Consistent patterns of variation in macrobenthic assemblages and environmental variables over multiple spatial scales using taxonomic and functional approaches

Marine Environmental Research 120 (2016) 191e201

Contents lists available at ScienceDirect

Marine Environmental Research journal homepage: www.elsevier.com/locate/marenvrev

Consistent patterns of variation in macrobenthic assemblages and environmental variables over multiple spatial scales using taxonomic and functional approaches Puri Veiga a, b, c, *, Ana Catarina Torres a, b, Fernando Aneiros d, e, Isabel Sousa-Pinto a, b, Jesús S. Troncoso d, e, Marcos Rubal a, b, d a Laboratory of Coastal Biodiversity, Interdisciplinary Centre of Marine and Environmental Research (CIIMAR/CIMAR), University of Porto, Rua dos Bragas 289, P 4050-123 Porto, Portugal b Department of Biology, Faculty of Sciences, University of Porto, Rua do Campo Alegre s/n 4150-181 Porto, Portugal c
n de Bioloxía Marin ~ a da Gran ~ a, Universidade de Santiago de Compostela, Casa do Ho
rreo, Rúa da Ribeira 1, 15590, A Gran ~ a, Ferrol, Spain Estacio d Departamento de Ecoloxía e Bioloxía Animal, Facultade de Ciencias do Mar, Universidade de Vigo, Campus Universitario Lagoas-Marcosende, 36310 Vigo, Spain e
n de Ciencias Marin ~ as de Toralla, Universidade de Vigo, Illa de Toralla, 36331 Vigo, Spain ECIMAT, Estacio

a r t i c l e i n f o

a b s t r a c t

Article history: Received 6 June 2016 Received in revised form 17 August 2016 Accepted 23 August 2016 Available online 27 August 2016

Spatial variability of environmental factors and macrobenthos, using species and functional groups, was examined over the same scales (100s of cm to >100 km) in intertidal sediments of two transitional water systems. The objectives were to test if functional groups were a good species surrogate and explore the relationship between environmental variables and macrobenthos. Environmental variables, diversity and the multivariate assemblage structure showed the highest variability at the scale of 10s of km. However, abundance was more variable at 10s of m. Consistent patterns were achieved using species and functional groups therefore, these may be a good species surrogate. Total carbon, salinity and silt/clay were the strongest correlated with macrobenthic assemblages. Results are valuable for design and interpretation of future monitoring programs including detection of anthropogenic disturbances in transitional systems and propose improvements in environmental variable sampling to refine the assessment of their relationship with biological data across spatial scales. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Benthic ecology Macrobenthic assemblages Taxonomic composition Functional groups Environmental variables Variability Spatial scale Nested design Transitional water systems Northeast Atlantic coast

1. Introduction In most ecological systems, the composition of assemblages is the result of complex interactions between abiotic and biotic factors (Cisneros et al., 2011; Kraufvelin et al., 2011; Kraan et al., 2015). Moreover, natural assemblages are complex and integrally variable in space and time (e.g. Ysebaert and Herman, 2002; Fraschetti et al., 2005; Rubal et al., 2014). Dealing with this natural variability has been challenging when trying to understand ecological processes

* Corresponding author. Laboratory of Coastal Biodiversity, Interdisciplinary Centre of Marine and Environmental Research (CIIMAR/CIMAR), University of Porto, Rua dos Bragas 289, P 4050-123 Porto, Portugal. E-mail address: [email protected] (P. Veiga). http://dx.doi.org/10.1016/j.marenvres.2016.08.011 0141-1136/© 2016 Elsevier Ltd. All rights reserved.

shaping the abundance and distribution of organisms (McGill, 2010). However, research has progressed from considering spatial variation of assemblages as ‘‘noise’’ to understanding that its knowledge is crucial because scales at which assemblages vary are also likely to be the scales at which ecological processes have great effects (Underwood et al., 2000; Kraufvelin et al., 2011; Valdivia et al., 2011). Therefore, the quantitative description and observation of patterns across a range of spatial scales is an essential step before explanatory models for patterns in assemblage structure can be proposed (Underwood et al., 2000; Hewitt et al., 2007). Indeed, they provide valuable insights about the mechanisms that probably strengthen diversity and affect the distribution of communities, supporting the development of conservation or environmental management strategies (McGill, 2010; Kraan et al., 2015).

192

P. Veiga et al. / Marine Environmental Research 120 (2016) 191e201

Insufficient knowledge of the scales at which relevant ecological processes act is a limitation to improve our understanding of biodiversity and ecosystem functioning and their underlying processes and often results in poor decision-making and environmental policy (Raffaelli and Friedlander, 2012; Yaffee, 1997). Studies focused on determining spatial variability of assemblages have received wide attention in rocky intertidal systems (e.g. Valdivia et al., 2011; Veiga et al., 2013; Díaz et al., 2015). The general picture emerging from these studies is that variability is larger at smaller spatial scales (i.e. cm or meters), which seems to be pervasive in marine systems (see review by Fraschetti et al., 2005). In soft bottom ecosystems, studies on spatial variation of macrobenthos (e.g. abundance, diversity, species composition) have often focused on large scales (i.e. 100 s m to km), examining patterns apparently governed by conspicuous environmental gradients such as those defined by tidal regime or wave exposition (e.g. De la Huz and Lastra, 2008; Schlacher and Thompson, 2013; Veiga et al., 2014). Smaller scales (i.e. cm or meters) have, however, not been usually incorporated into large-scale studies (but see Olabarria and Chapman, 2001; Chapman and Tolhurst, 2004, 2007; Chapman et al., 2010), and thus losing information about variability at these scales, because in many cases macrobenthic replicates are pooled (e.g. Jaramillo et al., 1995; Schlacher and Thompson, 2013). However, many ecological processes and environmental features, such as sedimentary properties, hydrodynamics or bioturbation, change at smaller scales (mm to < 1 m), and can influence the distribution of fauna within large patchy habitats (e.g. Sun et al., 1993; Passarelli et al., 2012; Jungerstam et al., 2014; Díaz et al., 2015). This is especially evident in transitional water systems (i.e. estuaries, fjords, lagoons and rías) and archipelagos, which are highly dynamic in their physical-chemical and hydro-morphologic features, resulting in a mosaic of habitats at relatively short distances (i.e. <100 m; Sigala et al., 2012; Díaz et al., 2015). The implementation of nested hierarchical designs that estimate variance components allows the examination of variability, both in univariate and multivariate contexts, at a range of spatial scales, from the sampling unit to geographical areas (Underwood, 1997; Underwood and Chapman, 1996, 1998a, b; Terlizzi et al., 2005; Kraufvelin et al., 2011). Analysis of variance tests the presence of significant variability at each scale but estimates of variance components also let us quantify the magnitude of variation for each scale independently of other scales (Morrisey et al., 1992). Hierarchical sampling designs are being increasingly used for studies of sedimentary systems, elucidating that abundance and diversity of macrobenthos show indeed considerable variability at different spatial scales (e.g. Morrisey et al., 1992; Ysebaert and Herman, 2002; Chapman and Tolhurst, 2004, 2007). However, many studies that have explored the relationship between environmental variables and benthic assemblages were based on environmental measures recorded from less sampling points compared with those for the fauna (e.g. De la Huz and Lastra, 2008; Kraufvelin et al., 2011; Veiga et al., 2011). This unbalanced design between environmental and biological data induces some error when an important amount of variability from biological data (different replicates) is linked to a single environmental measure (Kraufvelin et al., 2011). Despite this clear limitation, there are still few studies in sedimentary systems that examine multiple spatial scales of variability of both environmental variables and fauna (Chapman et al., 2010). In any case, the available literature has shown contradictory results. Some of them pointed out that physical features of habitats are sufficient to explain general patterns of benthos (Edgar and Barret, 2002; Ysebaert and Herman, 2002). On the contrary, others reported that those patterns are not clearly correlated with variation in environmental and sedimentary properties of the habitat (Tolhurst and Chapman, 2007; Chapman

