Marine Environmental Research xxx (2014) 1e9
Contents lists available at ScienceDirect
Marine Environmental Research journal homepage: www.elsevier.com/locate/marenvrev
Spatial and temporal benthic species assemblage responses with a deployed marine tidal energy device: A small scaled study Melanie Broadhurst*, C. David L. Orme 1 N2.3. Munro Building, Division of Biology, Department of Life Sciences, Imperial College London, Silwood Park Campus, Buckhurst Road, Berkshire SL5 7PY, UK
a r t i c l e i n f o
a b s t r a c t
Article history: Received 31 December 2013 Received in revised form 9 March 2014 Accepted 17 March 2014
The addition of man-made structures to the marine environment is known to increase the physical complexity of the seafloor, which can influence benthic species community patterns and habitat structure. However, knowledge of how deployed tidal energy device structures influence benthic communities is currently lacking. Here we examined species biodiversity, composition and habitat type surrounding a tidal energy device within the European Marine Energy Centre test site, Orkney. Commercial fishing and towed video camera techniques were used over three temporal periods, from 2009 to 2010. Our results showed increased species biodiversity and compositional differences within the device site, compared to a control site. Both sites largely comprised of crustacean species, omnivore or predatory feeding regimes and marine tide-swept EUNIS habitat types, which varied over the time. We conclude that the device could act as a localised artificial reef structure, but that further in-depth investigations are required. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Benthic ecology Community composition EUNIS habitat classification Fishing gear Towed benthic camera Marine tidal energy device offshore structure Anthropogenic ecological interactions
1. Introduction The benthic zone environment supports a variety of marine species, communities and habitats, which are influenced by a complex and dynamic interplay of chemical, physical and biological factors (Dounas et al., 2007; Ierodiaconou et al., 2011). Past studies indicate that the physical complexity of the seafloor can create a diversity of benthic habitats, exploited by a range of organisms (Ierodiaconou et al., 2011). The addition of new, permanent manmade structures to the seafloor can increase physical complexity and habitat diversity, with such structures acting as ‘artificial reefs’. These can provide additional ecological niches, leading to new trophic interactions, increased species biomass, abundance and recruitment opportunities (Langhamer, 2010). Consequently, man-made structures can influence existing benthic ecological communities and habitats (Inger et al., 2009; Page et al., 1999). This has been shown by a number of ecological and environmental impact investigations of the deployment and
* Corresponding author. Tel.: þ44 (0)7781453976; fax: þ44 (0)20 759 42339. E-mail addresses: [email protected] (M. Broadhurst), [email protected] (C.D.L. Orme). 1 Tel.: þ44 (0)20 7594 2352.
permanency of man-made structures within the marine environment. Research includes offshore oil and gas platforms, coastal defences and more recently, wind and wave renewable energy developments (Krone et al., 2013; Langhamer and Wilhelmsson, 2009; Langhamer et al., 2009; Munari, 2013; Page et al., 1999; Petersen and Malm, 2006; Vaselli et al., 2008; Wehkamp and Fischer, 2013; Wilhelmsson and Malm, 2008). Collectively, these studies identify a range of associated ecological and environmental effects from the deployment and operational life cycles of these structures: from changes in species biodiversity or assemblage composition to alteration of the hydrodynamic regimes present. For example, Krone et al. (2013) state that the development of offshore wind farms can provide hard substrate for biofouling species, such as the blue mussel, Mytilus edulis. Despite these fundamental studies, knowledge of how marine tidal energy devices influence the marine benthic environment is particularly limited. In general, these devices are fixed to the seafloor and comprise of moving parts designed to exploit tidal currents for electrical energy generation purposes. Throughout the UK and Europe, future marine tidal energy development plans range from multiple to hundreds of these devices permanently placed within one site. As such, their overall design, long-term presence and role within the marine environment will differ to other offshore structures and technology.
