Introduction

Peracarid crustaceans are among the most diverse and numerically dominant organisms of soft-bottom benthic faunas (Fincham 1974; Dauvin et al. 1994; Prato and Biandiolino 2005). Peracarids also play an important role in structuring of benthic assemblages (Duffy and Hay 2000) and are important source of food for other benthic animals and fishes of commercial importance (McDermott 1987; Dauvin 1988a; Beare and Moore 1996). Distribution and abundance of peracarids inhabiting marine sediments are influenced by a number of abiotic factors, such as sediment composition (Parker 1984; De Grave 1999) and organic content (Robertson et al. 1989). Many peracarid species are also sensitive to hydrocarbon pollution and other perturbations and, therefore, their abundance and species diversity may serve as indicators of environmental conditions (e.g., Marques and Bellan-Santini 1990; Corbera and Cardell 1995; Gómez-Gesteira and Dauvin 2000; Guerra-García and García-Gómez 2004).

The Galician rias (NW Spain) are a special and complex kind of estuarine system and have a great economic and social importance due to the presence of fisheries, bivalve culture and shellfish resources (Nombela et al. 1995; Figueiras et al. 2002). The rias also have a large diversity of sedimentary habitats inhabited by a particularly rich benthic fauna (Cadée 1968; López-Jamar and Mejuto 1985). Peracarid fauna of the Galician rias, however, are little known and few studies have been devoted to describe diversity, distribution and composition of peracarid assemblages on their soft-bottoms (Anadón 1975; Sánchez-Mata et al. 1993).

Composition and distribution of soft-bottom benthos are well-known in many areas of the Galician coast (Viéitez and Baz 1988; Junoy and Viéitez 1989; Mazé et al. 1990; Palacio et al. 1991; Currás and Mora 1991). There is, however, a lack of studies in some small rias such as the Ría de Aldán. This ría is located on the mouth of the Ría de Pontevedra and shows a variety of subtidal sediments, ranging from gravel to mud, at depths of between 3 and 45 m. Thus, the main objectives of this paper were to characterize the composition and distribution of the peracarid fauna on the subtidal soft-bottoms of the Ría de Aldán as well as studying the influence of several environmental variables on the distribution patterns of peracarids. This will also provide baseline data for further comparative analyses.

Material and methods

Study area

The Ría de Aldán is situated on the southern margin of the mouth of Ría de Pontevedra, between 42°16′40′′–42°20′50′′N and 8°49′–8°52′W. This ria has 7 km length, 3.5 km width and has a maximum depth of 45 m; its mouth is oriented northwards. Mean salinity values are around 36 ‰ in outer areas of the ria and there is gradual increase in salinity from the inner to the outer part of the ria (Parada 2004). The small Aldán River flows into the inner area. Both East and West margins are made up of rocky substratum which alternates with sandy beaches. The ria is greatly influenced by strong oceanic hydrodynamism which reaches the inner areas and reduces the effect of freshwater input from the Aldán River. The ria is influenced by the growing practice of bivalve culture on rafts, mostly in the inner areas of the ria; these practices are supposed to contribute to the increase of the content of silt/clay and organic matter in those areas, such as it occurs in other Galician rías.

Sample collection and processing

The sampling programme comprised 27 stations which covered the full extent of the ria in order to provide sufficient information on the distribution of peracarids (Fig. 1; Table 1). Quantitative sampling was done during July–August 1997 using a Van Veen grab with a sampling area of 0.056 m2. Five replicates were taken at each station, which accounted for a total area of 0.28 m2. Samples were sieved through a 0.5 mm mesh, and fixed in 10% buffered formaldehyde solution for later sorting and identification of the fauna. An additional sediment sample was taken at each station to analyse granulometric composition, carbonates and organic matter content. The following granulometric fractions were considered: gravel (GR >2 mm), very coarse sand (VCS 1–2 mm), coarse sand (CS 0.5–1 mm), medium sand (MS 0.25–0.5 mm), fine sand (FS 0.125–0.25 mm), very fine sand (VFS 0.063–0.125 mm), and silt/clay (<0.063 mm). Median grain size (Q50) and sort coefficient (S0) (Trask 1932) were also determined for each sample. Temperature was also measured in sediment, surface water and bottom water. Sediment types were characterized according to Junoy (1996). Calcium carbonate content (%) was estimated by the treatment of the sample with hydrochloric acid. The total organic matter content (TOM, %) was estimated from the weight loss after placing samples in a furnace for 4 h at 450°C.