and Tolhurst, 2004, 2007; Chapman et al., 2010). Therefore, the link between spatial variation in environmental factors and biological patterns of soft bottom benthic assemblages is still poorly understood (Edgar and Barret, 2002; Ysebaert and Herman, 2002). Macrobenthos plays an important role influencing the structure and functioning of ecosystems in transitional systems (Pratt et al., 2014). Most species display a sedentary lifestyle, intermediate trophic level positions, relatively long life-span, varying responses to changes in environmental conditions and importance in nutrient recycling. Because of that, macrobenthos may serve as an effective useful indicator of the ecological status of transitional water systems (Dauvin, 2007). In fact, macrobenthos is one of the biological elements described by the European Water Framework Directive (WDF, 2000/60/EC) to be used in defining ecological quality status in a water body. Therefore, macrobenthos has been a key element of many monitoring programmes. However, these programmes often have design issues. Firstly, they usually do not explicitly take into account distribution patterns at different spatial scales (Ysebaert and Herman, 2002). Secondly, most of them are mainly based on taxonomic composition and relative abundance of macrobenthic species. Nevertheless, the use of functional diversity (i.e. the diversity and range of functional traits present in the fauna of an ecosystem) is advocated because it is likely an imperative element of biodiversity for the ecosystem functioning (Hooper et al., 2005; Wright et al., 2006). In rocky shores, functional groups of macroalgae have been widely used, showing consistent spatial and temporal patterns with those found using species (e.g. Smale, 2010; Rubal et al., 2011; Veiga et al., 2013). However, whether this consistence occurs or not for macrobenthos in soft bottoms is yet poorly explored. The first aim of this study was to detect and quantify significant spatial scales of variability in environmental variables (i.e. coastal water and sediment features) and macrobenthic assemblages in intertidal sediments of transitional water systems at scales ranging from 100s of cm to >100 km. These data were used to test the hypotheses that (1) spatial variability of macrobenthic assemblages obtained by using taxonomic composition would be consistent with that resulting considering functional groups and (2) environmental variables with patterns of variation similar to those of macrobenthos will be highly correlated with its structure. 2. Materials and methods 2.1. Study area This study was done in the intertidal area of two transitional water systems: Ria de Aveiro and Ría de Vigo (Fig. 1). The Ría de Vigo is a partially stratified estuary on the NW coast of Spain. The combined effect of temperature and salinity results in the development of a strong stratification that leds to a positive circulation
pez et al., 2001). Water residence time varies from a few (Torres-Lo days, during upwelling or strong rainy events in winter, to five weeks in the downwelling season (Prego and Fraga, 1992). It shows a funnel-like morphology with its central axis lying in a SW-NE direction, with the Cíes islands at the entrance, acting as a shelter against waves. The Ria de Aveiro is located on the northwest coast of Portugal and connected to the Atlantic Ocean through an artificial inlet. It is classified as a bar-built estuary (Pritchard, 1967). It shows a rather complex topography with three main channels that radiate from the mouth with several branches, islands and mudflats resulting in extensive intertidal sand and mud flats (Dias et al., 1999). It has a length of 45 km and a maximum width of 10 km, thus covering an area of 43e47 km2 at low and high tide, respectively. The hydrological circulation is dominated by marine influence.

P. Veiga et al. / Marine Environmental Research 120 (2016) 191e201

193

multiparametric prove (Hach® HQ40). 2.3. Sampling processing

Fig. 1. Map showing the two studied systems: Vigo and Aveiro. Black stars indicate the four sampling localities.

Both systems are subjected to human perturbations such as urban activities (sewage runoff), harvesting, mainly focused on molluscs, and industrial discharges (Barroso et al., 2000; Prego et al., 2008). At each system, two localities were selected along the natural environmental gradient (e.g. salinity, wave exposure) present in both systems (Fig. 1).

2.2. Sampling Sampling was done during low spring tides in April 2014 following a fully nested hierarchical sampling design. The largest spatial scale examined was the system, which included two levels: Aveiro and Vigo, about 170 km apart (Fig. 1). Within each system, based on their size and the distance between both systems, two localities were randomly chosen separated by 10s of km (one order of magnitude smaller than the distance between systems). Similarly, within each locality, three plots (separated by 100s of m) were haphazardly chosen. At each plot, three replicate samples of macrobenthos were randomly collected 100s of cm apart with a 0.0113 m2 PVC core to a depth of 30 cm. Samples were immediately sieved though a 0.5-mm mesh bag, collected in a labelled plastic bag and preserved in 4% neutralised formaldehyde solution with Rose Bengal. Additionally, sediment samples were collected at the same plots as fauna (three replicates per plot), using a 0.001 m2 core to a depth of 10 cm, to analyse total carbon (TC), inorganic carbon (IC), total organic carbon (TOC), water content and grain size composition of the sediment. Moreover, three independent measures of temperature (both in the sediment and in water), salinity and oxygen were obtained from the coastal water by each plot with a

Macrofauna was sorted, identified to the lowest possible taxon (usually species level) and counted. Then, each species was assigned to a functional group according to a combination of biological traits related to their morphology, life habits and trophic guild (Table 1). Compiling reliable information on biological traits of benthic macrofauna is a time-consuming task, so the selection of the traits to be used must be based on a trade-off between their suitability for the aim of the study and the time and effort required to compile enough information on them for the taxa studied (Bremner et al., 2006). Considering this, and given that functional classification is used here for comparative purposes in terms of distribution patterns, not for addressing ecosystem functioning, the biological traits used were selected both for pragmatic and informative reasons: (1) a broad range of reliable information on these traits is available for the taxa studied and (2) they are widely used in the scientific literature for different kinds of functional ecology studies (e.g. Bremner et al., 2006; Hewitt et al., 2008; Paganelli et al., 2012). Moreover, the use of functional groups including several biological traits simplifies the functional approach and allows us to use the same statistical analyses as for taxonomic classification, thus enabling a direct comparison between both approaches (i.e. species and functional groups). The percentage of sediment water content (WC) was estimated from the difference between wet and dry weights of sediment; the latter measured after drying at 60
C up to constant weight. In order to study the granulometric composition of the sample, it was first wet-sieved using a 2-mm mesh. Coarse fraction (>2 mm) was dried and then sieved to sort the >4 mm and 2e4 mm fractions. Fine fraction (<2 mm) was dispersed by adding a solution containing sodium hexametaphosphate and sodium carbonate and then stirring for 2 h. After that, it was analysed by means of a Beckman Coulter LS 13 320 laser diffraction particle size analyser. The following sedimentary fractions were considered: coarse gravel (>4 mm), fine gravel (2e4 mm), very coarse sand (1e2 mm), coarse sand (0.5e1 mm), medium sand (0.25e0.5 mm), fine sand (0.125e0.25 mm), very fine sand (0.063e0.125 mm) and silt/clay (<0.063 mm). Using a previously dried and powdered portion of the sample, total carbon (TC), inorganic carbon (IC) and total organic carbon (TOC) were measured by means of a LECO CNS-2000 elemental macro analyser. 2.4. Data analyses All univariate and multivariate data were analysed using a balanced fully-nested design with three random factors: System (2 levels), Locality (2 levels, nested in system) and Plot (3 levels, nested in locality and system) considering 3 replicates. Spatial patterns of environmental variables, total number of individuals (N), number of taxa (S) and Shannon index (H0 ) of macrobenthos were examined by a three-way nested analysis of variance (ANOVA) based on the aforementioned design. Prior to the analysis, Cochran's C test was employed to assess homogeneity of variances. When necessary, data were Ln (Xþ1) transformed. The most stringent criterion of p < 0.01 was used to reject null hypotheses when variances were heterogeneous (Underwood, 1997). Mean square (MS) estimates of the ANOVAs were used to assess the variance associated with each examined spatial scale for both environmental variables and macrobenthos (N, S and H0 ). This was done by dividing the difference between the MS of the term of interest and the mean square of the term hierarchically below by the product of the levels of all terms below that of interest