http://dx.doi.org/10.1016/j.marenvres.2014.03.012 0141-1136/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Broadhurst, M., Orme, C.D.L., Spatial and temporal benthic species assemblage responses with a deployed marine tidal energy device: A small scaled study, Marine Environmental Research (2014), http://dx.doi.org/10.1016/j.marenvres.2014.03.012
2
M. Broadhurst, C.D.L. Orme / Marine Environmental Research xxx (2014) 1e9
Primarily, ecological baseline and environmental impact evidence is lacking due to the tidal energy industry being at an early research and design phase (Chen and Lam, 2014; Gill, 2005; Shields et al., 2009, 2011). There are also few accessible sites where devices are permanently deployed or are to be deployed to undertake in-depth ecological research studies, such as BACI experiments. In addition, device deployment activities by developers are sporadic due to these sites situated within marine environments which exhibit extreme tidal hydrodynamics and weather conditions. These factors have limited ecological research design, survey planning and constrained sampling effort to narrow time windows around slack water. As such, few quantitative studies have been undertaken in the past. Our aim was to begin to examine the influence of a permanently deployed, operational marine tidal energy test device upon local (mobile/non-mobile species) benthic species biodiversity, community composition, trophic functionality and habitat type. We used commercial fisheries potting equipment and benthic video photography techniques to sample within these extreme environments (Dunnington et al., 2005; Ierodiaconou et al., 2011; van Rein et al., 2011). We also began to explore the device’s influence across different fine spatial ( 0.05) were removed from the overall models to produce minimum adequate models (MAMs; Crawley, 2007). Models were assessed visually for unequal variance and non-normality of residuals by plots, and, where this occurred, the response variable was log transformed as this reduced rightskewness and stabilised variances (Clarke and Warwick, 2001). To determine species, taxonomic and feeding regime assemblage compositions, we first fourth-root transformed the datasets. This transformation was selected to down weight the contributions of the common and rare species, to view these datasets in more depth (Clarke and Warwick, 2001). We used analysis of similarities (ANOSIM in Primer v6) to identify significant species and assemblages compositional differences between the strings, survey sites and seasonal sampling periods (Langhamer et al., 2009). This routine was based on BrayeCurtis similarity resemblance matrix distances, which compares the ranks of the distance values within and between samples, grouped by a factor (Clarke and Warwick, 2001; Legendre and Legendre, 1998). The analysis produces an R value, which ranges from 1 to 1 and uses random permutations of the grouping factor to assess significance. Separate one-way ANOSIM routines were first run ‘within’ each survey site, with the string number selected as an exploratory factor. We then ran two-way analyses, with the survey site and seasonal sampling period selected as explanatory factors. We used the Primer v6 SIMPER procedure to explore which species and assemblages accounted for significant differences outlined by the ANOSIM results. This routine calculates a percentage contribution of each species within a pair of groups to the BrayeCurtis distances between those groups (Clarke and Gorley, 2006). The contribution similarity percentage for each species sample was based on 90% of the total percentage contribution, to omit any rare species or small contributions (Clarke and Gorley, 2006). For the assemblage analyses, we selected 100% of the
total percentage contribution, due to the small number of variable categories. 2.4.2. Benthic habitat assessment To examine the benthic habitat types within the two survey sites, we determined the mean percentage cover (%) and standard deviation of each physical and ecological category for each string (MESH, 2008; van Rein et al., 2011). We used separate one-way ANOSIM analyses to compare the composition of these combined categories between the strings, and survey sites. A hierarchical cluster analysis was also used; to examine the overall composition of these combined categories across all strings. This procedure was generated using the Primer v6 CLUSTER routine to produce a dendrogram, with group average linkage selected. The ANOSIM and CLUSTER routines were based on square-root transformations, with the combined dataset normalised prior to analysis, and the Euclidean distance to generate the distance matrix. These were selected as they are considered appropriate for the analysis of combined environmental datasets (Legendre and Legendre, 1998; Clarke and Warwick, 2001). We then selected benthic habitat descriptions for each string, using a combination of the video camera survey photography stills, raw video footage, bathymetry maps and water depth information (Connor et al., 2004; van Rein et al., 2011). Habitat classifications followed the EUNIS hierarchical habitat structure classification scheme (Brown and Collier, 2008; Connor et al., 2004; Davies et al., 2004; EEA, 2004). 