Fig. 1
figure 1

a Location of the Ría de Aldán, b location of sampling stations and spatial distribution of peracarid assemblages in the ria as determined by multivariate analyses and c bathymetry (m) in numbers and sediment type in the ria

Full size image
Table 1 Position, depth, sediment and water temperature and sedimentary characteristics of sampling stations in the Ría de Aldán
Full size table

Data analysis

Several univariate measures were calculated for each sampling station: total abundance (N), number of species (S), the Shannon–Wiener diversity index (H′, as log2), and Pielou’s evenness (J). For any given site, species with ≥4% of total abundance were considered as dominant (Field et al. 1982). Peracarid assemblages were determined through non-parametric multivariate techniques as described by Field et al. (1982) using the PRIMER v5.0 (Plymouth Routines in Multivariate Ecological Research) software package (Clarke and Warwick 1994). A similarity matrix was constructed by means of the Bray–Curtis similarity coefficient by first applying fourth root transformation on species abundance to minimise the contribution of the most abundant species. Differences in faunistic composition between sampling stations were tested using the one-way ANOSIM test. From the similarity matrix, classification of stations was done by cluster analysis based on the group-average sorting algorithm and an ordination by means of non-metric multidimensional scaling (nMDS). The SIMPER program was next used to identify species that greatly contributed to the differentiation of station groups. The species present in each group of stations were further classified according to the constancy and fidelity indexes (Glémarec 1964; Cabioch 1968).

The BIO-ENV procedure (belonging to the PRIMER package), and the canonical correspondence analysis (CCA, using the CANOCO v4.02, Canonical Community Ordination package; Ter Braak 1988) were used to research the possible relationship between peracarid distribution in the ria, and the measured environmental variables. The forward selection was used in the latter to detect which variables explained the most variance in the species data. All variables expressed in percentages were previously transformed by log (x + 1).

Results

Sediments

Sediments were mainly sandy in most of the ria (Fig. 1; Table 1). Coarser sandy granulometric fractions were more prevalent at the mouth of the ria, and muddy bottoms were restricted to inner and sheltered areas. There was a decrease in grain size and an increase in organic content from the outer areas to the inner areas of the ria.

Peracarid fauna

A total of 16,191 peracarid individuals were collected, comprising 125 species of peracarids belonging to five orders. Amphipods were the best represented in total number of species (79) and individuals (73.4% of numerical abundance) followed by isopods (20 species and 2.6% abundance) and cumaceans (14 species and 2.7% abundance). Tanaids and mysids were less diverse in number of species (5 and 7, respectively). Tanaids comprised 20.6% of the total abundance, mostly due to the abundance of Apseudes latreillii (Milne-Edwards, 1828) in some sites. Values of univariate measures are shown in Table 2. The lowest abundance values were recorded at medium sand (St. 13, 171 individuals/m2), while the highest numbers were recorded at muddy sand (St. 34, 12,229 individuals/m2). Number of species varied between 15 (St. 13) and 38 (St. 29, fine sand); diversity ranged from 0.79 (St. 21, medium sand) and 4.26 (St. 32, fine sand). Evenness showed low values on sediments with a high dominance of Siphonoecetes kroyeranus Bate, 1856 (St. 21; J, 0.18), Apseudes latreilli (St. 22, 24, 34; J, 0.30–0.52) and Gammarella fucicola (Leach, 1814) (St. 33; J, 0.44).

Table 2 Number of species (S), total abundance per m2 and 0.28 m2 (N), Shannon Wiener’s diversity index (H′, log2) and Pielou’s evenness (J) for each sampling station in the Ría de Aldán
Full size table

The dominant species in terms of abundance were the tanaid Apseudes latreillii and the amphipods Siphonoecetes kroyeranus, Photis longipes (della Valle, 1893), Gammarella fucicola, Aoridae spp. (undetermined females), Microdeutopus versiculatus (Bate, 1856), Guernea coalita (Norman, 1868), Perioculodes longimanus (Bate and Westwood, 1868), Leucothoe incisa Robertson, 1892, Ampithoe ramondi Audouin, 1826 and Ampelisca typica (Bate, 1856), and the cumacean Cumella sp. The remaining peracarid taxa comprised less than 25% of the total abundance.

The most widespread species in the ria (at least found in 15 sampling sites) were the amphipods Leucothoe incisa, Aoridae spp., Perioculodes longimanus, Photis longipes and Atylus vedlomensis (Bate and Westwood, 1862), the isopods Eurydice truncata (Norman, 1868) and Campecopea hirsuta (Montagu, 1804), and the cumacean Cumella sp. (Table 3). About 55% of the species were only found in one to four sites.