194

P. Veiga et al. / Marine Environmental Research 120 (2016) 191e201

Table 1 Summary of the considered traits for each functional group. Functional groups

Description

A B C D E F G H I J K L M N O P

Subsurface-feeding, predatory, free-living worms Surface-feeding, predatory, free-living worms Surface-feeding, omnivorous, tubicolous or burrow-dwelling worms Deposit-feeding, tubicolous or burrow-dwelling worms with ciliated feeding palps, tentacles, etc., collecting material on or close to the sediment surface Sediment-ingesting, subsurface, free-living or burrow-dwelling worms Omnivorous, free-living crustaceans Selective-feeding, free-living crustaceans Filter-feeding, benthopelagic crustaceans Filter-feeding, tubicolous worms Deposit-feeding, tubicolous crustaceans Predatory and scavenger, benthopelagic crustaceans Surface deposit-feeding bivalves Dependent in whole or part on chemosynthetic symbionts, free-living, sedentary bivalves Filter-feeding, infaunal bivalves Deposit-feeding, infaunal bivalves Microphytobenthos-grazing gastropods

(Underwood, 1997). Negative estimates of variance were removed from the analysis and all the other values recalculated under the assumption that they were sample underestimates of small or zero variances (Fletcher and Underwood, 2002). Estimates of spatial variance were reported as percentages of actual variances to establish the magnitude of each scale's contribution to patterns of macrobenthos and environmental factors. For estimates of variance, the analyses were done on untransformed data to provide variance components comparable across all data. Permutational multivariate analysis of variance PERMANOVA (Anderson, 2001) based on BrayeCurtis untransformed similarity matrixes, built from the abundance data considering species and functional groups independently, was used to analyse the spatial patterns of multivariate assemblage data. The model for these analyses was the same as for ANOVA case described above. All the analyses were done on untransformed data. The statistical significance of multivariate components of variance was tested using a maximum of 999 permutations under a reduced model with significance level set, a priori, at p < 0.05. When the number of unique permutations was less than 30, the statistical significance of multivariate components of variance was tested using Monte Carlo p-values (Terlizzi et al., 2005). The multivariate pseudo-variance components, which can be considered as analogues to the univariate ANOVA estimators, were used to calculate the components of variance associated with each spatial scale in a similar way to the one described above for the univariate analyses. Similar to variance components, negative values were also set to zero (Valdivia et al., 2011). The similarity between macrobenthic assemblages based on species and functional groups was studied by means of the RELATE test which calculates the Spearman Rank correlation for the two matrices of dissimilitude. In order to explore the relationship between univariate indices (i.e. N, S, H0 ) and the environmental variables, rank correlation analyses were done. Due to the non-normal distribution of the data, Spearman's rank correlation was used. The relationship between multivariate assemblage structure of macrobenthos and environmental variables was examined using the BIOENV procedure as an exploratory method (Clarke and Ainsworth, 1993). For BIOENV, environmental variables were previously normalised to give similar weight to variables measured on different units. Moreover, Draftsman plots were done to check the possible correlation of environmental variables. As TC and IC in the sediment showed a strong correlation (r > 0.95), IC was removed from the analyses whereas TC was maintained.

3. Results 3.1. Environmental variables ANOVA showed that environmental variables of the coastal water showed significant variability at the scales of locality and plot (Table 2). Estimates of components of variance, both for O2 and salinity, indicated that most of the variability occurred at the scale of locality (Fig. 2a,b). However, temperature of coastal water followed a different pattern, with most variability found at the scale of system (Fig. 2c). ANOVA for sedimentary fractions indicated significant variability at the scale of system only for coarse gravel (Table 3). Silt/ clay showed only significant variability at the scale of locality whereas coarse, fine and very fine sand showed significant variability at both the scales of locality and plot (Table 3). Finally, significant variability only at the scale of plot was found for fine gravel and for very coarse and medium sand (Table 3). Estimates of components of variance indicated that most of the sedimentary fractions showed the highest variability at the scale of locality (Fig. 3bed and f-h) with the exception of coarse gravel and medium sand (Fig. 3a, e). Moreover, ANOVA indicated that TC, IC, temperature and water content of the sediment showed significant variability both at the scales of locality and plot (Table 4). However, TOC only showed significant variability at the scale of locality (Table 4). Estimates of components of variance indicated that all the variables estimated in the sediment showed the highest variability at the scale of locality (Fig. 4aee).

Table 2 Results of ANOVAs testing differences in coastal water variables across the studied spatial scales. Sy: System; Lo: Locality; Pl: Plot; **: p < 0.01; ***: p < 0.001; ns: not significant; s: significant. Significant differences indicated in bold. Source of variation

df

Sy Lo (Sy) Pl (Sy x Lo) Residual Total Cochran's test Transformation

1 2 8 24 35

Salinity

O2

Temperature

MS

F

MS

F

MS

F

40.56 43.12 0.96 0.066

0.94 45.14*** 14.53***

317.85 828.97 31.24 0.14

0.38 26.54*** 232.15***

50.17 5.77 0.31 0.05

8.69 18.38** 6.90***

0.39ns none

0.27ns none

0.46s none

P. Veiga et al. / Marine Environmental Research 120 (2016) 191e201

100

O2

a

scale of locality (Fig. 5b, c). When multivariate data of macrofaunal assemblages were considered, at both species level and functional groups, PERMANOVA provided evidence that assemblage structure varied significantly at all the studied spatial scales except for system (Table 6). A breakdown of the components of pseudo-variance indicated a similar pattern to that found in S and H0 . Variability at the scale of locality was the major contributor to overall variability, considering data at both species level and functional groups (Fig. 6). Moreover, the RELATE test indicated a high correlation between the matrices obtained from data considering species and functional groups (R ¼ 0.874; p < 0.001).

80

60

40

20

3.3. Relationship between macrofaunal assemblages and environmental variables

0 System

100

Locality

Plot

Replicate

Salinity

b

80

60

40

20

0 System

100

195

Locality

Plot

Replicate

T

c

Results of the Spearman rank correlation analyses between univariate indices (i.e. N, S, H0 ) and environmental variables showed that IC, O2 and salinity of the coastal water were negatively correlated with N, S and H0 (Table 7). Moreover, content in very fine sand, silt/clay and TOC of sediment were positively correlated with N, S and H0 (Table 8). TOC, silt/clay and salinity seemed to play a more important role because they showed the highest correlation coefficients (Table 7). The BIOENV analysis (Global R: 0.567, p < 0.01), considering all the environmental variables, pointed out that the best correlation with multivariate assemblage structure of macrobenthos, based on species, was obtained with the combination of salinity, TC, silt/clay and medium sand (0.567, Table 8). However, the correlation was relatively similar with three variables, without selecting medium sand (0.551), and salinity alone showed a correlation of 0.405 (Table 8). Moreover, the BIOENV analysis (Global R: 0.537, p < 0.01) showed that the best correlation with multivariate assemblage structure of macrobenthos, based on functional groups, was obtained with the combination of TC and fine sand (0.537, Table 8). However, selecting only TC showed a correlation of 0.442 (Table 8). 4. Discussion

80

60

40

20

0 System

Locality

Plot

Replicate

Fig. 2. Univariate estimates of variance associated with each spatial scale in percentage of contribution for O2 (a), Salinity (b) and T: Temperature (c) of the coastal water.