3. Results 3.1. Benthic assemblage assessment: species biodiversity A total number of 10,889 individuals and 14 species were recorded during the 2009e2010 study, which comprised of 6 Crustacea, 3 Echinodermata, 4 Chordata and 1 Mollusca species. Within each survey site, we observed that species richness (S) was not significantly different. We observed that the device site comprised of significantly larger diversity, compared to the control site (mean S: device ¼ 7.22; control ¼ 3.33; Table 2). Across the seasonal sampling periods, we identified significant differences, with increased species richness in April 2010. The device site showed increased species richness over time, whilst the control site comprised of reduced diversity in October 2010 (Fig. 3). We observed similar ShannoneWiener diversity (H0 ) patterns, with no significant differences within each survey site, and significant differences between sites (mean H0 : device ¼ 1.80; control ¼ 1.22) and seasonal sampling periods. Pielou’s evenness (J0 ) diversity was not significantly different within each site or between survey sites (mean J’: device ¼ 0.920; control ¼ 0.934) and seasonal sampling periods.
Table 2 Generalised linear minimal adequate model (MAM) results for Species Richness (S) A), ShannoneWiener (H0 ) B), and Pielou’s evenness (J0 ) C) diversity measures recorded during the benthic assemblage assessment, 2009e2010, Isle of Eday. MAM model
Explanatory variable
A) Species richness (S)
Intercept Site Season Intercept Site Season
B) ShannoneWiener (H0 )
0
C) Pielou’s evenness (J )
F value
P
DF
24.71 7.56
Contents lists available at ScienceDirect
Marine Environmental Research journal homepage: www.elsevier.com/locate/marenvrev
Spatial and temporal benthic species assemblage responses with a deployed marine tidal energy device: A small scaled study Melanie Broadhurst*, C. David L. Orme 1 N2.3. Munro Building, Division of Biology, Department of Life Sciences, Imperial College London, Silwood Park Campus, Buckhurst Road, Berkshire SL5 7PY, UK
a r t i c l e i n f o
a b s t r a c t
Article history: Received 31 December 2013 Received in revised form 9 March 2014 Accepted 17 March 2014
The addition of man-made structures to the marine environment is known to increase the physical complexity of the seafloor, which can influence benthic species community patterns and habitat structure. However, knowledge of how deployed tidal energy device structures influence benthic communities is currently lacking. Here we examined species biodiversity, composition and habitat type surrounding a tidal energy device within the European Marine Energy Centre test site, Orkney. Commercial fishing and towed video camera techniques were used over three temporal periods, from 2009 to 2010. Our results showed increased species biodiversity and compositional differences within the device site, compared to a control site. Both sites largely comprised of crustacean species, omnivore or predatory feeding regimes and marine tide-swept EUNIS habitat types, which varied over the time. We conclude that the device could act as a localised artificial reef structure, but that further in-depth investigations are required. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Benthic ecology Community composition EUNIS habitat classification Fishing gear Towed benthic camera Marine tidal energy device offshore structure Anthropogenic ecological interactions
1. Introduction The benthic zone environment supports a variety of marine species, communities and habitats, which are influenced by a complex and dynamic interplay of chemical, physical and biological factors (Dounas et al., 2007; Ierodiaconou et al., 2011). Past studies indicate that the physical complexity of the seafloor can create a diversity of benthic habitats, exploited by a range of organisms (Ierodiaconou et al., 2011). The addition of new, permanent manmade structures to the seafloor can increase physical complexity and habitat diversity, with such structures acting as ‘artificial reefs’. These can provide additional ecological niches, leading to new trophic interactions, increased species biomass, abundance and recruitment opportunities (Langhamer, 2010). Consequently, man-made structures can influence existing benthic ecological communities and habitats (Inger et al., 2009; Page et al., 1999). This has been shown by a number of ecological and environmental impact investigations of the deployment and
* Corresponding author. Tel.: þ44 (0)7781453976; fax: þ44 (0)20 759 42339. E-mail addresses: [email protected] (M. Broadhurst), [email protected] (C.D.L. Orme). 1 Tel.: þ44 (0)20 7594 2352.