Table 3 List of the most abundant peracarid taxa in the Ría de Aldán (>30 individuals collected in total)
Full size table

Multivariate analyses

The ANOSIM test revealed significant differences in faunistic composition between all stations (global R, 0.865; P, 0.001), except between St. 2 and 17 (R, 0.148; P, 0.14), St. 17 and 19 (R, 0.108; P, 0.21), and St. 24 and 26 (R, 0.172; P, 0.05). The dendrogram obtained by cluster analysis showed the presence of two major groups of sites at a similarity level of 20% (Fig. 2): group A (finer sediments) and group B (coarser sediments). Group A was further subdivided into group A1 (St. 23, 27, 28, 31; mud) and group A2 (fine sand), the latter was subdivided into group A2a (St. 8, 9, 20, 21, 22, 24, 25, 26) and A2b (St. 29, 30, 32, 33, 34). Group B was subdivided in subgroup B1 (St. 13, 14; medium sand) and B2 (St. 2, 3, 6, 7, 12, 17, 18, 19; mostly coarse sand). nMDS ordination showed similar results to those of the dendrogram (Fig. 3). Relative abundance of peracarid orders in each assemblage is shown in Fig. 4.

Fig. 2
figure 2

Peracarid assemblages in the Ría de Aldán as determined by cluster analysis based on Bray–Curtis similarity coefficient

Full size image
Fig. 3
figure 3

nMDS ordination of sampling stations showing groups determined by cluster analysis and ordination of sites with values of some environmental variables superimposed (very fine sand, depth and fine sand)

Full size image
Fig. 4
figure 4

Relative abundance (%) of each peracarid order in the groups of stations determined by multivariate analyses

Full size image

Group A was located in the sheltered area of the ria. Subgroup A1 was constituted by muddy sites and had lower number of individuals and species than the other subgroups of A. The species that mostly contributed to similarities in A1 were Harpinia pectinata Sars, 1891, Leucothoe incisa and Tanaissus lilljeborgi (Stebbing, 1891), whereas Metaphoxus simplex Bate, 1857, Harpinia pectinata and Microdeutopus armatus Chevreux, 1887) were the most abundant species. Subgroup A2 was composed of sandy sediments (from muddy sand to coarse sand). The species which mostly contributed to characterize subgroup A2a (medium-fine sand) were Perioculodes longimanus, Leucothoe incisa (present in all stations), Siphonoecetes kroyeranus and Apseudes latreillii, the latter being the most abundant species. Subgroup A2b (fine-muddy sand) was located in the inner part of the ria and showed the highest number of individuals of all groups. This subgroup was characterized by Photis longipes, Microdeutopus versiculatus and Gammarella fucicola, which were present in all stations; those species and Apseudes latreillii were the most abundant.

Group B was located in the outer part of the ria. Subgroup B1 had less total abundance and number of species of peracarids than the other groups of the ría. Subgroup B2 showed the highest number of species of all groups; this subgroup was determined by Guernea coalita, Haplostylus sp. and Eurydice truncata, being Apseudes latreilli the most abundant species.

SIMPER analysis showed that Guernea coalita, Haplostylus sp. and Siphonoecetes kroyeranus explained most of dissimilarity between groups A2a and B2. Guernea coalita, Siphonoecetes kroyeranus and Iphinoe trispinosa (Goodsir, 1843) contributed greatly to the differentiation of B2 from B1. Apseudes latreillii, Siphonoecetes kroyeranus and Lepidepecreum longicornis (Bate and Westwood, 1862) differentiated group A2a from B1. Group A2a differed from A1 due to Harpinia pectinata, Siphonoecetes kroyeranus and Metaphoxus simplex. Gammarella fucicola and Microdeutopus versiculatus differentiated group A2a from A2b, whereas Harpinia pectinata and Microdeutopus versiculatus differentiated group A1 from A2b.

Species affinities

Cluster analysis done on the abundance data of the dominant species showed the existence of three major groups at similarity level of about 15% (Fig. 5). Group 1 included species mostly found in coarse sand (cluster group B2). Group 2 comprised species found in muddy and fine sand sediments. Subgroup 2a comprised species with higher abundance in muddy sediments (cluster group A1), while subgroup 2b was composed of species found in fine sand sediments (cluster group A2b). Group 3 included species mostly found in sediments composed of fine and medium sand (cluster group A2a).