3.2. Macrofaunal assemblages considering species and functional groups ANOVA analyses indicated that N and S showed significant variability at the scales of locality and plot (Table 5). However, H0 only showed significant variability at the scale of locality (Table 5). Estimates of variance indicated that N showed the highest variability at the scale of plot (Fig. 5a). However, a different pattern was found for S and H0 , in which the highest variability was found at the

The first step to identify and understand the drivers that shape the structure of natural assemblages is to define their distribution patterns, on the basis of which explanatory models of structuring processes can be proposed (Underwood et al., 2000). Our study shows an overview of spatial patterns of variability integrating both environmental factors (i.e. in coastal waters and sediment features) and macrobenthos (i.e. N, S, H0 and multivariate structure of assemblages) at four spatial scales from 100s of cm up to >100 km in two transitional water systems. There was a significant variability at the scales of locality (i.e. 10s of km) and plot (i.e. 10s of m) both for most of the environmental variables and for macrobenthos (i.e. N, S and multivariate assemblage structure). In general, analysis of variance components suggested that most variability in patterns of diversity (i.e. S and H0 ) was found at the scale of locality (10s of km), whereas the abundance of macrofauna was more variable at the scale of plot (100s of m). Morrisey et al. (1992) also found that S in subtidal sediments revealed most of the variation at the scale of locality (1 km), whereas N showed most variability at the scale of plot (10 m). Although the spatial scales considered by Morrisey et al. (1992) are an order of magnitude lower than those in our study, their results also indicated that diversity displayed more variability at a wider spatial scale than abundance did. Similarly, two distinct spatial scales, 1 km and <20 cm, explained the variability of S and N, respectively, in intertidal rocky assemblages (Archambault and Bourget, 1996). Our

196

P. Veiga et al. / Marine Environmental Research 120 (2016) 191e201

Table 3 Results of ANOVAs testing differences in sedimentary fractions across the studied spatial scales. Sy: System; Lo: Locality; Pl: Plot; **: p < 0.01; ***: p < 0.001; ns: not significant; s : significant. Significant differences indicated in bold. Source of variation

df

Coarse gravel MS

Sy Lo (Sy) Pl (Sy x Lo) Residual Total Cochran's test Transformation

1 2 8 24 35

88.91 0.18 19.82 9.80

Variance (%)

a

e

509.03 0.01 2.02

**

0.80s none df

Sy Lo (Sy) Pl (Sy x Lo) Residual Total Cochran's test Transformation

Fine gravel F

Coarse sand

F

MS

F

MS

F

408.38 190.72 46.38 2.03

2.14 4.11 22.85***

3584.00 3509.07 450.51 22.90

1.02 7.79 19.68***

160.34 1048.59 47.28 13.93

0.15 22.18*** 3.39**

0.45s none

Medium sand

1 2 8 24 35

Very coarse sand

MS

0.51s none

0.40s none

Very fine sand

Fine sand

Silt and clay

MS

F

MS

F

MS

F

MS

F

4731.57 145.32 368.07 31.46

32.56 0.39 11.70***

577.97 2225.92 151.84 9.40

0.26 14.66** 16.16***

351.10 684.46 8.30 1.04

0.51 82.44*** 8.00***

2.44 3280.20 110.36 59.49

0.00 29.72*** 1.85

0.42s none

0.56s none

Coarse gravel

b

Medium sand

f

Fine gravel

0.36ns none

c

Fine sand

g

0.58s none

Very coarse sand

Coarse sand

d

Very fine sand

Silt and clay

h

Fig. 3. Univariate estimates of variance associated with each spatial scale in percentage of contribution for different granulometric fractions of the sediment: Coarse gravel (a), Fine gravel (b), Very coarse sand (c), Coarse sand (d), Medium sand (e), Fine sand (f), Very fine sand (g) and Silt/clay (h).

Table 4 Results of ANOVAs testing differences in total carbon (TC), inorganic carbon (IC), total organic carbon (TOC), temperature (T) and water content (WC) in the sediment across the studied spatial scales. Sy: System; Lo: Locality; Pl: Plot; *: p < 0.05; ***: p < 0.001; ns: not significant; s: significant. Significant differences indicated in bold. Source of variation

Sy Lo (Sy) Pl (Sy x Lo) Residual Total Cochran's test Transformation

df

1 2 8 24 35

TC

IC

TOC

T

WC

MS

F

MS

F

MS

F

MS

F

MS

F

31.03 17.33 0.80 0.05

1.79 21.71*** 15.26***

22.82 23.21 0.69 0.04

0.98 33.72*** 19.06***

8.21 10.32 0.05 0.15

0.80 195.29*** 0.34

18.35 22.58 0.39 0.04

0.81 57.61*** 10.53***

1.45 0.97 0.03 0.01

1.49 39.11*** 3.09*

0.43s none

0.33ns none

results, however, contrast with many previous studies that reported the highest variability of N at the smallest scale (i.e. between replicates) and that variability usually decreases as the spatial scale increases (Olabarria and Chapman, 2001; Chapman and Tolhurst,

0.38ns none

0.47s none

0.25ns Ln (X þ 1)

2004, 2007; Chapman et al., 2010; Kraufvelin et al., 2011). These differences could be explained because transitional water systems are exposed to strong gradients and high environmental variability (Basset et al., 2013a, b; Reizopoulou et al., 2014) whereas previous

P. Veiga et al. / Marine Environmental Research 120 (2016) 191e201

70

TC

a

60

IC

b

100

70

50 60

40 30

40 30

40

20

20 20

10

Variance (%)

TOC

c

60

80

50

10

0

0 System

100

197

Locality

Plot

T

d

0 System

Replicate

80

80

Locality

Plot

Replicate

System

Locality

Plot

Replicate

WC

e

60

60 40 40 20

20 0

0 System

Locality

Plot

Replicate

System

Locality

Plot

Replicate

Fig. 4. Univariate estimates of variance associated with each spatial scale in percentage of contribution for TC: total carbon (a), IC: inorganic carbon (b), TOC: total organic carbon (c), T: temperature (d) and WC: water content (e) in the sediment.

Table 5 Results of ANOVAs testing for differences in the total number of individuals (N), total number of taxa (S) and Shannon's diversity index (H0 ) of macrobenthos across the studied spatial scales. Sy: System; Lo: Locality; Pl: Plot; *: p < 0.05; **: p < 0.01; ***: p < 0.001; ns: not significant; s: significant. Significant differences indicated in bold. Source of variation

df

Sy Lo (Sy) Pl (Sy x Lo) Residual Total Cochran's test Transformation

1 2 8 24 35

N

H0

S

MS

F

MS

F

MS

F

1.58 20.72 4.40 0.24

0.08 4.71* 18.39***

53.78 197.89 13.00 4.33

0.27 15.22** 3.00*

0.10 3.77 0.20 0.13

0.03 18.62** 1.62

0.22ns Ln (Xþ1)