permanency of man-made structures within the marine environment. Research includes offshore oil and gas platforms, coastal defences and more recently, wind and wave renewable energy developments (Krone et al., 2013; Langhamer and Wilhelmsson, 2009; Langhamer et al., 2009; Munari, 2013; Page et al., 1999; Petersen and Malm, 2006; Vaselli et al., 2008; Wehkamp and Fischer, 2013; Wilhelmsson and Malm, 2008). Collectively, these studies identify a range of associated ecological and environmental effects from the deployment and operational life cycles of these structures: from changes in species biodiversity or assemblage composition to alteration of the hydrodynamic regimes present. For example, Krone et al. (2013) state that the development of offshore wind farms can provide hard substrate for biofouling species, such as the blue mussel, Mytilus edulis. Despite these fundamental studies, knowledge of how marine tidal energy devices influence the marine benthic environment is particularly limited. In general, these devices are fixed to the seafloor and comprise of moving parts designed to exploit tidal currents for electrical energy generation purposes. Throughout the UK and Europe, future marine tidal energy development plans range from multiple to hundreds of these devices permanently placed within one site. As such, their overall design, long-term presence and role within the marine environment will differ to other offshore structures and technology.
http://dx.doi.org/10.1016/j.marenvres.2014.03.012 0141-1136/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Broadhurst, M., Orme, C.D.L., Spatial and temporal benthic species assemblage responses with a deployed marine tidal energy device: A small scaled study, Marine Environmental Research (2014), http://dx.doi.org/10.1016/j.marenvres.2014.03.012
2
M. Broadhurst, C.D.L. Orme / Marine Environmental Research xxx (2014) 1e9
Primarily, ecological baseline and environmental impact evidence is lacking due to the tidal energy industry being at an early research and design phase (Chen and Lam, 2014; Gill, 2005; Shields et al., 2009, 2011). There are also few accessible sites where devices are permanently deployed or are to be deployed to undertake in-depth ecological research studies, such as BACI experiments. In addition, device deployment activities by developers are sporadic due to these sites situated within marine environments which exhibit extreme tidal hydrodynamics and weather conditions. These factors have limited ecological research design, survey planning and constrained sampling effort to narrow time windows around slack water. As such, few quantitative studies have been undertaken in the past. Our aim was to begin to examine the influence of a permanently deployed, operational marine tidal energy test device upon local (mobile/non-mobile species) benthic species biodiversity, community composition, trophic functionality and habitat type. We used commercial fisheries potting equipment and benthic video photography techniques to sample within these extreme environments (Dunnington et al., 2005; Ierodiaconou et al., 2011; van Rein et al., 2011). We also began to explore the device’s influence across different fine spatial ( 0.05) were removed from the overall models to produce minimum adequate models (MAMs; Crawley, 2007). Models were assessed visually for unequal variance and non-normality of residuals by plots, and, where this occurred, the response variable was log transformed as this reduced rightskewness and stabilised variances (Clarke and Warwick, 2001). To determine species, taxonomic and feeding regime assemblage compositions, we first fourth-root transformed the datasets. This transformation was selected to down weight the contributions of the common and rare species, to view these datasets in more depth (Clarke and Warwick, 2001). We used analysis of similarities (ANOSIM in Primer v6) to identify significant species and assemblages compositional differences between the strings, survey sites and seasonal sampling periods (Langhamer et al., 2009). This routine was based on BrayeCurtis similarity resemblance matrix distances, which compares the