Fig. 5
figure 5

Dendrogram based on cluster analysis showing the classification of species with a numerical dominance ≥4% at any given site. Species code are given in Table 3, except: Amp sp., Ampelisca sp.; Amp spi, Ampelisca spinipes; Aph bis, Apherusa bispinosa; Bat ele, Bathyporeia elegans; Bod sco, Bodotria scorpioides; Che ass, Cheirocratus assimilis; Con cyl, Conilera cylindracea; Gas san, Gastrosaccus sanctus; Idu lon, Idunella longirostris; Iph tri, Iphinoe trispinosa; Lep hir, Leptocheirus hirsutimanus; Lep lon, Lepidepecreum longicornis; Mon car, Monoculodes carinatus; Nat neg, Natatolana neglecta; Pse lon, Pseudocuma longicorne; Orc nan, Orchomenella nana; Zeu nor, Zeuxo normani

Full size image

Relation with environmental variables

The BIO-ENV procedure showed that the combination of gravel, very fine sand, silt/clay and depth had the highest correlation with faunistic data (ρ w 0.516). Very fine sand was the variable that alone showed the highest correlation (ρ w 0.277), followed by depth (ρ w 0.270) and fine sand (ρ w 0.238).

The nMDS ordination of sites with superimposed values of the mentioned variables showed that stations appeared distributed from right to left following an increase in very fine sand and fine sand fractions in the sediment, accompanied by decreasing values of depth (Fig. 3).

The forward selection of CCA selected median grain size, silt/clay, sorting coefficient and depth as the variables explaining most of the variance in the species data (P: 0.002). Axes I and II were the most important in the CCA ordination, accumulating 21.2% of species variance and 27.9% of species-environment variance. Cluster groups with higher content of coarser granulometric fractions were distributed on the left of the ordination, while assemblages located in fine sand-mud were distributed on the right, following a gradient defined by a decrease in median grain size (Fig. 6).

Fig. 6
figure 6

Canonical correspondence analysis (CCA) ordination of stations and environmental variables relative to axes I and II for the Ría de Aldán. Gravel, G; very coarse sand, VCS; coarse sand, CS; medium sand, MS; fine sand, FS; very fine sand, VFS; median grain size, Q50; sorting coefficient, S0; bottom water temperature, bottom temp; surface water temperature, surf temp; sediment temperature, sed temp; calcium carbonate content, carb; total organic matter content, TOM

Full size image

Discussion

Peracarid diversity

The peracarid fauna diversity of Ría de Aldán was higher than in other European temperate waters, having been recorded the typical species from shallow sediments of those waters (Dauvin et al. 1994; Conradi et al. 1997; Cunha et al. 1999). The total number of peracarid species founded at Ría de Aldán was 125, while Dauvin et al. (1994) reported 99 peracarid species from the western English channel (circalittoral suprabenthic coarse sand community), Cunha et al. (1999) reported 61 peracarid species from Ría de Aveiro (Portugal), and Conradi and López González (2001) reported 67 peracarid species from Algeciras Bay. Table 4 shows a comparison of the peracarid fauna between the Ría de Aldán and other nearby geographical areas.

Table 4 Comparison of peracarid diversity between the Ría de Aldán (this work) and other nearby geographical areas (Galicia, Basque country and Portuguese coast)
Full size table

The large peracarid diversity at Ría de Aldán was mainly due to the contribution of amphipods (79 species). For example, Garmendia et al. (1998) found 66 amphipod species in Ría de Ares-Betanzos, Conradi et al. (1997), recorded the presence of 53 amphipod species in Algeciras Bay, Parker found 26 subtidal amphipod species in Belfast Lough, Jimeno and Turon (1995) found 71 gammaridean amphipods along the Catalonian coast and Arresti et al. (1986) found 40 amphipod species in the Abra de Bilbao (País Vasco, Spain). Again, this gammaridean diversity was larger than in other similar areas, apart from the English Channel where Dauvin et al. (2000) reported 142 gammaridean amphipods.

Thus, the total number of species of the Ría de Aldán is high enough to consider those soft-bottoms as particularly rich in peracarids. Polychaetes and molluscs also show a high number of species in this ria (Lourido et al. 2006, 2008). This fact may be due to the granulometric heterogeneity existing in this area. Normally, heterogeneous sediments provide many microhabitats which may support a greater biodiversity of species than homogenous sediments do (Gray 1974).