0.31ns none

0.41s none

studies have explored more homogeneous systems. The environmental gradients have a strong selective influence on macrofauna that may potentially colonise these areas, resulting in complex spatial patterns of biodiversity and species distribution (Basset et al., 2013a). Therefore, the strong variability of environmental factors at the scale of locality in our study seems to explain the higher variability of macrobenthic assemblages at this spatial scale. Results based on the multivariate structure of assemblages, considering both species identity and functional groups, showed significant variability at the spatial scales of locality and plot and identical pattern of variability (i.e. Locality > Replicate > Plot > System). Therefore, functional groups and species composition seem to be modulated by processes acting at the same spatial scales. Similarly, Chapman (1998) reported that spatial patterns could be consistently identified using different levels of taxonomic resolution. She also pointed out the advantages of using coarser taxonomic resolution in benthic sampling programs as this makes the process easier and quicker (Chapman, 1998). Moreover, the use of higher levels of taxonomic resolution or functional groups also allows comparing assemblages with

different species composition (Warwick and Clarke, 1993; Anderson et al., 2005; Mouillot et al., 2006). However, Bell (2007) found that, on coral reef sponge assemblages, functional composition was strongly influenced by depth but not by intrinsic differences between sites; in contrast, patterns of species composition were determined by sedimentation levels, which differed between sites. Moreover, Wong and Dowd (2015) found that taxonomic diversity increased across a gradient of seagrass habitats (i.e. bare sediment, bed edge and bed interior). However, functional diversity did not differ among habitats or showed a weaker pattern across habitats relative to taxonomic diversity (Wong and Dowd, 2015). Therefore, factors responsible for spatial variation on species composition do not always match with those responsible for patterns of functional composition. Consequently, this emphasizes the need of preliminary studies to test the validity of functional groups as surrogate of species. A number of issues have been highlighted when assessing the ecological quality of transitional water systems on taxonomic composition, e.g. taxonomic peculiarities, the reduced species pool and the poor knowledge about species tolerance or resistance to disturbance (Mouillot et al., 2006; Marchini et al., 2008). In this frame, the use of functional groups to assess changes in transitional water systems, produced by natural variability or anthropogenic disturbance, may be useful because functional groups show a more noticeable response to environmental pressure than species richness (Basset et al., 2013a). Nevertheless, if functional groups are to be potentially used as surrogates, they should adequately represent patterns of spatial variability in assemblages at multiple spatial scales (Smale, 2010). In this way, our study confirmed that the considered functional groups of macrobenthos might be used as surrogates in monitoring studies of temperate transitional water systems, because their spatial pattern was remarkably consistent with that achieved using species. It is important to notice that these results could change if a different functional classification or different biological traits

198

P. Veiga et al. / Marine Environmental Research 120 (2016) 191e201

70 60

N

a

50 40 30 20 10 0 System

80

Locality

Plot

Replicate

Plot

Replicate

Plot

Replicate

S

b

Variance (%)

60

40

20

0 System

70 60

Locality

H′

c

50 40 30 20 10 0 System

Locality

Fig. 5. Univariate estimates of variance associated with each spatial scale in percentage of contribution for N: total number of individuals (a), S: total number of taxa (b) and H’: Shannon's diversity index (c).

were selected (Bremner et al., 2006) and when different areas are studied. However, the interest of our results relies on the widely extended use of the biological traits considered here in different functional ecological studies (e.g. Bremner et al., 2006; Hewitt et al., 2008; Paganelli et al., 2012). Moreover, these biological traits have important effects on ecosystem functioning, e.g. nutrient fluxes across the sediment-water interface, bioturbation and irrigation, sediment stability/transport and carbon seques€rnroos and Bonsdorff, 2012). tration (Sigala et al., 2012; To

Therefore, this functional classification could be useful in future ecosystem functioning studies in temperate transitional water systems. If spatial variability of macrobenthic assemblages is related to environmental variables, one can suppose that the latter should also show similar patterns of variation at the same scales (Tolhurst and Chapman, 2007). However, variability patterns of abundance (i.e. Plot > Replicate > Locality > System) did not match with those of any of the studied environmental variables. In this way, abundance patterns could be shaped by biotic interactions, which have €m been considered to be more relevant at smaller scales (Bergstro et al., 2002; Jungerstam et al., 2014) but the significant correlations suggest that environmental factors may play also a role. Our results pointed out that most of the studied environmental variables also showed their highest variability at the scale of locality (e.g. salinity, silt/clay, TC, TOC) these might have contributed to spatial patterns of S, H0 and multivariate assemblage structure of macrobenthos at this scale. Salinity, TOC and silt/clay content of sediment were significantly correlated with N, S and H0 of macrobenthos, although correlation coefficients were lower for N than for S and H0 , suggesting that, as explained above, other factors might be shaping N patterns. The BIOENV analyses showed significant relationships between variability patterns of environmental variables and those of multivariate structure of macrobenthos, considering both species and functional groups. TC, that may be considered a proxy for food resources, seems to be a meaningful variable because it was selected for both species and functional groups. Moreover, salinity and silt/clay content seem to shape the multivariate structure of macrobenthic assemblages based on species directly or indirectly by influencing other variables and processes such as oxygen diffusion, pH, redox potential and nutrient advection (Consentino and Giacobbe, 2008). Previous studies have also shown a close relationship between environmental variables and patterns of benthos (Edgar and Barret, 2002; Ysebaert and Herman, 2002; Anderson et al., 2004). In contrast with our results, Chapman and Tolhurst (2004, 2007) showed that sediment properties (i.e. water content, chlorophyll, carbohydrates, grain size and organic matter) were uncorrelated with patterns of macrobenthos in intertidal mudflats from mangrove forest. Chapman and Tolhurst (2004) found that chlorophyll a of sediment was the variable which correlated best with patterns of benthos but the relationship was much weaker (r ¼ 0.10) than that found with environmental variables considered in our study. Tolhurst and Chapman (2007) suggest that interactions between sediment properties and benthos may be weaker than originally proposed, particularly for homogeneous habitats, as extensive mud or sand flats. Therefore, in transitional water systems, which include very distinct habitats, a stronger correlation between environmental features and patterns of benthos could be expected, as our results have shown. In conclusion, this study showed that spatial patterns of macrobenthic variables presented most variability at the scale of locality (10s of km) but abundance was more variable at the scale of plot (10s of m). Moreover, results pointed that TC, TOC, silt/clay content and salinity were the main drivers in shaping macrobenthic assemblage in the studied systems. In contrast with this result, the abundance was probably controlled by biological drivers that generally operate at smaller scales than 10s of meters. Finally, the extremely low variability between the two studied systems could be explained as both are within the same biogeographical area and no significant changes in oceanographic variables such as sea surface temperature can be found between them. This results contrast with most of the previous

P. Veiga et al. / Marine Environmental Research 120 (2016) 191e201

199

Table 6 Results of PERMANOVAs testing differences in macrobenthic assemblage structure across the studied spatial scales. Sy: System; Lo: Locality; Pl: Plot. Analyses based on BrayeCurtis dissimilarity matrixes from untransformed data considering species and functional groups. All tests used 999 random permutations. When the number of unique permutations were lower than 30, Monte Carlo p-values were considered. Significant differences indicated in bold. Source of variation

df

Sy Lo (Sy) Pl (Sy x Lo) Residual Total

1 2 8 24 35

Species

Functional groups

MS

Pseudo-F

Unique perms

MS

Pseudo-F

Unique perms

11,762 22,774 4865.4 1593.3

0.52 4.68** 3.05**

3 984 996

10,301 20,370 4170.6 1405.4

0.51 4.88** 2.97**

3 998 997

Species Functional groups

50

Pseudovariance (%)

40

30

20

10

0 System

Locality

Plot

Replicate

Fig. 6. Multivariate estimates of pseudo-variance associated with each spatial scale in percentage of contribution for the total assemblage considering species and functional groups.

