- Email: [email protected]
EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study
SEARES-01321; No of Pages 10 Journal of Sea Research xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Sea Research journal homepage: www.elsevier.com/locate/seares
EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study Pedro Monteiro ⁎, Luis Bentes, Frederico Oliveira, Carlos M.L. Afonso, Mafalda O. Rangel, Jorge M.S. Gonçalves Centre of Marine Sciences (CCMAR), Universidade do Algarve, Ed. 7, Campus de Gambelas, 8005-139 Faro, Portugal
a r t i c l e
i n f o
Article history: Received 31 January 2014 Received in revised form 11 August 2014 Accepted 28 November 2014 Available online xxxx Keywords: EUNIS classification Habitat mapping Thresholds Biological zonation Kinetic wave energy Seaweeds
a b s t r a c t The European Nature Information System (EUNIS) has been implemented for the establishment of a marine European habitats inventory. Its hierarchical classification is defined and relies on environmental variables which primarily constrain biological communities (e.g. substrate types, sea energy level, depth and light penetration). The EUNIS habitat classification scheme relies on thresholds (e.g. fraction of light and energy) which are based on expert judgment or on the empirical analysis of the above environmental data. The present paper proposes to establish and validate an appropriate threshold for energy classes (high, moderate and low) and for subtidal biological zonation (infralittoral and circalittoral) suitable for EUNIS habitat classification of the Western Iberian coast. Kinetic wave-induced energy and the fraction of photosynthetically available light exerted on the marine bottom were respectively assigned to the presence of kelp (Saccorhiza polyschides, Laminaria hyperborea and Laminaria ochroleuca) and seaweed species in general. Both data were statistically described, ordered from the largest to the smallest and percentile analyses were independently performed. The threshold between infralittoral and circalittoral was based on the first quartile while the ‘moderate energy’ class was established between the 12.5 and 87.5 percentiles. To avoid data dependence on sampling locations and assess the confidence interval a bootstrap technique was applied. According to this analysis, more than 75% of seaweeds are present at locations where more than 3.65% of the surface light reaches the sea bottom. The range of energy levels estimated using S. polyschides data, indicate that on the Iberian West coast the ‘moderate energy’ areas are between 0.00303 and 0.04385 N/m2 of wave-induced energy. The lack of agreement between different studies in different regions of Europe suggests the need for more standardization in the future. However, the obtained thresholds in the present study will be very useful in the near future to implement and establish the Iberian EUNIS habitats inventory. © 2014 Elsevier B.V. All rights reserved.
1. Introduction The effects of human activities on the marine environment are increasingly evident and impossible to be ignored. Industrial fishing and pollution as well as several other anthropogenic impacts have been causing damage to marine species and their habitats. An ecosystem-based management approach within marine spatial planning requires standardized classifications and terminology across all possible habitats (Costello, 2009; Hiscock and Connor, 1991). There have been progresses into the development of habitat classifications and different schemes have been implemented, such as in Europe (Connor et al., 1995, 1997a, 1997b, 2003, 2004; Davies et al., 2004; Devillers and Devillers-Terschuren, 1996; Devillers et al., 1991), Canada (Zacharias et al., 1998), Australia (Butler et al., 2001), United States (Allee et al., 2000), New Zealand (Snelder et al., 2006) and in South Africa (SANBI, 2009).
⁎ Corresponding author. Tel.: +351 289 800 900; fax: +351 289 800051. E-mail address: [email protected] (P. Monteiro).
The system by which European habitats have been classified is called the European Nature Information System (EUNIS) (Connor et al., 2003, 2004; Davies et al., 2004; Evans, 2012). The marine section of the EUNIS classification was originally based on the UK BioMar classification (Connor et al., 1995, 1997a, 1997b). It was developed in collaboration with experts from all over Europe and managed by the European Topic Centre for Nature Protection and Biodiversity (ETC/NPB) for the European Environment Agency (EEA) and the European Environmental Information Observation Network (EIONET) (Davies et al., 2004). This system covers all types of natural and artificial habitats and is theoretically under development as new information becomes available (Howell, 2010). A habitat is the area where an organism, or population or communities of species occur (e.g. mussel beds on circalittoral rock) i.e. the type of environment where a species could potentially establish itself (Olenin and Ducrotoy, 2006). For the purposes of EUNIS habitat classification the term is used as the place where a biological community normally lives, characterized by its physical features and by the communities or assemblages of species that live there (Davies et al., 2004). Thus it is defined by the combination of several key environmental variables
http://dx.doi.org/10.1016/j.seares.2014.11.007 1385-1101/© 2014 Elsevier B.V. All rights reserved.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
2
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
(e.g. substrate types, exposure to wave action, strength of tidal currents, salinity, temperature, hydrographic regime), referred to as the ‘habitat envelope’, together with its associated biological community, operating at particular spatial and temporal scales (ICES, 2001). Hierarchical segmentation within the upper marine division of the EUNIS classification systems is defined by a few environmental variables which primarily drive/structure the distribution of biological communities such as substrate types, energy level (wave action and tidal current), depth and light penetration (Connor et al., 1995, 1997a, 1997b, 2003, 2004; Davies et al., 2004). The EUNIS marine classification scheme is initially divided up on the basis of biological zones (e.g. infralittoral and circalittoral) and substrate types (e.g. sand, rock) as well as on the depth range (level 2). At level 3 the segmentation is broadly based on ecological region (Atlantic, Baltic, Mediterranean and Black Sea) and three energy levels of the coast (low, moderate and high), while associated biological communities begin to appear on lower levels (Davies et al., 2004). Referred to as a ‘top-down’ approach, this classification scheme relies on thresholds (e.g. biological zonation and sea energy) that can be based on the literature, expert judgment or on the empirical analysis of the above environmental data variables (Cameron and Askew, 2011). The biological zones are classically obtained through the characterization of the vertical distribution of certain indicator species, such as macrophytes and algae, or sessile fauna such as sponges and ascidians which are less tolerant to light (Logan et al., 1984; Pérès, 1967; Pérès and Picard, 1964). The presence of macrophytes and algae (sea grasses, kelp and other seaweeds) is directly correlated to the fraction of light reaching the seabed (Birkett et al., 1998). Studies have stated that kelp species and seagrass (e.g. Posidonia oceanica) require a minimum of 1% of surface light to be able to develop (Ballesta et al., 2000; McBreen and Askew, 2011; McBreen et al., 2011b). This threshold value has been tested for different European regions by several marine mapping projects (e.g. 2.36% for the North Sea and Celtic Sea in MESH and 4.5% for the North and Celtic Sea and 1% for the West Mediterranean Sea in EUSeaMap) (Cameron and Askew, 2011; Coltman et al., 2008). McBreen and Askew (2011) in the UKSeaMap 2010 defined the hard threshold to be the value above which 75% of kelp biotopes were classed as infralittoral and obtained 1%. On the other hand, the concept of what constitutes ‘low’ or ‘high energy’ on the sea bottom is far from being easily established
(Galparsoro et al., 2012). Energy can be characterized in a variety of ways that account for effects due to tidal currents and waves (McBreen et al., 2011a). In previous European mapping projects, sea energy expressed as Newton per square meter (N/m2) has been divided in three range classes based on wave and/or current. ‘Moderate energy’ was characterized using a tidal effect between 1.8 and 4.0 N/m2 in the MESH project for the Atlantic North (Coltman et al., 2008). The UKSeaMap project, combining the tidal and wave effects, computed respectively 0.13–1.16 N/m2 and 0.21–1.2 N/m2 for ‘Moderate energy’ class (McBreen et al., 2011a). Due to regional specificity and a low level of agreement between studies across Europe (Galparsoro et al., 2012) it is not adequate to use thresholds from one region to another one. Validated thresholds which could provide confidence in output maps for the Iberian coast are therefore desirable. The aim of the present paper is to validate and establish appropriate thresholds on sea energy and biological zonation to fit the EUNIS classification system for the Western Atlantic Iberian coast, using environmental (fraction of photosynthetically available light and kinetic wave energy) and biological data (presence of seaweeds species). 2. Material and methods 2.1. Study area and biological data This study was carried out within the framework of the MeshAtlantic project (www.meshatlantic.eu/) which was built on the approach of the MESH (Coltman et al., 2008) and EUSeaMap European projects (Cameron and Askew, 2011). Using the EUNIS classification system of habitats (version 2007-11), the MeshAtlantic project (2010–2013) proposed to collate and provide harmonized habitat maps for the Atlantic area, including the Iberian coast. Biological and depth data used to establish thresholds between biological zones were collected from a previous regional project (Rensub) carried out seasonally over 7 years (2003–2010) in the South coast of Portugal (Gonçalves et al., 2004, 2007, 2008, 2010). The Rensub project mapped marine habitat/biotopes off the Algarve coast between the localities of ‘Faro’ and ‘Lagos’ (340 km2) from 0 to 30 m depth (Algarve study area) (Fig. 1). Seaweed percentage cover (Chlorophyta — green algae, Heterokontophyta — brown algae and Rhodophyta
Fig. 1. Sampling station from the Rensub data base project (2003–2010) used for computing light threshold.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
— red algae) was evaluated using underwater visual census (SCUBA diving) and quadrat method (50 × 50 cm) on rocky areas. A total of 101 sampling stations were studied from the Rensub data base, with 3 replicates each giving a total number of 303 quadrates evaluated (Fig. 1). To establish classes of wave energy exerted on the sea bottom, the presence of kelp species (including the orders Laminariales and Tilopteridales: Saccorhiza polyschides, Laminaria hyperborea and Laminaria ochroleuca) occurring along the entire mainland Portugal coast (Fig. 2) was analyzed. These biological data were provided from a previous project (FindKelp project) which aimed to assess the Portuguese kelp meadow (from 15th of May to 31st August 2008) distribution status (Assis et al., 2009). 2.2. Environment data collection 2.2.1. Light penetration Light penetration is a limiting factor for seaweed development in coastal waters. A data layer of the diffuse attenuation coefficient photosynthetically available radiation (KdPAR) was used to compute
3
the surface light reaching the sea bottom or the fraction of photosynthetically available light. This variable relies on the entire 400– 700 nm spectral range which is the photosynthetically active region (Morel et al., 2007). Diffuse attenuation coefficient data was obtained from the MERIS sensors (Medium Resolution Imaging Spectrometer Instrument) aboard the European Envisat satellite. Computed KdPAR data layer was a mean over the period 2005–2009, with a full resolution of 250 m (Saulquin et al., 2013).