ranks of the distance values within and between samples, grouped by a factor (Clarke and Warwick, 2001; Legendre and Legendre, 1998). The analysis produces an R value, which ranges from 1 to 1 and uses random permutations of the grouping factor to assess significance. Separate one-way ANOSIM routines were first run ‘within’ each survey site, with the string number selected as an exploratory factor. We then ran two-way analyses, with the survey site and seasonal sampling period selected as explanatory factors. We used the Primer v6 SIMPER procedure to explore which species and assemblages accounted for significant differences outlined by the ANOSIM results. This routine calculates a percentage contribution of each species within a pair of groups to the BrayeCurtis distances between those groups (Clarke and Gorley, 2006). The contribution similarity percentage for each species sample was based on 90% of the total percentage contribution, to omit any rare species or small contributions (Clarke and Gorley, 2006). For the assemblage analyses, we selected 100% of the
total percentage contribution, due to the small number of variable categories. 2.4.2. Benthic habitat assessment To examine the benthic habitat types within the two survey sites, we determined the mean percentage cover (%) and standard deviation of each physical and ecological category for each string (MESH, 2008; van Rein et al., 2011). We used separate one-way ANOSIM analyses to compare the composition of these combined categories between the strings, and survey sites. A hierarchical cluster analysis was also used; to examine the overall composition of these combined categories across all strings. This procedure was generated using the Primer v6 CLUSTER routine to produce a dendrogram, with group average linkage selected. The ANOSIM and CLUSTER routines were based on square-root transformations, with the combined dataset normalised prior to analysis, and the Euclidean distance to generate the distance matrix. These were selected as they are considered appropriate for the analysis of combined environmental datasets (Legendre and Legendre, 1998; Clarke and Warwick, 2001). We then selected benthic habitat descriptions for each string, using a combination of the video camera survey photography stills, raw video footage, bathymetry maps and water depth information (Connor et al., 2004; van Rein et al., 2011). Habitat classifications followed the EUNIS hierarchical habitat structure classification scheme (Brown and Collier, 2008; Connor et al., 2004; Davies et al., 2004; EEA, 2004). 3. Results 3.1. Benthic assemblage assessment: species biodiversity A total number of 10,889 individuals and 14 species were recorded during the 2009e2010 study, which comprised of 6 Crustacea, 3 Echinodermata, 4 Chordata and 1 Mollusca species. Within each survey site, we observed that species richness (S) was not significantly different. We observed that the device site comprised of significantly larger diversity, compared to the control site (mean S: device ¼ 7.22; control ¼ 3.33; Table 2). Across the seasonal sampling periods, we identified significant differences, with increased species richness in April 2010. The device site showed increased species richness over time, whilst the control site comprised of reduced diversity in October 2010 (Fig. 3). We observed similar ShannoneWiener diversity (H0 ) patterns, with no significant differences within each survey site, and significant differences between sites (mean H0 : device ¼ 1.80; control ¼ 1.22) and seasonal sampling periods. Pielou’s evenness (J0 ) diversity was not significantly different within each site or between survey sites (mean J’: device ¼ 0.920; control ¼ 0.934) and seasonal sampling periods.
Table 2 Generalised linear minimal adequate model (MAM) results for Species Richness (S) A), ShannoneWiener (H0 ) B), and Pielou’s evenness (J0 ) C) diversity measures recorded during the benthic assemblage assessment, 2009e2010, Isle of Eday. MAM model
Explanatory variable
A) Species richness (S)
Intercept Site Season Intercept Site Season
B) ShannoneWiener (H0 )
0
C) Pielou’s evenness (J )
F value
P
DF
24.71 7.56