Table 7 Spearman rank correlations (r) between total number of individuals (N), total number of taxa (S) and Shannon's diversity index (H0 ) and environmental variables. * : p < 0.05; **: p < 0.01; ***: p < 0.001. Environmental variables

N

O2 Salinity Water temperature Coarse gravel Fine gravel Very coarse sand Coarse sand Medium sand Fine sand Very fine sand Silt/clay Total carbon of sediment Inorganic carbon of sediment Total organic carbon of sediment Sediment temperature Sediment water content

H0

S *

0.41 0.50** 0.01 0.01 0.03 0.19 0.24 0.49** 0.26 0.38* 0.53*** 0.09 0.43** 0.69*** 0.02 0.13

***

0.53 0.61*** 0.21 0.01 0.04 0.30 0.36* 0.42* 0.10 0.58*** 0.65*** 0.13 0.53*** 0.76*** 0.16 0.22

0.62*** 0.67*** 0.12 0.16 0.10 0.41* 0.40* 0.22 0.04 0.62*** 0.70*** 0.2 0.53** 0.67*** 0.11 0.07

literature (Olabarria and Chapman, 2001; Chapman and Tolhurst, 2004, 2007; Chapman et al., 2010; Kraufvelin et al., 2011) that identify small spatial scales as the most variable. Therefore, this study stresses the importance of multiple spatial scale studies, particularly in very variable systems, which does not fit with previously described generalities. These deviations from generality show that our understanding of benthic assemblages is site specific and thus, results from studies considering different habitats, spatial scales or different components of the benthos cannot be extrapolated as generalities. Moreover, functional groups may be a good surrogate for species in future monitoring or impact assessment studies as they showed the same spatial pattern as using species. These results improve our capacity to design and understanding results of monitoring programs aimed to detect anthropogenic perturbations and to develop ecosystem management plans. This is critical because the major constraints to implement conservation strategies in marine ecosystems are the general lack of baseline data prior to impacts and substantial gaps in the current knowledge of natural patterns of variability at different scales of assemblages (Claudet and Fraschetti, 2010). This is particularly relevant in intertidal transitional systems, that usually support an important urban and industrial development and are characterised by a high degree of variability and thus, the incorporation of scale issues in their conservations is needed. Finally, the relationships between environmental variables and macrobenthic assemblages were significant and relevant environmental variables showed a similar pattern of variability to that of macrobenthos. Therefore, the use of independent replicates of environmental variables at the same spatial scales as macrobenthos showed to be a useful strategy to relate biological and environmental patterns of variability. Therefore, this replication in environmental variable sampling will improve our ability to explore the relationship between the environment and assemblage structure fundamental to predictive modelling that offers an integrated vision useful for ecosystem management (Gogina et al., 2010).

Table 8 Summary of the results from BIOENV analyses. Correlations obtained between an increasing number of environmental variables and the structure of macrofaunal assemblages, considering species and functional groups. TC: Total carbon. The best combination of variables indicated in bold. Species

Functional groups

No of variables

Correlation

Selections

No of variables

Correlation

Selections

4 3 3 2 2 1 1 1 1

0.567 0.551 0.531 0.511 0.497 0.405 0.339 0.302 0.162

Salinity; TC; Silt/Clay; Medium sand Salinity; Silt/Clay; TC Salinity; Medium sand; TC Silt/Clay; TC Salinity; TC Salinity TC Silt/Clay Medium sand

2 1 1

0.537 0.442 0.187

TC; Fine sand TC Fine sand

200

P. Veiga et al. / Marine Environmental Research 120 (2016) 191e201

Acknowledgments Authors would like to thank Dr. Juan Moreira for reviewing a previous draft of this paper. We are also grateful to three anonymous referees for all the helpful comments and suggestions, which greatly improved this paper. Financial support was partially provided by the European Regional Development Fund (ERDF) through ^ncia Estrate
gico the programme POFC-COMPETE, ‘Quadro de Refere ~o para a Cie ^ncia e a TecNacional (QREN), the Portuguese Fundaça nologia (FCT) through the project PEst-C/MAR/LA0015/2011 and by FEDER GRC2013-004 (Xunta de Galicia). During this study, P. Veiga (SFRH/BPD/81582/2011) and M. Rubal (SFRH/BDP/104225/2014) ~o para a were supported by postdoctoral grants awarded by Fundaça ^ncia e a Tecnologia (FCT, Portugal). F. Aneiros was supported by Cie FPU program (Spanish Education Ministry, AP2010-2010). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.marenvres.2016.08.011. References Anderson, M.J., 2001. A new method for non-parametric multivariate analysis of variance. Austral Ecol. 26, 32e46. http://dx.doi.org/10.1111/j.14429993.2001.01070.pp.x. Anderson, M.J., Ford, R.B., Feary, D.A., Honeywill, C., 2004. Quantitative measures of sedimentation in an estuarine system and its relationship with intertidal softsediment infauna. Mar. Ecol. Prog. Ser. 272, 33e48. http://dx.doi.org/10.3354/ meps272033. Anderson, M.J., Connell, S.D., Gillanders, B.M., Diebel, C.E., Blom, W.M., Saunders, J.E., Landers, T.J., 2005. Relationships between taxonomic resolution and spatial scales of multivariate variation. J. Anim. Ecol. 74, 636e646. http:// dx.doi.org/10.1111/j.1365-2656.2005.00959.x. Archambault, P., Bourget, E., 1996. Scales of coastal heterogeneity and benthic intertidal species richness, diversity and abundance. Mar. Ecol. Prog. Ser. 136, 111e121. http://dx.doi.org/10.3354/meps136111. Barroso, C.M., Moreira, M.H., Gibbs, P.E., 2000. Comparison of imposex and intersex development in four prosobranch species for TBT monitoring of a southern European estuarine system (Ria de Aveiro, NW Portugal). Mar. Ecol. Prog. Ser. 201, 221e232. http://dx.doi.org/10.3354/meps201221. Basset, A., Barbone, E., Elliott, M., Li, B.-L., Jorgensen, S.E., Lucena-Moya, P., Pardo, I., Mouillot, D., 2013a. A unifying approach to understanding transitional waters: fundamental properties emerging from ecotone ecosystems. Estuar. Coast. Shelf Sci. 132, 5e16. http://dx.doi.org/10.1016/j.ecss.2012.04.012. Basset, A., Elliott, M., West, R.J., Wilson, J.G., 2013b. Estuarine and lagoon biodiversity and their natural goods and services. Estuar. Coast. Shelf Sci. 132, 1e4. http://dx.doi.org/10.1016/j.ecss.2013.05.018. Bell, J.J., 2007. Contrasting patterns of species and functional composition of coral reef sponge assemblages. Mar. Ecol. Prog. Ser. 339, 73e81. http://dx.doi.org/ 10.3354/meps339073. €m, U., Englund, G., Bonsdroff, E., 2002. Small-scale spatial structure of Baltic Bergstro Sea zoobenthosdinferring processes from patterns. J. Exp. Mar. Biol. Ecol. 281, 123e136. http://dx.doi.org/10.1016/S0022-0981(02)00440-9. Bremner, J., Rogers, S.I., Frid, C.L.J., 2006. Methods for describing ecological functioning of marine benthic assemblages using biological traits analysis (BTA). Ecol. Indic. 6, 609e622. http://dx.doi.org/10.1016/j.ecolind.2005.08.026. Chapman, M.G., 1998. Relationships between spatial patterns of benthic assemblages in a mangrove forest using different levels of taxonomic resolution. Mar. Ecol. Prog. Ser. 162, 71e78. http://dx.doi.org/10.3354/meps162071. Chapman, M.G., Tolhurst, T.J., 2004. The relationship between invertebrate assemblages and bio-dependant properties of sediment in urbanized temperate mangrove forests. J. Exp. Mar. Biol. Ecol. 304, 51e73. http://dx.doi.org/10.1016/ j.jembe.2003.11.019. Chapman, M.G., Tolhurst, T.J., 2007. Relationships between benthic macrofauna and biogeochemical properties of sediments at different spatial scales and among different habitats in mangrove forests. J. Exp. Mar. Biol. Ecol. 343, 96e109. http://dx.doi.org/10.1016/j.jembe.2003.11.019. Chapman, M.G., Tolhurst, T.J., Murphy, R.J., Underwood, A.J., 2010. Complex and inconsistent patterns of variation in benthos, micro-algae and sediment over multiple spatial scales. Mar. Ecol. Prog. Ser. 398, 33e47. http://dx.doi.org/ 10.3354/meps08328. Cisneros, K.O., Smit, A.J., Laudien, J., Schoeman, D.S., 2011. Complex, dynamic combination of physical, chemical and nutritional variables controls spatiotemporal variation of sandy beach community structure. PLoS One 6 (8), e23724. http://dx.doi.org/10.1371/journal.pone.0023724. Clarke, K.R., Ainsworth, M., 1993. A method of linking multivariate community structure to environmental variables. Mar. Ecol. Prog. Ser. 92, 205e219.