2.2.2. Wave energy The wave energy layer for the Western Iberian Sea area computed using the WAVE WATCH (WW3) model (IST — Wave Forecast for the Portuguese coast posted at www.maretec.mohid.com/ww3/) was analyzed. The wave induced energy layer was used as a surrogate of sea energy to distinguish between energy categories (high, moderate and low) for subtidal EUNIS Level 3 rocky habitats. The energy layers (0.05° resolution) expressed in terms of kinetic energy acting on the seabed were produced by the Research Centre on Marine Environmental and Technology (MARETEC; http://www.maretec.org/) (Fig. 3).
Fig. 2. Sampling station from the FindKelp data base project (from 15th of May to 31st August 2008) used for computing of wave energy threshold.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
4
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
The presence of kelp (Fig. 2) was used to test and split wave energy classes based on the kinetic wave energy data model. To establish energy class limits the values below percentile 90% (p90) of wave-induced energy was investigated. This statistical output layer which excludes extreme events was used in agreement with previous European mapping projects (e.g. MESH and EUSeaMap). According to Cameron and Askew (2011) it is important to filter out major events by taking high percentile statistics over as long periods as possible. 2.3. Threshold analysis 2.3.1. Biological zonation A data layer of the KdPAR was cross-tabulated with the field data on the presence/absence of seaweeds. Marine Geospatial Ecology Tools (MGET package) (Roberts et al., 2010) for ArcGIS software was used for the cross-tabulation. KdPAR with in situ depth was used to compute a mean value of the fraction of photosynthetically available light (FR) at each sampling location. Depth data measured during seaweed sampling were corrected to the mean water level (vertical datum).
The computation was performed according to the following formula (Hamdi et al., 2010): ð−hKdPARÞ
FR ¼ e
100:
Mann–Whitney U tests (Zar, 1999) were carried out to compare the fraction of light assigned to the presence against the absence of each seaweed taxon. The analysis was carried out independently considering all four seasons and for the summer season alone. Sampling of FR assigned to the presence of seaweeds was analyzed by descriptive statistic. Data was ordered from the largest to the smallest, with the observations falling below 25% of all assessed values assigned to the lower quartile (Q1/4) (Zar, 1999). This lower quartile splits the data and indicates the amount of light where 75% of the sample is present, which could define the hard photic threshold. In agreement with the method used in the MESH (Coltman et al., 2008) and EUSeaMap project (Cameron and Askew, 2011; McBreen and Askew, 2011) this quartile was established as a threshold between the infralittoral and circalittoral. The bootstrap technique was implemented on the R software (R Development Core Team, 2008) to avoid data dependence on sampling
Fig. 3. Wave-induced energy model (90th percentile) of the Western Iberian coast.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
locations and to assess the relative confidence. Bootstrapping uses resampling with replacement to estimate the statistics sampling distribution and estimates standard errors and confidence intervals (CI) for that particular statistic (Haukoos and Lewis, 2005). The resample was based on 1000 theoretical samples of the original set and has the same number of samples as the original. 2.3.2. Sea bed wave energy Energy exerted on the seabed can be characterized in a variety of ways that account for effects due to tidal currents or waves, or their combined effects (Bettencourt et al., 2004). However, due to higher range values and effect in shallower areas the wave energy was deemed an appropriate surrogate for sea energy levels. Values of kinetic wave energy were associated with the presence of both kelp species (S. polyschides, L. hyperborea and L. ochroleuca). These cross-tabulated values of wave energy data were ordered using the same method as used for the biological zonation. The optimal range of wave energy for both kelp species were analyzed separately by descriptive statistic. This analysis took into account that kelp tends to grow in areas highly to moderately exposed to wave or current environments (Birkett et al., 1998; Mumford, 2007). However, according to the same authors the liver weed L. hyperborea is for instance unable to survive where wave action is extreme. In the absence of an abundant habitat description database useful for validation like the Marine Recorder database used in the UKSeaMap and EUSeaMap (Cameron and Askew, 2011; McBreen et al., 2011a) an alternative approach was required. Therefore, we used the presence of kelp as a marker of a ‘moderately exposed’ location. Particular attention was given to the kelp species S. polyschides due to its broad distribution along the Portuguese
5
continental coast and representativeness along the wave energy gradient. In the present study the 12.5 and 87.5 percentiles of wave energy were used in order to remove extreme values splitting the data that indicate the amount of energy appropriate for the kelp species. To avoid data dependence on sampling locations and to assess the relative confidence, a bootstrap technique was also implemented (Haukoos and Lewis, 2005; R Development Core Team, 2008). Standard errors (SE) and confidence intervals (CI) for that particular statistic were estimated for this approach. 3. Results 3.1. Biological zonation Taking into consideration only the presence of seaweed (green, brown and red algae) no statistically significant difference was found (Mann–Whitney U tests; p = 0.436) among all seasons and the summer season alone, i.e. the difference in the median fraction of light values was not great enough to exclude the possibility that the difference is due to random sampling variability. The average of the fraction of light values was higher than the standard deviations for all of the analyzed datasets, indicating that the data was not scattered over a large range of values. The green algae (20 green algae species: e.g. Codium adhaerens, Codium effusum, Codium tomentosum, Codium vermilara and Valonia utricularis) were present between the fraction of light values of 2.4% and 33.0% (mean ± SD = 9.0 ± 8.2%; median = 6.6%; range = 30.6%). The brown algae (36 brown algae species: e.g. Cystoseira usneoides, Dictyota dichotoma, Halopteris filicina and Taonia atomaria)
Fig. 4. Histogram of the original fraction of light data assigned to both seaweed taxa (a, b and c) and boxplot presenting bootstrap statistics (d). The P25% lines highlight the bootstrap's 1st quartile results.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
6
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
(mean ± SD = 6.4 ± 4.6%; median = 5.5%; range = 32.2%) and red algae (94 red algae species: e.g. Plocamium cartilagineum, Peyssonnelia rubra, Sphaerococcus coronopifolius, Peyssonnelia squamaria, Asparagopsis armata, Rhodymenia holmesii and Gelidium latifolium) (mean ± SD = 6.5 ± 5.2%; median = 5.2%; range = 32.2%) were presented between the fraction of light values of 0.8% and 33.0%. Mann–Whitney U tests were carried out between the presence and absence within each algae taxon (Chlorophyta — green algae, Heterokontophyta — brown algae and Rhodophyta — red algae). Significant differences were observed for the three performed tests (Mann–Whitney U tests; p = b0.001) for each algae taxon. The difference in the median fraction of light values was larger than would be expected by chance between the presence and absence within each taxon. The estimated 1st quartile (P25%) through the bootstrap technique indicated that 75% of green and brown algae were located in areas with more than 4.63% (CI: 4.62–4.65%) and 4.05% (CI: 4.04–4.06%) of surface light reaching the seabed. Red algae were on the other hand located in areas with more than 3.65% (CI: 3.64–3.66%) of surface light reaching the seabed (Fig. 4). The computed bias for both seaweed taxa indicates an adequate accurate model being less than 2% in relation to the estimated first quartile value (green algae: − 0.086; red algae: 0.068; brown algae: 0.003). Both thresholds are located between 20 and 30 meter depth, being closer to the shore on the West side coast. The biological threshold between the infralittoral and the circalittoral is highlighted in Fig. 5 using the minimum value and first quartile for red algae (0.80 and 3.65%).
0.6099 N/m2, with a mean value of 0.0011 N/m2. Values of kinetic wave energy observed for each kelp species indicated a decrease of the average value from Saccorhiza polyschides to Laminaria hyperborea and Laminaria ochroleuca, with the same arising in range values. S. polyschides was observed in coastal areas with values of energy between 0.00005 and 0.19916 N/m2 (mean ± SD = 002747 ± 0.02949 N/m2; median = 0.02756 N/m2; range = 0.19911 N/m2) (Table 1). Estimated percentile analysis through the bootstrap technique (12.5% and 87.5%) on data for S. polyschides pointed out that the energy class was between 0.00303 and 0.04385 N/m2 (Table 1). L. hyperborea, a species that is unable to survive in extreme wave action conditions, was observed in places with more than 0.00611N/m2 and less than 0.04918 N/m2 (Table 1). Computed bias for both thresholds indicates acceptable accurate model simulation, being 13.0% and 0.7% for the percentiles 12.5% and 87.5% respectively. The ‘moderate energy’ class assigned to the presence of S. polyschides is pointed out in the histogram using original data analysis (Fig. 6a) and in the boxplot after applying the bootstrap technique (Fig. 6b). Fig. 7 shows the distribution map of energy classes set using S. polyschides for the entire Iberian West coast. According to the established classes, the ‘high energy’ class is mainly observed in the North coast of Portugal while the Algarve South coast is mainly covered by areas with low energy levels and in minor percentage with the ‘moderate energy’ levels.
