacea Asterias rubens
Average Average Abundance Abundance 2005 2006 0,0 19,0 17,6 1,5
% Contribution 21,8 18,4
0,0
16,7
14,0
10,6
15,2
12,1
0,0 3,7 2,1 1,7
11,4 3,1 2,1 0,9
8,9 7,1 3,5 2,4
Table 2. Taxa contribution to 90% of dissimilarities in the relative abundance of taxa between macrofauna coverage on wave power foundations sampled 2005 and 2006. Species numbers and species richness increased with time and we also found a higher diversity on foundations in 2006 than in 2005 (Fig.3). Cover of epibiota was significantly higher on vertical than on horizontal surfaces in both 2005 and 2006 (R=0.3; p=0.002 and R=0.8; p=0.001 respectively). The nMDS plot illustrated this relationship of horizontal and vertical surfaces showing a clear clustering (Fig. 2).
Figure 1. nMDS plot of macrobenthos communities sampled on wave power foundations in 2005 (blue) and 2006 (green), based on BrayCurtis similarity of forth root transformed species coverage.
Figure 2. nMDS plot of macrobenthos communities sampled on horizontal (green) and vertical (blue) surfaces, on wave power foundations, based on Bray-Curtis similarity of forth root transformed species coverage.
3
Comparing sides, more species and higher diversity occurred at the vertical side than on the horizontal side (Fig.4). The species richness showed only a trend towards a higher richness on the vertical side but there was no significant difference (Fig. 4). There were no differences in colonisation among foundations (2005: R=0.093; p=0.04 2006: R=0.07; p=0.008), nor between the foundations with holes compared to foundation without holes (2005: R=-0.02; p=0.07, 2006: R=-0.001; p=0.5). However, in 2006, when looking only at the vertical surface and comparing foundations with and without holes, they differed slightly (R=0.2; p=0.002), mainly attributed to more seas squirts on the foundations with holes. In the survey of fish and shellfish, a total of 18 fish of 6 species, 38 crabs (C. pagurus), and two lobsters (Homarus gammarus) were recorded in association with the five foundations. Typical hard bottom associated species such as lobsters and goldsinny wrasse were
b) p