Claudet, J., Fraschetti, S., 2010. Human-driven impacts on marine habitats: a regional meta-analysis in the Mediterranean Sea. Biol. Cons. 143, 2195e2206. http://dx.doi.org/10.1016/j.biocon.2010.06.004. Consentino, A., Giacobbe, S., 2008. Distribution and functional response of sublittoral soft bottom assemblages to sedimentary constraints. Estuar. Coast. Shelf Sci. 79, 263e276. http://dx.doi.org/10.1016/j.ecss.2008.04.003. Dauvin, J.C., 2007. Paradox of estuarine quality: benthic indicators and indices, consensus or debate for the future. Mar. Pollut. Bull. 55, 271e281. http:// dx.doi.org/10.1016/j.marpolbul.2006.08.017. De la Huz, R., Lastra, M., 2008. Effects of morphodynamic state on macrofauna community of exposed sandy beaches on Galician coast (NWSpain). Mar. Ecol. 29 (Suppl. 1), 150e159. http://dx.doi.org/10.1111/j.1439-0485.2008.00230.x. Dias, J.M., Lopes, J.F., Dekeyser, I., 1999. Hydrological characterization of Ria de Aveiro, Portugal, in early summer. Oceanol. Acta 22, 473e485. http://dx.doi.org/ 10.1016/S0399-1784(00)87681-1. Díaz, E.R., Erlandsson, J., Westerbom, M., Kraufvelin, P., 2015. Depth-related spatial patterns of sublittoral blue mussel beds and their associated macrofauna diversity revealed by geostatistical analyses. Mar. Ecol. Prog. Ser. 540, 121e134. http://dx.doi.org/10.3354/meps11461. Edgar, G.J., Barret, N.S., 2002. Benthic macrofauna in Tasmanian estuaries: scales of distribution and relationships with environmental variables. J. Exp. Mar. Biol. Ecol. 270, 1e24. http://dx.doi.org/10.1016/S0022-0981(02)00014-X. Fletcher, D.J., Underwood, A.J., 2002. How to cope with negative estimates of components of variance in ecological field studies. J. Exp. Mar. Biol. Ecol. 273, 89e95. http://dx.doi.org/10.1016/S0022-0981(02)00142-9. Fraschetti, S., Terlizzi, A., Benedetti-Cecchi, L., 2005. Patterns of distribution of marine assemblages from rocky shores: evidence of relevant scales of variation. Mar. Ecol. Prog. Ser. 296, 13e29. http://dx.doi.org/10.3354/meps296013. Gogina, M., Glockzin, M., Zettler, M.L., 2010. Distribution of benthic macrofaunal communities in the western Baltic Sea with regard to near-bottom environmental parameters. 1. Causal analysis. J. Mar. Syst. 79, 112e123. http:// dx.doi.org/10.1016/j.jmarsys.2009.07.006. Hewitt, J.E., Thrush, S.F., Dayton, P.K., Bonsdorff, E., 2007. The effect of spatial and temporal heterogeneity on the design and analysis of empirical studies of scaledependent systems. Am. Nat. 169, 398e408. Hewitt, J.E., Thrush, S.F., Dayton, P.D., 2008. Habitat variation, species diversity and ecological functioning in a marine system. J. Exp. Mar. Biol. Ecol. 366, 116e122. http://dx.doi.org/10.1016/j.jembe.2008.07.016. Hooper, D.U., Chapin, F.S., Ewel, J.J., Hector, A., Inchausti, P., Lavorel, S., Lawton, J.H., 2005. Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecol. Monogr. 75, 3e35. http://dx.doi.org/10.1890/04-0922. Jaramillo, E., McLachlan, A., Dugan, J., 1995. Total simple area and estimates of species richness in exposed sandy beaches. Mar. Ecol. Prog. Ser. 119, 311e314. Jungerstam, J., Erlandsson, J., McQuaid, C.D., Porri, F., Westerbom, M., Kraufvelin, P., 2014. Is habitat amount important for biodiversity in rocky shore systems? A study of South African mussel assemblages. Mar. Biol. 161, 1507e1519. http:// dx.doi.org/10.1007/s00227-014-2436-4. Kraan, C., Dormann, C.F., Greenfield, B.L., Thrush, S.F., 2015. Cross-scale variation in biodiversity-environment links illustrated by coastal sandflat communities. PLoS One 10 (11), e0142411. http://dx.doi.org/10.1371/journal.pone.0142411. Kraufvelin, P., Perus, J., Bonsdorff, E., 2011. Scale-dependent distribution of softbottom infauna and possible structuring forces in low diversity systems. Mar. Ecol. Prog. Ser. 426, 13e28. http://dx.doi.org/10.3354/meps09038. Marchini, A., Munari, C., Mistri, M., 2008. Functions and ecological status of eight Italian lagoons examined using biological traits analysis (BTA). Mar. Pollut. Bull. 56, 1076e1085. http://dx.doi.org/10.1016/j.marpolbul.2008.03.027. McGill, B.J., 2010. Matters of scale. Science 328, 575e576. http://dx.doi.org/10.1126/ science.1188528. PMID:20431001. Morrisey, D.J., Howitt, L., Underwood, A.J., Stark, J.S., 1992. Spatial variation in soft sediment benthos. Mar. Ecol. Prog. Ser. 81, 197e204. Mouillot, D., Spatharis, S., Reizopoulou, S., Laugier, T., Sabetta, L., Basset, A., Chi, T.D., 2006. Alternatives to taxonomic-based approaches to assess changes in transitional water communities Aquat. Conserv 16, 469e482. http://dx.doi.org/ 10.1002/aqc.769. Olabarria, C., Chapman, M.G., 2001. Comparison of patterns of spatial variation of microgastropods between two contrasting intertidal habitats. Mar. Ecol. Prog. Ser. 220, 201e211. http://dx.doi.org/10.3354/meps220201. Paganelli, D., Marchini, A., Occhipinti-Ambrogi, A., 2012. Functional structure of marine benthic assemblages using biological traits analysis (BTA): a study along the Emilia-Romagna coastline (Italy, North-West Adriatic sea). Estuar. Coast. Shelf Sci. 96, 245e256. http://dx.doi.org/10.1016/j.ecss.2011.11.014. Passarelli, C., Olivier, F., Paterson, D.M., Hubas, C., 2012. Impacts of biogenic structures on benthic assemblages: microbes, meiofauna, macrofauna and related ecosystem functions. Mar. Ecol. Prog. Ser. 465, 85e97. http://dx.doi.org/10.3354/ meps09915. Pratt, D.R., Lohrer, A.M., Pilditch, C.A., Thrush, S.F., 2014. Changes in ecosystem function across sedimentary gradients in estuaries. Ecosystems 17, 182e194. http://dx.doi.org/10.1007/s10021-013-9716-6. Prego, R., Fraga, F., 1992. A simple model to calculate the residual flows in a Spanish ria. Hydrographic consequences in the ria of Vigo. Estuar. Coast. Shelf Sci. 34, 603e615. http://dx.doi.org/10.1016/S0272-7714(05)80065-4. Prego, R., Filgueiras, A.V., Santos-Echeandía, J., 2008. Temporal and spatial changes of total and labile metal concentration in the surface sediments of the Vigo Ría (NW Iberian Peninsula): influence of anthropogenic sources. Mar. Pollut. Bull. 56, 1031e1042. http://dx.doi.org/10.1016/j.marpolbul.2008.01.036.