3.2. Wave energy threshold
The EUNIS habitat classification scheme which relies on thresholds is based on few environmental variables that mainly structure the biological communities (Cameron and Askew, 2011). Accurate and validated thresholds (e.g. biological zonation and coastal energy) to
Predicted percentile values (p90%) of kinetic wave-induced energy for the Portuguese mainland coast ranged from 0 to a maximum of
4. Discussion
Fig. 5. Biological threshold between the infralittoral and the circalittoral zones in the Algarve study area (a) using different fractions of light: b) minimum observed value (0.80%) and c) 1st quartile (P25%) through the bootstrap technique (3.65%) for red algae species.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
7
Table 1 Kinetic wave energy (N/m2) values observed (original) with the presence of kelp species and main results through the bootstrap technique. S. polyschides Bootstrap Mean Standard error − confidence (95%) + confidence (95%) BIAS Original data Max Min Range Mean n samples
12.5% 0.00303 0.00142 0.00295 0.00312 0.00039
L. hyperborea 87.5% 0.04385 0.00147 0.04376 0.04394 0.00032
0.19916 0.00005 0.19911 0.02747 143
implement EUNIS habitat mapping are therefore desirable. The regional specificity and low level of agreement between available studies reinforce the need to compute and establish the thresholds obtained by the present study for the Western Atlantic Iberian coast: subtidal biological zonation and coastal energy level thresholds. The sub-littoral seabed is in general a heterogeneous environment that supports a wide biodiversity. Broad scale patterns of zonation have been mostly described qualitatively and some generalizations on the organization of marine communities were established (e.g. Pérès, 1967; Pérès and Picard, 1964; Saldanha, 1995). Those generalizations have stated that changes in depth and in environmental factors such as light availability are important physical reasons for the zonation of coastal communities. Typically sub-littoral ecological zones and boundaries are obtained through the characterization of the vertical distribution of certain indicator species, namely the abundance and frequency of macrophytes and algae, and sessile fauna such as sponges and ascidians which are less tolerant to light (Logan et al., 1984). The infralittoral zone is characterized as a submerged zone that extends to the compatible limit of marine phanerogam or photophile species. In contrary, the circalittoral zone is
12.5% 0.00611 0.00219 0.00597 0.00624 −0.00025
0.07299 0.00281 0.07017 0.02326 10
L. ochroleuca 87.5% 0.04918 0.01581 0.04820 0.05016 0.00019
12.5% 0.00005 0.00073 0.00000 0.00009 0.00010
87.5% 0.03087 0.00616 0.03049 0.03125 −0.00312
0.03335 0.00000 0.03335 0.01046 23
characterized by the presence of certain sciaphilic algae species and abundant fauna communities (e.g. Pérès, 1967; Pérès and Picard, 1964). The present study emphasizes the importance of the light fraction that reaches the sea bottom as a significant factor for algae zonation and consequently to other sea organisms. The results point out that the transition between the infralittoral and circalittoral zones might occur when no more than 3.65% of light reaches the seabed. This statement was based on the 1st quartile of the ordered data which was used as indicative value. This indicative value suggests that 75% of red algae are located in areas in which more than 3.65% of surface light reaches the seabed. The value of 2.36% of surface incident light was obtained from the MESH study area through the same indicative quartile (1st quartile) (Coltman et al., 2008). The UKSeaMap project results indicate that the transition between the circalittoral and the infralittoral zones occurs when less than 1% of light reaches the seabed (McBreen and Askew, 2011). This difference has been attributed to the improved resolution of light data (from 9 to 4 km) and increased number of biological records in the UKSeaMap 2010 analysis (McBreen and Askew, 2011). McBreen and Askew (2011) in the UKSeaMap 2010 project also defined the hard threshold to be close to 1% of light reaching the seabed
Fig. 6. Histogram and boxplot presenting bootstrap statistics for kinetic energy assigned to the presence of Saccorhiza polyschides.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
8
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
which was defined by the 1st quartile, above which 75% of kelp biotopes were present. Using a slightly different approach the computed value was around 4.5% for the North and Celtic Sea in the EUSeaMap project (Cameron and Askew, 2011). All these disagreements highlight the fact that, although one study may suggest one threshold value, it cannot be assumed that it should be applied to other datasets or regions without a new analysis being undertaken. The minimum value obtained in the present study point out that both brown and red algae are still present with at least more than 0.80% of light reaching the seabed. This value seems to be the limit for the development of most sciaphilic algae species, which are mostly adapted to live in low light environments. The minimum light value of 0.80% for the presence of brown and red algae is closer to the value of 1% reported in the literature for other European regions and used on the UKSeaMap project. The indicative value of 3.65% seems very realistic and closer to the 20 to 24 meter depth given by the Portuguese literature (e.g. Saldanha, 1995) as the boundary between the infralittoral and circalittoral zones. Nevertheless, being the minimum value acceptable, within the MeshAtlantic framework it was decided to use the minimum light value of 0.80% (rounded to 1%) in order to
standardize in the European level the broad scale habitat maps produced (Vasquez et al., in press). Differences between the present study and others that have been carried out in the European coast might be induced by different indicator species used (e.g. seagrass, kelp and other macroalgae), habitat disparity or expected differences in environmental factors (e.g. temperature, turbidity and salinity). Among these limitations, there are also different sources of light attenuation and higher resolution that should be considered in future attempts to establish thresholds which could provide finer tuning in small areas (Ellwood et al., 2011). Further studies are needed because the light is only one of the various environmental factors affecting the presence of algae. The hydrodynamic energy regime is also an important environmental factor, playing a significant role in the sedimentary cover and on benthic species settlement, therefore contributing to structure the biological communities. Thus, as previously mentioned, it is carefully considered in the EUNIS classification scheme, being explicit on the rocky area habitats (Connor et al., 1997a, 1997b, 2003, 2004; Davies et al., 2004). Given the available energy model our approach characterized ‘moderate energy’ for the Iberian Atlantic coast between 0.00303
Fig. 7. EUNIS energy classes established based on Saccorhiza polyschides spatial distribution and kinetic wave energy (N/m2).
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
N/m2 and 0.04385 N/m2. After mapping the results for the entire Iberian coast, it was observed that this is only roughly in agreement with the discrete thresholds already established in published literature (Bettencourt et al., 2004). According to these authors the Portuguese coast is broadly divided in three different hydrodynamic regime areas. The energy of the coast rises according to Bettencourt et al. (2004), from the South to the North, being sheltered from the South Portuguese border to Lagos, moderately exposed from Lagos to Cape Carvoeiro and exposed from Cape Carvoeiro to the Northern Portuguese border. The results obtained in the present paper showed that the trend should be much less discrete, conditioned by bathymetry characteristics and orientation of the coast. As shown, S. polyschides could be used to establish energy thresholds for the West Iberian coast since the values of kinetic wave energy that the species can endure, along with its wide distribution along the mainland Portuguese coast seem to specify this kelp species as a suitable indicator species of moderate class energy, allowing the definition of two additional EUNIS energy classes (high and low energy classes). Concerning the energy classes, our approach with S. polyschides gave more realistic results to create wave energy classes when compared to pilot studies that establish areas for energy production in mainland Portugal (e.g. POEM, 2014). We have shown that the identified energy classes have a more scattered distribution along the Portuguese coast. However, the higher energy level was predominantly observed on the North while the moderate and low energy levels were more common in the southern coast. As expected, the computed thresholds also differed from those attained by the previous EU funded projects (1.8– 4.0 N/m2 in MESH and 0.21–1.2 N/m2 in EUSeaMap) mainly because they are based on different models. The difference of range among studies highlights the fact that it cannot be applied to other regions without a new analysis being undertaken. 5. Conclusions This study allowed the computation of appropriate thresholds on biological zonation and wave energy for the EUNIS habitat classification of the Western Atlantic Iberian coast. For that purpose, it has established a link between the spatial distributions of benthic macroalgae, light penetration and kinetic wave energy. Furthermore, this study highlighted the importance of estimating the variability associated with these thresholds, mainly at a regional or national scale, and the need of additional base studies on the Atlantic Iberian coast. In order to improve the threshold on sea energy the integration of current energy in the analysis is important. The study highlights the need for an integrated work in a European scale which can contribute for a better standardization of both EUNIS thresholds studied here. However, the present study will be very useful in the near future to implement and establish the West Iberian marine EUNIS habitats inventory. Acknowledgments This work was made possible by the availability of biological historical data from a previous research project, Rensub (CCMAR-ARH Algarve: ‘Cartography and characterization of the marine communities off the National Underwater Ecological Reserve’). We would like to thanks Jacques Populus and Mickael Vasquez from the IFREMER (Institut Français de Recherche pour L'Exploitation de la Mer: http://wwz.ifremer.fr/institut) for providing the coefficient photosynthetically available radiation (KdPAR) layer. We would like to acknowledge professor Ramiro Neves and its Research team (MARETEC — http://www.maretec.org/) from the Instituto Superior Técnico (IST) the Engineering School of Lisbon Technical University for providing us with data of wave and currents. We also would like to kindly acknowledge our colleagues Alexandra Cunha and Jorge Assis from CCMAR, University of Algarve for providing data on the presence of kelp species at the Portuguese coast (FindKelp
9
project). Finally special thanks to all the MeshAtlantic partners for the opportunity to discuss this subject. We are grateful for the comments and suggestions of two referees who contributed greatly to improving the manuscript.
References Allee, R.J., Dethier, M., Brown, D., Deegan, L., Ford, G.R., Hourigan, T.R., Maragos, J., Schoch, C., Sealy, K., Twilley, R., Weinstein, M.P., Yoklavich, M., 2000. Marine and estuarine ecosystem and habitat classification. NOAA Technical Memorandum, NMFS-F/SPO, Silver Spring, Maryland (43 pp.). Assis, J., Tavares, D., Tavares, J., Cunha, A., Alberto, F., Serrão, E.A., 2009. Findkelp, a GISbased community participation project to assess Portuguese kelp conservation status. J. Coast. Res. Special Issue 56, 1469–1473. Ballesta, L., Pergent, G., Pasqualini, V., Pergent-Martini, C., 2000. Distribution and dynamics of Posidonia oceanica beds along the Albères coastline. Life Sci. 323, 407–414. Bettencourt, A.M., Bricker, S.B., Ferreira, J.G., Franco, A., Marques, J.C., Melo, J.J., Nobre, A., Ramos, L., Reis, C.S., Salas, F., Silva, M.C., Simas, T., Wolff, W., 2004. Typology and reference conditions for Portuguese transitional and coastal waters. Development of Guidelines for the Application of the European Union Water Framework Directive. INAG/IMAR, Lisboa, Portugal (100 pp.). Birkett, D.A., Maggs, C.A., Dring, M.J., Boaden, P.J.S., Seed, R., 1998. Infralittoral reef biotopes with kelp species (volume VII). An overview of dynamic and sensitivity characteristics for conservation management of marine SACs. Scottish Association of Marine Science (UK Marine SACs Project) (174 pp.). Butler, A., Harris, P.T., Lyne, V., Heap, A., Passlow, V.L., Smith, R.N.P., 2001. An interim, draft bioregionalisation for the continental slope and deeper waters of the south-east marine region of Australia. Report to the National Oceans Office, Geoscience Australia and CSIRO, Hobart (35 pp.). Cameron, A., Askew, N. (Eds.), 2011. EUSeaMap — Preparatory Action for Development and Assessment of a European Broad-scale Seabed Habitat Map Final Report (226 pp. Available at http://jncc.defra.gov.uk/euseamap). Coltman, N., Golding, N., Verling, E., 2008. Developing a broadscale predictive EUNIS habitat map for the MESH study area. Available at, http://www.searchmesh.net/ pdf/MESH%20EUNIS%20model.pdf (16 pp.). Connor, D.W., Hiscock, K., Foster-Smith, R.L., Covey, R., 1995. A classification system for benthic marine biotopes. In: Eleftheriou, A., Ansell, A.D., Smith, C.J. (Eds.), Biology and Ecology of Shallow Coastal Waters. Proceedings of the 28th European Marine Biology Symposium. Olsen and Olsen, Fredensborg, pp. 155–165. Connor, D.W., Brazier, D.P., Hill, T.O., Northen, K.O., 1997a. Marine Nature Conservation Review: marine biotope classification for Britain and Ireland. Volume 1. Littoral biotopes. Version 97.06. JNCC Report, No. 229. Connor, D.W., Dalkin, M.J., Hill, T.O., Holt, R.H.F., Sanderson, W.G., 1997b. Marine Nature Conservation Review: marine biotope classification for Britain and Ireland. Volume 2. Sublittoral biotopes. Version 97.06. JNCC Report, No. 230. Connor, D.W., Allen, J.H., Golding, N., Lieberknecht, L.M., Northen, K.O., Reker, J.B., 2003. The national marine habitat classification for Britain and Ireland. Version 03.02. Introductory text. Joint Nature Conservation Committee, Peterborough (42 pp. Available at www.jncc.defra.gov.uk/marinehabitatclassification Accessed: July 2014). Connor, D., Allen, J.H., Golding, N., Howell, K.L., Lieberknecht, L.M., Northen, K.O., Reker, J., 2004. The Marine Habitat Classification for Britain and Ireland. Version 04.05. Introduction. Joint Nature Conservation Committee, Peterborough (49 pp. Available at http://jncc.defra.gov.uk/MarineHabitatClassification). Costello, M.J., 2009. Distinguishing marine habitat classification concepts for ecological data management. Mar. Ecol. Prog. Ser. 397, 253–268. http://dx.doi.org/10.3354/ meps08317. Davies, C.E., Moss, D., Hill, M.O., 2004. EUNIS habitat classification revised 2004. Report to the European Topic Centre on Nature Protection and Biodiversity. European Environment Agency (307 pp. Available at http://www.eunis.eea.europa.eu/habitatscodebrowser.jsp). Devillers, P., Devillers-Terschuren, J., 1996. A classification of Palaearctic habitats. Nature and Environment, No 78. Council of Europe, Strasbourg. Devillers, P., Devillers-Terschuren, J., Ledant, J.P., 1991. CORINE Biotopes Manual vol. 2. Habitats of the European Community. Office for Official Publications of the European Communities, Luxembourg. Ellwood, H.N., Askew, N.A., Cameron, J., McBreen, F., 2011. UKSeaMap 2010. Marine Ecological Surveys Limited (MESL): Technical Report 7. Joint Nature Conservation Committee, Peterborough (23 pp.). Evans, D., 2012. The EUNIS habitats classification — past, present and future. Rev. Investig. Mar. 19 (2), 28–29. Galparsoro, I., Connor, D.W., Borja, A., Aish, A., Amorim, P., Bajjouk, T., Chambers, C., Coggan, R., Dirberg, G., Ellwood, H., et al., 2012. Using EUNIS habitat classification for benthic mapping in European seas: present concerns and future needs. Mar. Pollut. Bull. 64 (12), 2630–2638. Gonçalves, J.M.S., Monteiro, P., Coelho, R., Afonso, C., Ribeiro, J., Almeida, C., Veiga, P., Machado, D., Berecibar, E., Oliveira, F., Bentes, L., 2004. Mapeamento de biocenoses marinhas da Reserva Ecológica Nacional Submarina entre Albufeira e Vale do Lobo. CCDR Algarve, Universidade do Algarve, CCMAR, Faro (182 pp.). Gonçalves, J.M.S., Monteiro, P., Coelho, R., Afonso, C., Almeida, C., Veiga, P., Machado, M., Machado, D., Oliveira, F., Ribeiro, J., Abecasis, D., Primo, L., Tavares, D., FernándezCarvalho, J., Abreu, S., Fonseca, L., Erzini, K., Bentes, L., 2007. Cartografia e caracterização das biocenoses marinhas da Reserva Ecológica Nacional Submarina
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
10
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
entre a Galé e a barra Nova do Ancão. Relatório Final. CCDR Algarve, Universidade do Algarve, CCMAR, Faro (250 pp. + anexos). Gonçalves, J.M.S., Monteiro, P., Afonso, C., Almeida, C., Oliveira, F., Rangel, M., Ribeiro, J., Machado, M., Veiga, P., Abecasis, D., Pires, F., Fonseca, L., Erzini, K., Bentes, L., 2008. Cartografia e caracterização das biocenoses marinhas da Reserva Ecológica Nacional Submarina entre a Galé e a foz do rio Arade. Relatório Final CCDR Algarve. Universidade do Algarve, CCMAR, Faro (144 pp. + Anexos). Gonçalves, J.M.S., Monteiro, P., Afonso, C., Oliveira, F., Rangel, M., Machado, M., Veiga, P., Leite, L., Sousa, I., Bentes, L., Fonseca, L., Erzini, K., 2010. Cartography and characterization of the marine communities off the National Underwater Ecological Reserve between the Arade River and Ponta da Piedade. Final Report. ARH Algarve. University of Algarve, CCMAR, Faro (122 pp. + Anexos). Hamdi, A., Vasquez, M., Populus, J.Y., 2010. Cartographie des habitats physiques Eunis — Côtes de France. Convention Ifremer/AAMP. no 09/12177764/FY. Available at. http://archimer.ifremer.fr/doc/00026/13751/. Haukoos, J.S., Lewis, R.J., 2005. Advanced statistics: bootstrapping confidence intervals for statistics with “difficult” distributions. Acad. Emerg. Med. 12 (4), 360–365. http://dx. doi.org/10.1197/j.aem.2004.11.018. Hiscock, K., Connor, D.W., 1991. Benthic marine habitats and communities in Great Britain: the development of an MNCR classification. JNCC Report 6; Marine Nature Conservation Review Report No. MNCR/OR/14. Joint Nature Conservation Committee, Peterborough (92 pp.). Howell, K.L., 2010. A benthic classification system to aid in the implementation of marine protected area networks in the deep/high seas of the NE Atlantic. Biol. Conserv. 143, 1041–1056. ICES, 2001. Report of the Working Group on Marine Habitat Mapping (WGMHM), 3–6 April 200, Galway, Ireland, ICES CM 2000/E/08 (42 pp.). Logan, A., Page, F.H., Thomas, M.L.H., 1984. Depth zonation of epibenthos on sublittoral hard substrates off Deer Island, Bay of Fundy, Canada. Estuar. Coast. Shelf Sci. 18 (5), 571–592. http://dx.doi.org/10.1016/0272-7714(84) 90091-X. McBreen, F., Askew, N., 2011. Light penetration layer. UKSeaMap 2010. Technical Report 2. JNCC, Peterborough (7 pp.). McBreen, F., Askew, N., Cameron, A., 2011a. UKSeaMap 2010. Technical Report 4: Energy, JNCC Report (38 pp.). McBreen, F., Askew, N., Cameron, A., Connor, D., Ellwood, H., Carter, A., 2011b. UKSeaMap 2010: predictive mapping of seabed habitats in UK waters. JNCC Report, No. 446 (103 pp.). Morel, A., Huot, Y., Gentili, B., Werdell, P.J., Hooker, S.B., Franz, B.A., 2007. Examining the consistency of products derived from various ocean color sensors in open ocean (Case 1) waters in the perspective of a multi-sensor approach. Remote Sens. Environ. 111, 69–88.
Mumford, T.F., 2007. Kelp and eelgrass in Puget Sound. Technical report 2007-05. Prepared in Support of the Puget Sound Nearshore Partnership. Washington Department of Natural Resources, Aquatic Resources Division (27 pp.). Olenin, S., Ducrotoy, J.P., 2006. The concept of biotope in marine ecology and coastal management. Mar. Pollut. Bull. 53, 20–29. Pérès, J.M., 1967. The Mediterranean benthos. Oceanogr. Mar. Biol. Annu. Rev. 5, 449–533. Pérès, J.M., Picard, J., 1964. Nouveau manuel de bionomie benthique de la mer Mediterranée. Reel Trav. Stat. Mar. Endoume 31 (47), 1–137. POEM, 2014. Portuguese Marine Spatial Planning. Available at, http://poem.inag.pt/ (Accessed on January, 29th). R Development Core Team, 2008. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria 3-900051-07-0 (URL http://www.R-project.org). Roberts, J.J., Best, B.D., Dunn, C.D., Treml, E.A., Halpin, P.N., 2010. Marine Geospatial Ecology Tools: an integrated framework for ecological geoprocessing with ArcGIS, Python, R, MATLAB, and C++. Environ. Model. Softw. 25, 1197–1207. http://dx.doi. org/10.1016/j.envsoft.2010.03.029. Saldanha, L., 1995. Fauna Submarina Atlântica. Portugal continental, Açores, Madeira. Publicações Europa América (364 pp.). SANBI, 2009. Further development of a proposed national wetland classification system for South Africa. Primary Project Report. Prepared by the Freshwater Consulting Group (FCG) for the South African National Biodiversity Institute (SANBI) (74 pp.). Saulquin, B., Hamdi, A., Gohin, F., Populus, J., Mangin, A., d' Andon, O.F., 2013. Estimation of the diffuse attenuation coefficient KdPAR using MERIS and application to seabed habitat mapping. Remote Sens. Environ. 128, 224–233. http://dx.doi.org/10.1016/j. rse.2012.10.002. Snelder, T.H., Leathwick, J.R., Dey, K.L., Rowden, A.A., Weatherhead, M.A., Fenwick, G.D., Francis, M.P., Gorman, R.M., Grieve, J.M., Hadfield, M.G., Hewitt, J.E., Richardson, K.M., Uddstrom, M.J., Zeldis, J.R., 2006. Development of an ecological marine classification in the New Zealand region. Environ. Manag. 39 (1), 12–29. Vasquez, M., Populus, J., Mata Chacón, D., Sanz Alonso, J.L., Tempera, F., Gonçalves, J.M.S., O'Keeffe, E., Galparsoro, I., Bentes, L., Amorim, P., Henriques, V., McGrath, F., Monteiro, P., Mendes, B., Freitas, R., Martins, R., 2014. Mapping seabed habitats of the north-east Atlantic using environmental data. J. Sea Res http://dx.doi.org/10.1016/j.seares.2014. 09.011 (in press). Zacharias, M.A., Howes, D.E., Harper, J.R., Wainwright, P., 1998. The British Columbia marine ecosystem classification: rationale, development, and verification. Coast. Manag. 26, 105–124. Zar, J.H., 1999. Biostatistical Analysis. 4th edn. Prentice-Hall, Upper Saddle River, NJ (663 pp.).