P. Veiga et al. / Marine Environmental Research 120 (2016) 191e201 Pritchard, D.W., 1967. What is an estuary: physical view point. In: Lauff, G.E. (Ed.), Estuaries. American Association for the Advancement of Science, Washington DC, pp. 3e5. Raffaelli, D., Friedlander, A.M., 2012. Biodiversity and ecosystem functioning: an ecosystem-level approach. In: Solan, M., Aspden, R.J., Paterson, D.M. (Eds.), Marine Biodiversity and Ecosystem Functioning: Frameworks, Methodologies, and Integration. Oxford University Press, Oxford, pp. 149e163. Reizopoulou, S., Simboura, N., Barbone, E., Aleffi, F., Basset, A., Nicolaidou, A., 2014. Biodiversity in transitional waters: steeper ecotone, lower diversity. Mar. Ecol. 35, 78e84. http://dx.doi.org/10.1111/maec.12121. Rubal, M., Veiga, P., Vieira, R., Sousa-Pinto, I., 2011. Seasonal patterns of tidepool macroalgal assemblages in the North of Portugal. Consistence between species and functional group approaches. J. Sea Res. 66, 187e194. http://dx.doi.org/ 10.1016/j.seares.2011.07.003. Rubal, M., Veiga, P., Reis, P.A., Bertocci, I., Sousa-Pinto, I., 2014. Effects of subtle pollution at different levels of biological organization on species-rich assemblages. Environ. Pollut. 191, 101e110. http://dx.doi.org/10.1016/ j.envpol.2014.04.019. Schlacher, T.A., Thompson, L., 2013. Spatial structure on ocean-exposed sandy beaches: faunal zonation metrics and their variability. Mar. Ecol. Prog. Ser. 478, 43e55. http://dx.doi.org/10.3354/meps10205. Sigala, K., Reizopoulou, S., Basset, A., Nicolaidou, A., 2012. Functional diversity in three Mediterranean transitional water ecosystems. Estuar. Coast. Shelf Sci. 110, 202e209. http://dx.doi.org/10.1016/j.ecss.2012.06.002. Smale, D.A., 2010. Monitoring marine macroalgae: the influence of spatial scale on the usefulness of biodiversity surrogates. Divers. Distrib. 16, 985e995. http:// dx.doi.org/10.1111/j.1472-4642.2010.00709.x. Sun, B., Fleeger, J.W., Carney, R.S., 1993. Sediment microtopography and the smallscale spatial distribution of meiofauna. J. Exp. Mar. Biol. Ecol. 167, 73e90. http:// dx.doi.org/10.1016/0022-0981(93)90185-Q. Terlizzi, A., Benedetti-Cecchi, L., Bevilacqua, S., Fraschetti, S., Guidetti, P., Anderson, M.J., 2005. Multivariate and univariate asymmetrical analyses in environmental impact assessment: a case study of Mediterranean subtidal sessile assemblages. Mar. Ecol. Prog. Ser. 289, 27e42. http://dx.doi.org/10.3354/ meps289027. Tolhurst, T.J., Chapman, M.G., 2007. Patterns in biogeochemical properties of sediments and benthic animals among different habitats in mangrove forests. Austral Ecol. 32, 775e788. http://dx.doi.org/10.1111/j.1442-9993.2007.01764.x. €rnroos, A., Bonsdorff, E., 2012. Developing the multitrait concept for functional To diversity: lessons from a system rich in functions but poor in species. Ecol. Appl. 22, 2221e2236. http://dx.doi.org/10.1890/11-2042.1.
pez, S., Varela, R.A., Delhez, E., 2001. Residual circulation and thermohaTorres-Lo line distribution of the Ría de Vigo: a 3-D hydrodynamical model. Sci. Mar. 65, 277e289. Underwood, A.J., 1997. Experiments in Ecology: Their Logical Design and Interpretation Using Analysis of Variances. Cambridge University Press, Cambridge,

201

p. 504. Underwood, A.J., Chapman, M.G., 1996. Scales of spatial patterns of distribution of intertidal invertebrates. Oecologia 107, 212e224. http://dx.doi.org/10.1007/ BF00327905. Underwood, A.J., Chapman, M.G., 1998a. A method for analysing spatial scales of variation in composition of assemblages. Oecologia 117, 570e578. http://www. jstor.org/stable/4222201. Underwood, A.J., Chapman, M.G., 1998b. Spatial analyses of intertidal assemblages on sheltered rocky shores. Aust. J. Ecol. 23, 138e157. http://dx.doi.org/10.1111/ j.1442-9993.1998.tb00712.x. Underwood, A.J., Chapman, M.G., Connell, S.D., 2000. Observations in ecology: you can't make progress on processes without understanding the patterns. J. Exp. Mar. Biol. Ecol. 250, 97e115. http://dx.doi.org/10.1016/S0022-0981(00)00181-7. Valdivia, N., Scrosati, R.A., Molis, M., Knox, A.S., 2011. Variation in community structure across vertical intertidal stress gradients: how does it compare with horizontal variation at different scales? PLoS One 6 (8), e24062. http:// dx.doi.org/10.1371/journal.pone.0024062. Veiga, P., Besteiro, C., Rubal, M., 2011. Meiobenthic communities structure at different spatial scales on sandy beaches from a ria environment: the role of abiotic factors. Mar. Biol. Res. 7, 425e435. http://dx.doi.org/10.1080/ 17451000.2010.527987. Veiga, P., Rubal, M., Vieira, R., Arenas, F., Sousa-Pinto, I., 2013. Spatial variability in intertidal macroalgal assemblages on the North Portuguese coast. Consistence between species and functional group approaches. Helgol. Mar. Res. 67, 191e201. http://dx.doi.org/10.1007/s10152-012-0315-2. Veiga, P., Rubal, M., Cacabelos, E., Maldonado, C., Sousa-Pinto, I., 2014. Spatial variability of macrobenthic zonation on exposed sandy beaches. J. Sea Res. 90, 1e9. http://dx.doi.org/10.1016/j.seares.2014.02.009. Warwick, R.M., Clarke, K.R., 1993. Comparing the severity of disturbance: a metaanalysis of marine macrobenthic community data. Mar. Ecol. Prog. Ser. 92, 221e231. Wong, M.C., Dowd, M., 2015. Patterns in taxonomic and functional diversity of macrobenthic invertebrates across seagrass habitats: a case study in Atlantic Canada. Estuar. Coast 38, 2323e2336. http://dx.doi.org/10.1007/s12237-0159967-x. Wright, J.P., Naeem, S., Hector, A., Lehman, C., Reich, P.B., Schmid, B., Tilman, D., 2006. Conventional functional classification schemes underestimate the relationship with ecosystem functioning. Ecol. Lett. 9, 111e120. http://dx.doi.org/ 10.1111/j.1461-0248.2005.00850.x. Yaffee, S.L., 1997. Why environmental policy nightmares recur. Conserv. Biol. 11, 328e337. http://dx.doi.org/10.1046/j.1523-1739.1997.95204.x. Ysebaert, T., Herman, P.M.J., 2002. Spatial and temporal variation in benthic macrofauna and relationships with environmental variables in an estuarine, intertidal soft-sediment environment. Mar. Ecol. Prog. Ser. 244, 105e124. http:// dx.doi.org/10.3354/meps244105.