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
Contents lists available at ScienceDirect
Journal of Sea Research journal homepage: www.elsevier.com/locate/seares
EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study Pedro Monteiro ⁎, Luis Bentes, Frederico Oliveira, Carlos M.L. Afonso, Mafalda O. Rangel, Jorge M.S. Gonçalves Centre of Marine Sciences (CCMAR), Universidade do Algarve, Ed. 7, Campus de Gambelas, 8005-139 Faro, Portugal
a r t i c l e
i n f o
Article history: Received 31 January 2014 Received in revised form 11 August 2014 Accepted 28 November 2014 Available online xxxx Keywords: EUNIS classification Habitat mapping Thresholds Biological zonation Kinetic wave energy Seaweeds
a b s t r a c t The European Nature Information System (EUNIS) has been implemented for the establishment of a marine European habitats inventory. Its hierarchical classification is defined and relies on environmental variables which primarily constrain biological communities (e.g. substrate types, sea energy level, depth and light penetration). The EUNIS habitat classification scheme relies on thresholds (e.g. fraction of light and energy) which are based on expert judgment or on the empirical analysis of the above environmental data. The present paper proposes to establish and validate an appropriate threshold for energy classes (high, moderate and low) and for subtidal biological zonation (infralittoral and circalittoral) suitable for EUNIS habitat classification of the Western Iberian coast. Kinetic wave-induced energy and the fraction of photosynthetically available light exerted on the marine bottom were respectively assigned to the presence of kelp (Saccorhiza polyschides, Laminaria hyperborea and Laminaria ochroleuca) and seaweed species in general. Both data were statistically described, ordered from the largest to the smallest and percentile analyses were independently performed. The threshold between infralittoral and circalittoral was based on the first quartile while the ‘moderate energy’ class was established between the 12.5 and 87.5 percentiles. To avoid data dependence on sampling locations and assess the confidence interval a bootstrap technique was applied. According to this analysis, more than 75% of seaweeds are present at locations where more than 3.65% of the surface light reaches the sea bottom. The range of energy levels estimated using S. polyschides data, indicate that on the Iberian West coast the ‘moderate energy’ areas are between 0.00303 and 0.04385 N/m2 of wave-induced energy. The lack of agreement between different studies in different regions of Europe suggests the need for more standardization in the future. However, the obtained thresholds in the present study will be very useful in the near future to implement and establish the Iberian EUNIS habitats inventory. © 2014 Elsevier B.V. All rights reserved.
1. Introduction The effects of human activities on the marine environment are increasingly evident and impossible to be ignored. Industrial fishing and pollution as well as several other anthropogenic impacts have been causing damage to marine species and their habitats. An ecosystem-based management approach within marine spatial planning requires standardized classifications and terminology across all possible habitats (Costello, 2009; Hiscock and Connor, 1991). There have been progresses into the development of habitat classifications and different schemes have been implemented, such as in Europe (Connor et al., 1995, 1997a, 1997b, 2003, 2004; Davies et al., 2004; Devillers and Devillers-Terschuren, 1996; Devillers et al., 1991), Canada (Zacharias et al., 1998), Australia (Butler et al., 2001), United States (Allee et al., 2000), New Zealand (Snelder et al., 2006) and in South Africa (SANBI, 2009).
⁎ Corresponding author. Tel.: +351 289 800 900; fax: +351 289 800051. E-mail address: [email protected] (P. Monteiro).
The system by which European habitats have been classified is called the European Nature Information System (EUNIS) (Connor et al., 2003, 2004; Davies et al., 2004; Evans, 2012). The marine section of the EUNIS classification was originally based on the UK BioMar classification (Connor et al., 1995, 1997a, 1997b). It was developed in collaboration with experts from all over Europe and managed by the European Topic Centre for Nature Protection and Biodiversity (ETC/NPB) for the European Environment Agency (EEA) and the European Environmental Information Observation Network (EIONET) (Davies et al., 2004). This system covers all types of natural and artificial habitats and is theoretically under development as new information becomes available (Howell, 2010). A habitat is the area where an organism, or population or communities of species occur (e.g. mussel beds on circalittoral rock) i.e. the type of environment where a species could potentially establish itself (Olenin and Ducrotoy, 2006). For the purposes of EUNIS habitat classification the term is used as the place where a biological community normally lives, characterized by its physical features and by the communities or assemblages of species that live there (Davies et al., 2004). Thus it is defined by the combination of several key environmental variables
http://dx.doi.org/10.1016/j.seares.2014.11.007 1385-1101/© 2014 Elsevier B.V. All rights reserved.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
2
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
(e.g. substrate types, exposure to wave action, strength of tidal currents, salinity, temperature, hydrographic regime), referred to as the ‘habitat envelope’, together with its associated biological community, operating at particular spatial and temporal scales (ICES, 2001). Hierarchical segmentation within the upper marine division of the EUNIS classification systems is defined by a few environmental variables which primarily drive/structure the distribution of biological communities such as substrate types, energy level (wave action and tidal current), depth and light penetration (Connor et al., 1995, 1997a, 1997b, 2003, 2004; Davies et al., 2004). The EUNIS marine classification scheme is initially divided up on the basis of biological zones (e.g. infralittoral and circalittoral) and substrate types (e.g. sand, rock) as well as on the depth range (level 2). At level 3 the segmentation is broadly based on ecological region (Atlantic, Baltic, Mediterranean and Black Sea) and three energy levels of the coast (low, moderate and high), while associated biological communities begin to appear on lower levels (Davies et al., 2004). Referred to as a ‘top-down’ approach, this classification scheme relies on thresholds (e.g. biological zonation and sea energy) that can be based on the literature, expert judgment or on the empirical analysis of the above environmental data variables (Cameron and Askew, 2011). The biological zones are classically obtained through the characterization of the vertical distribution of certain indicator species, such as macrophytes and algae, or sessile fauna such as sponges and ascidians which are less tolerant to light (Logan et al., 1984; Pérès, 1967; Pérès and Picard, 1964). The presence of macrophytes and algae (sea grasses, kelp and other seaweeds) is directly correlated to the fraction of light reaching the seabed (Birkett et al., 1998). Studies have stated that kelp species and seagrass (e.g. Posidonia oceanica) require a minimum of 1% of surface light to be able to develop (Ballesta et al., 2000; McBreen and Askew, 2011; McBreen et al., 2011b). This threshold value has been tested for different European regions by several marine mapping projects (e.g. 2.36% for the North Sea and Celtic Sea in MESH and 4.5% for the North and Celtic Sea and 1% for the West Mediterranean Sea in EUSeaMap) (Cameron and Askew, 2011; Coltman et al., 2008). McBreen and Askew (2011) in the UKSeaMap 2010 defined the hard threshold to be the value above which 75% of kelp biotopes were classed as infralittoral and obtained 1%. On the other hand, the concept of what constitutes ‘low’ or ‘high energy’ on the sea bottom is far from being easily established
(Galparsoro et al., 2012). Energy can be characterized in a variety of ways that account for effects due to tidal currents and waves (McBreen et al., 2011a). In previous European mapping projects, sea energy expressed as Newton per square meter (N/m2) has been divided in three range classes based on wave and/or current. ‘Moderate energy’ was characterized using a tidal effect between 1.8 and 4.0 N/m2 in the MESH project for the Atlantic North (Coltman et al., 2008). The UKSeaMap project, combining the tidal and wave effects, computed respectively 0.13–1.16 N/m2 and 0.21–1.2 N/m2 for ‘Moderate energy’ class (McBreen et al., 2011a). Due to regional specificity and a low level of agreement between studies across Europe (Galparsoro et al., 2012) it is not adequate to use thresholds from one region to another one. Validated thresholds which could provide confidence in output maps for the Iberian coast are therefore desirable. The aim of the present paper is to validate and establish appropriate thresholds on sea energy and biological zonation to fit the EUNIS classification system for the Western Atlantic Iberian coast, using environmental (fraction of photosynthetically available light and kinetic wave energy) and biological data (presence of seaweeds species). 2. Material and methods 2.1. Study area and biological data This study was carried out within the framework of the MeshAtlantic project (www.meshatlantic.eu/) which was built on the approach of the MESH (Coltman et al., 2008) and EUSeaMap European projects (Cameron and Askew, 2011). Using the EUNIS classification system of habitats (version 2007-11), the MeshAtlantic project (2010–2013) proposed to collate and provide harmonized habitat maps for the Atlantic area, including the Iberian coast. Biological and depth data used to establish thresholds between biological zones were collected from a previous regional project (Rensub) carried out seasonally over 7 years (2003–2010) in the South coast of Portugal (Gonçalves et al., 2004, 2007, 2008, 2010). The Rensub project mapped marine habitat/biotopes off the Algarve coast between the localities of ‘Faro’ and ‘Lagos’ (340 km2) from 0 to 30 m depth (Algarve study area) (Fig. 1). Seaweed percentage cover (Chlorophyta — green algae, Heterokontophyta — brown algae and Rhodophyta
Fig. 1. Sampling station from the Rensub data base project (2003–2010) used for computing light threshold.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
— red algae) was evaluated using underwater visual census (SCUBA diving) and quadrat method (50 × 50 cm) on rocky areas. A total of 101 sampling stations were studied from the Rensub data base, with 3 replicates each giving a total number of 303 quadrates evaluated (Fig. 1). To establish classes of wave energy exerted on the sea bottom, the presence of kelp species (including the orders Laminariales and Tilopteridales: Saccorhiza polyschides, Laminaria hyperborea and Laminaria ochroleuca) occurring along the entire mainland Portugal coast (Fig. 2) was analyzed. These biological data were provided from a previous project (FindKelp project) which aimed to assess the Portuguese kelp meadow (from 15th of May to 31st August 2008) distribution status (Assis et al., 2009). 2.2. Environment data collection 2.2.1. Light penetration Light penetration is a limiting factor for seaweed development in coastal waters. A data layer of the diffuse attenuation coefficient photosynthetically available radiation (KdPAR) was used to compute
3
the surface light reaching the sea bottom or the fraction of photosynthetically available light. This variable relies on the entire 400– 700 nm spectral range which is the photosynthetically active region (Morel et al., 2007). Diffuse attenuation coefficient data was obtained from the MERIS sensors (Medium Resolution Imaging Spectrometer Instrument) aboard the European Envisat satellite. Computed KdPAR data layer was a mean over the period 2005–2009, with a full resolution of 250 m (Saulquin et al., 2013).
2.2.2. Wave energy The wave energy layer for the Western Iberian Sea area computed using the WAVE WATCH (WW3) model (IST — Wave Forecast for the Portuguese coast posted at www.maretec.mohid.com/ww3/) was analyzed. The wave induced energy layer was used as a surrogate of sea energy to distinguish between energy categories (high, moderate and low) for subtidal EUNIS Level 3 rocky habitats. The energy layers (0.05° resolution) expressed in terms of kinetic energy acting on the seabed were produced by the Research Centre on Marine Environmental and Technology (MARETEC; http://www.maretec.org/) (Fig. 3).
Fig. 2. Sampling station from the FindKelp data base project (from 15th of May to 31st August 2008) used for computing of wave energy threshold.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
4
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
The presence of kelp (Fig. 2) was used to test and split wave energy classes based on the kinetic wave energy data model. To establish energy class limits the values below percentile 90% (p90) of wave-induced energy was investigated. This statistical output layer which excludes extreme events was used in agreement with previous European mapping projects (e.g. MESH and EUSeaMap). According to Cameron and Askew (2011) it is important to filter out major events by taking high percentile statistics over as long periods as possible. 2.3. Threshold analysis 2.3.1. Biological zonation A data layer of the KdPAR was cross-tabulated with the field data on the presence/absence of seaweeds. Marine Geospatial Ecology Tools (MGET package) (Roberts et al., 2010) for ArcGIS software was used for the cross-tabulation. KdPAR with in situ depth was used to compute a mean value of the fraction of photosynthetically available light (FR) at each sampling location. Depth data measured during seaweed sampling were corrected to the mean water level (vertical datum).
The computation was performed according to the following formula (Hamdi et al., 2010): ð−hKdPARÞ
FR ¼ e
100:
Mann–Whitney U tests (Zar, 1999) were carried out to compare the fraction of light assigned to the presence against the absence of each seaweed taxon. The analysis was carried out independently considering all four seasons and for the summer season alone. Sampling of FR assigned to the presence of seaweeds was analyzed by descriptive statistic. Data was ordered from the largest to the smallest, with the observations falling below 25% of all assessed values assigned to the lower quartile (Q1/4) (Zar, 1999). This lower quartile splits the data and indicates the amount of light where 75% of the sample is present, which could define the hard photic threshold. In agreement with the method used in the MESH (Coltman et al., 2008) and EUSeaMap project (Cameron and Askew, 2011; McBreen and Askew, 2011) this quartile was established as a threshold between the infralittoral and circalittoral. The bootstrap technique was implemented on the R software (R Development Core Team, 2008) to avoid data dependence on sampling
Fig. 3. Wave-induced energy model (90th percentile) of the Western Iberian coast.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
locations and to assess the relative confidence. Bootstrapping uses resampling with replacement to estimate the statistics sampling distribution and estimates standard errors and confidence intervals (CI) for that particular statistic (Haukoos and Lewis, 2005). The resample was based on 1000 theoretical samples of the original set and has the same number of samples as the original. 2.3.2. Sea bed wave energy Energy exerted on the seabed can be characterized in a variety of ways that account for effects due to tidal currents or waves, or their combined effects (Bettencourt et al., 2004). However, due to higher range values and effect in shallower areas the wave energy was deemed an appropriate surrogate for sea energy levels. Values of kinetic wave energy were associated with the presence of both kelp species (S. polyschides, L. hyperborea and L. ochroleuca). These cross-tabulated values of wave energy data were ordered using the same method as used for the biological zonation. The optimal range of wave energy for both kelp species were analyzed separately by descriptive statistic. This analysis took into account that kelp tends to grow in areas highly to moderately exposed to wave or current environments (Birkett et al., 1998; Mumford, 2007). However, according to the same authors the liver weed L. hyperborea is for instance unable to survive where wave action is extreme. In the absence of an abundant habitat description database useful for validation like the Marine Recorder database used in the UKSeaMap and EUSeaMap (Cameron and Askew, 2011; McBreen et al., 2011a) an alternative approach was required. Therefore, we used the presence of kelp as a marker of a ‘moderately exposed’ location. Particular attention was given to the kelp species S. polyschides due to its broad distribution along the Portuguese
5
continental coast and representativeness along the wave energy gradient. In the present study the 12.5 and 87.5 percentiles of wave energy were used in order to remove extreme values splitting the data that indicate the amount of energy appropriate for the kelp species. To avoid data dependence on sampling locations and to assess the relative confidence, a bootstrap technique was also implemented (Haukoos and Lewis, 2005; R Development Core Team, 2008). Standard errors (SE) and confidence intervals (CI) for that particular statistic were estimated for this approach. 3. Results 3.1. Biological zonation Taking into consideration only the presence of seaweed (green, brown and red algae) no statistically significant difference was found (Mann–Whitney U tests; p = 0.436) among all seasons and the summer season alone, i.e. the difference in the median fraction of light values was not great enough to exclude the possibility that the difference is due to random sampling variability. The average of the fraction of light values was higher than the standard deviations for all of the analyzed datasets, indicating that the data was not scattered over a large range of values. The green algae (20 green algae species: e.g. Codium adhaerens, Codium effusum, Codium tomentosum, Codium vermilara and Valonia utricularis) were present between the fraction of light values of 2.4% and 33.0% (mean ± SD = 9.0 ± 8.2%; median = 6.6%; range = 30.6%). The brown algae (36 brown algae species: e.g. Cystoseira usneoides, Dictyota dichotoma, Halopteris filicina and Taonia atomaria)
Fig. 4. Histogram of the original fraction of light data assigned to both seaweed taxa (a, b and c) and boxplot presenting bootstrap statistics (d). The P25% lines highlight the bootstrap's 1st quartile results.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
6
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
(mean ± SD = 6.4 ± 4.6%; median = 5.5%; range = 32.2%) and red algae (94 red algae species: e.g. Plocamium cartilagineum, Peyssonnelia rubra, Sphaerococcus coronopifolius, Peyssonnelia squamaria, Asparagopsis armata, Rhodymenia holmesii and Gelidium latifolium) (mean ± SD = 6.5 ± 5.2%; median = 5.2%; range = 32.2%) were presented between the fraction of light values of 0.8% and 33.0%. Mann–Whitney U tests were carried out between the presence and absence within each algae taxon (Chlorophyta — green algae, Heterokontophyta — brown algae and Rhodophyta — red algae). Significant differences were observed for the three performed tests (Mann–Whitney U tests; p = b0.001) for each algae taxon. The difference in the median fraction of light values was larger than would be expected by chance between the presence and absence within each taxon. The estimated 1st quartile (P25%) through the bootstrap technique indicated that 75% of green and brown algae were located in areas with more than 4.63% (CI: 4.62–4.65%) and 4.05% (CI: 4.04–4.06%) of surface light reaching the seabed. Red algae were on the other hand located in areas with more than 3.65% (CI: 3.64–3.66%) of surface light reaching the seabed (Fig. 4). The computed bias for both seaweed taxa indicates an adequate accurate model being less than 2% in relation to the estimated first quartile value (green algae: − 0.086; red algae: 0.068; brown algae: 0.003). Both thresholds are located between 20 and 30 meter depth, being closer to the shore on the West side coast. The biological threshold between the infralittoral and the circalittoral is highlighted in Fig. 5 using the minimum value and first quartile for red algae (0.80 and 3.65%).
0.6099 N/m2, with a mean value of 0.0011 N/m2. Values of kinetic wave energy observed for each kelp species indicated a decrease of the average value from Saccorhiza polyschides to Laminaria hyperborea and Laminaria ochroleuca, with the same arising in range values. S. polyschides was observed in coastal areas with values of energy between 0.00005 and 0.19916 N/m2 (mean ± SD = 002747 ± 0.02949 N/m2; median = 0.02756 N/m2; range = 0.19911 N/m2) (Table 1). Estimated percentile analysis through the bootstrap technique (12.5% and 87.5%) on data for S. polyschides pointed out that the energy class was between 0.00303 and 0.04385 N/m2 (Table 1). L. hyperborea, a species that is unable to survive in extreme wave action conditions, was observed in places with more than 0.00611N/m2 and less than 0.04918 N/m2 (Table 1). Computed bias for both thresholds indicates acceptable accurate model simulation, being 13.0% and 0.7% for the percentiles 12.5% and 87.5% respectively. The ‘moderate energy’ class assigned to the presence of S. polyschides is pointed out in the histogram using original data analysis (Fig. 6a) and in the boxplot after applying the bootstrap technique (Fig. 6b). Fig. 7 shows the distribution map of energy classes set using S. polyschides for the entire Iberian West coast. According to the established classes, the ‘high energy’ class is mainly observed in the North coast of Portugal while the Algarve South coast is mainly covered by areas with low energy levels and in minor percentage with the ‘moderate energy’ levels.
3.2. Wave energy threshold
The EUNIS habitat classification scheme which relies on thresholds is based on few environmental variables that mainly structure the biological communities (Cameron and Askew, 2011). Accurate and validated thresholds (e.g. biological zonation and coastal energy) to
Predicted percentile values (p90%) of kinetic wave-induced energy for the Portuguese mainland coast ranged from 0 to a maximum of
4. Discussion
Fig. 5. Biological threshold between the infralittoral and the circalittoral zones in the Algarve study area (a) using different fractions of light: b) minimum observed value (0.80%) and c) 1st quartile (P25%) through the bootstrap technique (3.65%) for red algae species.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
7
Table 1 Kinetic wave energy (N/m2) values observed (original) with the presence of kelp species and main results through the bootstrap technique. S. polyschides Bootstrap Mean Standard error − confidence (95%) + confidence (95%) BIAS Original data Max Min Range Mean n samples
12.5% 0.00303 0.00142 0.00295 0.00312 0.00039
L. hyperborea 87.5% 0.04385 0.00147 0.04376 0.04394 0.00032
0.19916 0.00005 0.19911 0.02747 143
implement EUNIS habitat mapping are therefore desirable. The regional specificity and low level of agreement between available studies reinforce the need to compute and establish the thresholds obtained by the present study for the Western Atlantic Iberian coast: subtidal biological zonation and coastal energy level thresholds. The sub-littoral seabed is in general a heterogeneous environment that supports a wide biodiversity. Broad scale patterns of zonation have been mostly described qualitatively and some generalizations on the organization of marine communities were established (e.g. Pérès, 1967; Pérès and Picard, 1964; Saldanha, 1995). Those generalizations have stated that changes in depth and in environmental factors such as light availability are important physical reasons for the zonation of coastal communities. Typically sub-littoral ecological zones and boundaries are obtained through the characterization of the vertical distribution of certain indicator species, namely the abundance and frequency of macrophytes and algae, and sessile fauna such as sponges and ascidians which are less tolerant to light (Logan et al., 1984). The infralittoral zone is characterized as a submerged zone that extends to the compatible limit of marine phanerogam or photophile species. In contrary, the circalittoral zone is
12.5% 0.00611 0.00219 0.00597 0.00624 −0.00025
0.07299 0.00281 0.07017 0.02326 10
L. ochroleuca 87.5% 0.04918 0.01581 0.04820 0.05016 0.00019
12.5% 0.00005 0.00073 0.00000 0.00009 0.00010
87.5% 0.03087 0.00616 0.03049 0.03125 −0.00312
0.03335 0.00000 0.03335 0.01046 23
characterized by the presence of certain sciaphilic algae species and abundant fauna communities (e.g. Pérès, 1967; Pérès and Picard, 1964). The present study emphasizes the importance of the light fraction that reaches the sea bottom as a significant factor for algae zonation and consequently to other sea organisms. The results point out that the transition between the infralittoral and circalittoral zones might occur when no more than 3.65% of light reaches the seabed. This statement was based on the 1st quartile of the ordered data which was used as indicative value. This indicative value suggests that 75% of red algae are located in areas in which more than 3.65% of surface light reaches the seabed. The value of 2.36% of surface incident light was obtained from the MESH study area through the same indicative quartile (1st quartile) (Coltman et al., 2008). The UKSeaMap project results indicate that the transition between the circalittoral and the infralittoral zones occurs when less than 1% of light reaches the seabed (McBreen and Askew, 2011). This difference has been attributed to the improved resolution of light data (from 9 to 4 km) and increased number of biological records in the UKSeaMap 2010 analysis (McBreen and Askew, 2011). McBreen and Askew (2011) in the UKSeaMap 2010 project also defined the hard threshold to be close to 1% of light reaching the seabed
Fig. 6. Histogram and boxplot presenting bootstrap statistics for kinetic energy assigned to the presence of Saccorhiza polyschides.
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
8
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
which was defined by the 1st quartile, above which 75% of kelp biotopes were present. Using a slightly different approach the computed value was around 4.5% for the North and Celtic Sea in the EUSeaMap project (Cameron and Askew, 2011). All these disagreements highlight the fact that, although one study may suggest one threshold value, it cannot be assumed that it should be applied to other datasets or regions without a new analysis being undertaken. The minimum value obtained in the present study point out that both brown and red algae are still present with at least more than 0.80% of light reaching the seabed. This value seems to be the limit for the development of most sciaphilic algae species, which are mostly adapted to live in low light environments. The minimum light value of 0.80% for the presence of brown and red algae is closer to the value of 1% reported in the literature for other European regions and used on the UKSeaMap project. The indicative value of 3.65% seems very realistic and closer to the 20 to 24 meter depth given by the Portuguese literature (e.g. Saldanha, 1995) as the boundary between the infralittoral and circalittoral zones. Nevertheless, being the minimum value acceptable, within the MeshAtlantic framework it was decided to use the minimum light value of 0.80% (rounded to 1%) in order to
standardize in the European level the broad scale habitat maps produced (Vasquez et al., in press). Differences between the present study and others that have been carried out in the European coast might be induced by different indicator species used (e.g. seagrass, kelp and other macroalgae), habitat disparity or expected differences in environmental factors (e.g. temperature, turbidity and salinity). Among these limitations, there are also different sources of light attenuation and higher resolution that should be considered in future attempts to establish thresholds which could provide finer tuning in small areas (Ellwood et al., 2011). Further studies are needed because the light is only one of the various environmental factors affecting the presence of algae. The hydrodynamic energy regime is also an important environmental factor, playing a significant role in the sedimentary cover and on benthic species settlement, therefore contributing to structure the biological communities. Thus, as previously mentioned, it is carefully considered in the EUNIS classification scheme, being explicit on the rocky area habitats (Connor et al., 1997a, 1997b, 2003, 2004; Davies et al., 2004). Given the available energy model our approach characterized ‘moderate energy’ for the Iberian Atlantic coast between 0.00303
Fig. 7. EUNIS energy classes established based on Saccorhiza polyschides spatial distribution and kinetic wave energy (N/m2).
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
N/m2 and 0.04385 N/m2. After mapping the results for the entire Iberian coast, it was observed that this is only roughly in agreement with the discrete thresholds already established in published literature (Bettencourt et al., 2004). According to these authors the Portuguese coast is broadly divided in three different hydrodynamic regime areas. The energy of the coast rises according to Bettencourt et al. (2004), from the South to the North, being sheltered from the South Portuguese border to Lagos, moderately exposed from Lagos to Cape Carvoeiro and exposed from Cape Carvoeiro to the Northern Portuguese border. The results obtained in the present paper showed that the trend should be much less discrete, conditioned by bathymetry characteristics and orientation of the coast. As shown, S. polyschides could be used to establish energy thresholds for the West Iberian coast since the values of kinetic wave energy that the species can endure, along with its wide distribution along the mainland Portuguese coast seem to specify this kelp species as a suitable indicator species of moderate class energy, allowing the definition of two additional EUNIS energy classes (high and low energy classes). Concerning the energy classes, our approach with S. polyschides gave more realistic results to create wave energy classes when compared to pilot studies that establish areas for energy production in mainland Portugal (e.g. POEM, 2014). We have shown that the identified energy classes have a more scattered distribution along the Portuguese coast. However, the higher energy level was predominantly observed on the North while the moderate and low energy levels were more common in the southern coast. As expected, the computed thresholds also differed from those attained by the previous EU funded projects (1.8– 4.0 N/m2 in MESH and 0.21–1.2 N/m2 in EUSeaMap) mainly because they are based on different models. The difference of range among studies highlights the fact that it cannot be applied to other regions without a new analysis being undertaken. 5. Conclusions This study allowed the computation of appropriate thresholds on biological zonation and wave energy for the EUNIS habitat classification of the Western Atlantic Iberian coast. For that purpose, it has established a link between the spatial distributions of benthic macroalgae, light penetration and kinetic wave energy. Furthermore, this study highlighted the importance of estimating the variability associated with these thresholds, mainly at a regional or national scale, and the need of additional base studies on the Atlantic Iberian coast. In order to improve the threshold on sea energy the integration of current energy in the analysis is important. The study highlights the need for an integrated work in a European scale which can contribute for a better standardization of both EUNIS thresholds studied here. However, the present study will be very useful in the near future to implement and establish the West Iberian marine EUNIS habitats inventory. Acknowledgments This work was made possible by the availability of biological historical data from a previous research project, Rensub (CCMAR-ARH Algarve: ‘Cartography and characterization of the marine communities off the National Underwater Ecological Reserve’). We would like to thanks Jacques Populus and Mickael Vasquez from the IFREMER (Institut Français de Recherche pour L'Exploitation de la Mer: http://wwz.ifremer.fr/institut) for providing the coefficient photosynthetically available radiation (KdPAR) layer. We would like to acknowledge professor Ramiro Neves and its Research team (MARETEC — http://www.maretec.org/) from the Instituto Superior Técnico (IST) the Engineering School of Lisbon Technical University for providing us with data of wave and currents. We also would like to kindly acknowledge our colleagues Alexandra Cunha and Jorge Assis from CCMAR, University of Algarve for providing data on the presence of kelp species at the Portuguese coast (FindKelp
9
project). Finally special thanks to all the MeshAtlantic partners for the opportunity to discuss this subject. We are grateful for the comments and suggestions of two referees who contributed greatly to improving the manuscript.
References Allee, R.J., Dethier, M., Brown, D., Deegan, L., Ford, G.R., Hourigan, T.R., Maragos, J., Schoch, C., Sealy, K., Twilley, R., Weinstein, M.P., Yoklavich, M., 2000. Marine and estuarine ecosystem and habitat classification. NOAA Technical Memorandum, NMFS-F/SPO, Silver Spring, Maryland (43 pp.). Assis, J., Tavares, D., Tavares, J., Cunha, A., Alberto, F., Serrão, E.A., 2009. Findkelp, a GISbased community participation project to assess Portuguese kelp conservation status. J. Coast. Res. Special Issue 56, 1469–1473. Ballesta, L., Pergent, G., Pasqualini, V., Pergent-Martini, C., 2000. Distribution and dynamics of Posidonia oceanica beds along the Albères coastline. Life Sci. 323, 407–414. Bettencourt, A.M., Bricker, S.B., Ferreira, J.G., Franco, A., Marques, J.C., Melo, J.J., Nobre, A., Ramos, L., Reis, C.S., Salas, F., Silva, M.C., Simas, T., Wolff, W., 2004. Typology and reference conditions for Portuguese transitional and coastal waters. Development of Guidelines for the Application of the European Union Water Framework Directive. INAG/IMAR, Lisboa, Portugal (100 pp.). Birkett, D.A., Maggs, C.A., Dring, M.J., Boaden, P.J.S., Seed, R., 1998. Infralittoral reef biotopes with kelp species (volume VII). An overview of dynamic and sensitivity characteristics for conservation management of marine SACs. Scottish Association of Marine Science (UK Marine SACs Project) (174 pp.). Butler, A., Harris, P.T., Lyne, V., Heap, A., Passlow, V.L., Smith, R.N.P., 2001. An interim, draft bioregionalisation for the continental slope and deeper waters of the south-east marine region of Australia. Report to the National Oceans Office, Geoscience Australia and CSIRO, Hobart (35 pp.). Cameron, A., Askew, N. (Eds.), 2011. EUSeaMap — Preparatory Action for Development and Assessment of a European Broad-scale Seabed Habitat Map Final Report (226 pp. Available at http://jncc.defra.gov.uk/euseamap). Coltman, N., Golding, N., Verling, E., 2008. Developing a broadscale predictive EUNIS habitat map for the MESH study area. Available at, http://www.searchmesh.net/ pdf/MESH%20EUNIS%20model.pdf (16 pp.). Connor, D.W., Hiscock, K., Foster-Smith, R.L., Covey, R., 1995. A classification system for benthic marine biotopes. In: Eleftheriou, A., Ansell, A.D., Smith, C.J. (Eds.), Biology and Ecology of Shallow Coastal Waters. Proceedings of the 28th European Marine Biology Symposium. Olsen and Olsen, Fredensborg, pp. 155–165. Connor, D.W., Brazier, D.P., Hill, T.O., Northen, K.O., 1997a. Marine Nature Conservation Review: marine biotope classification for Britain and Ireland. Volume 1. Littoral biotopes. Version 97.06. JNCC Report, No. 229. Connor, D.W., Dalkin, M.J., Hill, T.O., Holt, R.H.F., Sanderson, W.G., 1997b. Marine Nature Conservation Review: marine biotope classification for Britain and Ireland. Volume 2. Sublittoral biotopes. Version 97.06. JNCC Report, No. 230. Connor, D.W., Allen, J.H., Golding, N., Lieberknecht, L.M., Northen, K.O., Reker, J.B., 2003. The national marine habitat classification for Britain and Ireland. Version 03.02. Introductory text. Joint Nature Conservation Committee, Peterborough (42 pp. Available at www.jncc.defra.gov.uk/marinehabitatclassification Accessed: July 2014). Connor, D., Allen, J.H., Golding, N., Howell, K.L., Lieberknecht, L.M., Northen, K.O., Reker, J., 2004. The Marine Habitat Classification for Britain and Ireland. Version 04.05. Introduction. Joint Nature Conservation Committee, Peterborough (49 pp. Available at http://jncc.defra.gov.uk/MarineHabitatClassification). Costello, M.J., 2009. Distinguishing marine habitat classification concepts for ecological data management. Mar. Ecol. Prog. Ser. 397, 253–268. http://dx.doi.org/10.3354/ meps08317. Davies, C.E., Moss, D., Hill, M.O., 2004. EUNIS habitat classification revised 2004. Report to the European Topic Centre on Nature Protection and Biodiversity. European Environment Agency (307 pp. Available at http://www.eunis.eea.europa.eu/habitatscodebrowser.jsp). Devillers, P., Devillers-Terschuren, J., 1996. A classification of Palaearctic habitats. Nature and Environment, No 78. Council of Europe, Strasbourg. Devillers, P., Devillers-Terschuren, J., Ledant, J.P., 1991. CORINE Biotopes Manual vol. 2. Habitats of the European Community. Office for Official Publications of the European Communities, Luxembourg. Ellwood, H.N., Askew, N.A., Cameron, J., McBreen, F., 2011. UKSeaMap 2010. Marine Ecological Surveys Limited (MESL): Technical Report 7. Joint Nature Conservation Committee, Peterborough (23 pp.). Evans, D., 2012. The EUNIS habitats classification — past, present and future. Rev. Investig. Mar. 19 (2), 28–29. Galparsoro, I., Connor, D.W., Borja, A., Aish, A., Amorim, P., Bajjouk, T., Chambers, C., Coggan, R., Dirberg, G., Ellwood, H., et al., 2012. Using EUNIS habitat classification for benthic mapping in European seas: present concerns and future needs. Mar. Pollut. Bull. 64 (12), 2630–2638. Gonçalves, J.M.S., Monteiro, P., Coelho, R., Afonso, C., Ribeiro, J., Almeida, C., Veiga, P., Machado, D., Berecibar, E., Oliveira, F., Bentes, L., 2004. Mapeamento de biocenoses marinhas da Reserva Ecológica Nacional Submarina entre Albufeira e Vale do Lobo. CCDR Algarve, Universidade do Algarve, CCMAR, Faro (182 pp.). Gonçalves, J.M.S., Monteiro, P., Coelho, R., Afonso, C., Almeida, C., Veiga, P., Machado, M., Machado, D., Oliveira, F., Ribeiro, J., Abecasis, D., Primo, L., Tavares, D., FernándezCarvalho, J., Abreu, S., Fonseca, L., Erzini, K., Bentes, L., 2007. Cartografia e caracterização das biocenoses marinhas da Reserva Ecológica Nacional Submarina
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007
10
P. Monteiro et al. / Journal of Sea Research xxx (2014) xxx–xxx
entre a Galé e a barra Nova do Ancão. Relatório Final. CCDR Algarve, Universidade do Algarve, CCMAR, Faro (250 pp. + anexos). Gonçalves, J.M.S., Monteiro, P., Afonso, C., Almeida, C., Oliveira, F., Rangel, M., Ribeiro, J., Machado, M., Veiga, P., Abecasis, D., Pires, F., Fonseca, L., Erzini, K., Bentes, L., 2008. Cartografia e caracterização das biocenoses marinhas da Reserva Ecológica Nacional Submarina entre a Galé e a foz do rio Arade. Relatório Final CCDR Algarve. Universidade do Algarve, CCMAR, Faro (144 pp. + Anexos). Gonçalves, J.M.S., Monteiro, P., Afonso, C., Oliveira, F., Rangel, M., Machado, M., Veiga, P., Leite, L., Sousa, I., Bentes, L., Fonseca, L., Erzini, K., 2010. Cartography and characterization of the marine communities off the National Underwater Ecological Reserve between the Arade River and Ponta da Piedade. Final Report. ARH Algarve. University of Algarve, CCMAR, Faro (122 pp. + Anexos). Hamdi, A., Vasquez, M., Populus, J.Y., 2010. Cartographie des habitats physiques Eunis — Côtes de France. Convention Ifremer/AAMP. no 09/12177764/FY. Available at. http://archimer.ifremer.fr/doc/00026/13751/. Haukoos, J.S., Lewis, R.J., 2005. Advanced statistics: bootstrapping confidence intervals for statistics with “difficult” distributions. Acad. Emerg. Med. 12 (4), 360–365. http://dx. doi.org/10.1197/j.aem.2004.11.018. Hiscock, K., Connor, D.W., 1991. Benthic marine habitats and communities in Great Britain: the development of an MNCR classification. JNCC Report 6; Marine Nature Conservation Review Report No. MNCR/OR/14. Joint Nature Conservation Committee, Peterborough (92 pp.). Howell, K.L., 2010. A benthic classification system to aid in the implementation of marine protected area networks in the deep/high seas of the NE Atlantic. Biol. Conserv. 143, 1041–1056. ICES, 2001. Report of the Working Group on Marine Habitat Mapping (WGMHM), 3–6 April 200, Galway, Ireland, ICES CM 2000/E/08 (42 pp.). Logan, A., Page, F.H., Thomas, M.L.H., 1984. Depth zonation of epibenthos on sublittoral hard substrates off Deer Island, Bay of Fundy, Canada. Estuar. Coast. Shelf Sci. 18 (5), 571–592. http://dx.doi.org/10.1016/0272-7714(84) 90091-X. McBreen, F., Askew, N., 2011. Light penetration layer. UKSeaMap 2010. Technical Report 2. JNCC, Peterborough (7 pp.). McBreen, F., Askew, N., Cameron, A., 2011a. UKSeaMap 2010. Technical Report 4: Energy, JNCC Report (38 pp.). McBreen, F., Askew, N., Cameron, A., Connor, D., Ellwood, H., Carter, A., 2011b. UKSeaMap 2010: predictive mapping of seabed habitats in UK waters. JNCC Report, No. 446 (103 pp.). Morel, A., Huot, Y., Gentili, B., Werdell, P.J., Hooker, S.B., Franz, B.A., 2007. Examining the consistency of products derived from various ocean color sensors in open ocean (Case 1) waters in the perspective of a multi-sensor approach. Remote Sens. Environ. 111, 69–88.
Mumford, T.F., 2007. Kelp and eelgrass in Puget Sound. Technical report 2007-05. Prepared in Support of the Puget Sound Nearshore Partnership. Washington Department of Natural Resources, Aquatic Resources Division (27 pp.). Olenin, S., Ducrotoy, J.P., 2006. The concept of biotope in marine ecology and coastal management. Mar. Pollut. Bull. 53, 20–29. Pérès, J.M., 1967. The Mediterranean benthos. Oceanogr. Mar. Biol. Annu. Rev. 5, 449–533. Pérès, J.M., Picard, J., 1964. Nouveau manuel de bionomie benthique de la mer Mediterranée. Reel Trav. Stat. Mar. Endoume 31 (47), 1–137. POEM, 2014. Portuguese Marine Spatial Planning. Available at, http://poem.inag.pt/ (Accessed on January, 29th). R Development Core Team, 2008. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria 3-900051-07-0 (URL http://www.R-project.org). Roberts, J.J., Best, B.D., Dunn, C.D., Treml, E.A., Halpin, P.N., 2010. Marine Geospatial Ecology Tools: an integrated framework for ecological geoprocessing with ArcGIS, Python, R, MATLAB, and C++. Environ. Model. Softw. 25, 1197–1207. http://dx.doi. org/10.1016/j.envsoft.2010.03.029. Saldanha, L., 1995. Fauna Submarina Atlântica. Portugal continental, Açores, Madeira. Publicações Europa América (364 pp.). SANBI, 2009. Further development of a proposed national wetland classification system for South Africa. Primary Project Report. Prepared by the Freshwater Consulting Group (FCG) for the South African National Biodiversity Institute (SANBI) (74 pp.). Saulquin, B., Hamdi, A., Gohin, F., Populus, J., Mangin, A., d' Andon, O.F., 2013. Estimation of the diffuse attenuation coefficient KdPAR using MERIS and application to seabed habitat mapping. Remote Sens. Environ. 128, 224–233. http://dx.doi.org/10.1016/j. rse.2012.10.002. Snelder, T.H., Leathwick, J.R., Dey, K.L., Rowden, A.A., Weatherhead, M.A., Fenwick, G.D., Francis, M.P., Gorman, R.M., Grieve, J.M., Hadfield, M.G., Hewitt, J.E., Richardson, K.M., Uddstrom, M.J., Zeldis, J.R., 2006. Development of an ecological marine classification in the New Zealand region. Environ. Manag. 39 (1), 12–29. Vasquez, M., Populus, J., Mata Chacón, D., Sanz Alonso, J.L., Tempera, F., Gonçalves, J.M.S., O'Keeffe, E., Galparsoro, I., Bentes, L., Amorim, P., Henriques, V., McGrath, F., Monteiro, P., Mendes, B., Freitas, R., Martins, R., 2014. Mapping seabed habitats of the north-east Atlantic using environmental data. J. Sea Res http://dx.doi.org/10.1016/j.seares.2014. 09.011 (in press). Zacharias, M.A., Howes, D.E., Harper, J.R., Wainwright, P., 1998. The British Columbia marine ecosystem classification: rationale, development, and verification. Coast. Manag. 26, 105–124. Zar, J.H., 1999. Biostatistical Analysis. 4th edn. Prentice-Hall, Upper Saddle River, NJ (663 pp.).
Please cite this article as: Monteiro, P., et al., EUNIS habitat's thresholds for the Western coast of the Iberian Peninsula — A Portuguese case study, J. Sea Res. (2014), http://dx.doi.org/10.1016/j.seares.2014.11.007