The second peracarid groups in number of species were isopods and cumaceans; whereas, the former was more widespread, the latter was more abundant, both being found from gravel to muddy sediments. Isopods live in most marine habitats, being more diverse in the deep sea (Kensley 1998). Cumaceans can live in all kind of benthic habitats, mud, sand, gravel and on natural rock formations associated with algae or with sessile invertebrates (Alfonso et al. 1998). Nevertheless, other authors state that cumaceans may be more abundant with higher organic matter content and higher proportion of silt/clay in the sediment (Corbera and Cardell 1995).

In the Ría de Aldán, cumaceans showed their highest abundances in muddy sand, being Cumella sp. and Iphinoe serrata Norman, 1867 the most abundant species.

Mysids were poorly represented in our samples. This may happen due to the fact that mysids are sampled more efficiently by using epibenthic sledges (Brandt 1995) rather than the grab used here. Tanaids showed a small number of species but a high number of individuals. This fact is due to the numerical dominance of Apseudes latreillii at some stations. Although this species may proliferate in conditions of organic enrichment (Grall and Glémarec 1997), content of organic matter was not high at stations where Apseudes latreillii was dominant. In the Ría de Aldán, this species showed a high number of individuals in different kind of sediments [1,490 individuals/m2 in st. 34 (muddy sand), 641 in st. 24 (coarse sand) and 605 in st. 22 (fine sand)].

Peracarid assemblages and environmental conditions

Two major peracarids assemblages were determined in Ría de Aldán through multivariate analyses whose distribution agreed mostly with that of granulometric fractions. Similarly, the distribution of molluscan and polychaete assemblages in the Ría de Aldán also shows the same pattern (Lourido et al. 2006, 2008).

Groups A1 and A2b can be included within the ‘Abra alba community’ (Petersen 1918). This community has been reported along European coasts in different types of muddy sediments (Glémarec 1964; Cabioch 1968; Gentil et al. 1986; Carpentier et al. 1997) as well as in other Galician rias (Cadée 1968; Olabarría et al. 1998; Moreira et al. 2005). The most abundant peracarids in the muddy sediments of group A1 were Metaphoxus simplex (exclusive), Harpinia pectinata (constant) and Microdeutopus armatus (exclusive). Constant species of this group were Eudorella truncatula (Bate, 1859), Tanaissus lilljeborgi, Leucothoe incisa and Leptochelia savignyi (Kröyer, 1842). In group A2b (muddy sand and fine sand sediment), Photis longipes (constant), Gammarella fucicola (constant) and Apseudes latreillii had a high abundance (>1,500 individuals/m2). Dexamine spinosa (Montagu, 1813) and Phtisica marina Slabber, 1,769 were constant and exclusive species of this group.

Faunal composition of group A2a (fine sand) can be ascribed to the Venus gallina community (Thorson 1957). In this community, Siphonoecetes kroyeranus (constant) and Apseudes latreillii (constant) were the most abundant species. Furthermore, Bathyporeia elegans Watkin, 1938 was constant and exclusive, Ampelisca sp. exclusive and Perioculodes longimanus, Ampelisca brevicornis (Costa, 1853) and Leucothoe incisa were constant.

The most abundant peracarid in group B1 (medium sand) was the amphipod Siphonoecetes kroyeranus. The characteristic species were Eurydice truncata, Iphinoe trispinosa, Lepidepecreum longicornis and Pseudocuma longicorne (Bate, 1858) (all constant) and the amphipod Leucothoe spinicarpa (Abildgaard, 1789) (exclusive). The mysid Gastrosaccus spinifer (Goës, 1864) (constant and elective) was also a typical species of this kind of sediment.

The coarse sandy sediments of group B2 has a fauna that could be included among the different varieties of the ‘Branchiostoma lanceolatum-Venus fasciata community’ (Thorson 1957). Several authors have reported the presence of similar faunal associations in other areas of Galicia such as the Ría da Coruña (López-Jamar and Mejuto 1985), the Ría de Ares-Betanzos (Troncoso et al. 1993, 2005) and the Ensenada de Baiona (Moreira et al. 2005), and in other areas outside of Galicia, such as the Baie de Morlaix in the Manche Occidental (Dauvin 1988b).

Apseudes latreillii was the most abundant peracarid of this group. Guernea coalita, Haplostylus sp. and Atylus vedlomensis were constant species, while Socarnes erythrophthalmus Robertson, 1892 and Ceradocus semiserratus (Bate, 1862) and Ampelisca spinipes Boeck, 1861 were exclusive.

Among factors determining distribution of peracarids and composition of assemblages in sediments are temperature, stability of substrate, grain size, organic matter content, food availability, burrowing ability, the role of pollutants and diel activity changes determined by specific behavioural patterns (endogenous rhythms, predation, etc.) (Corbera and Cardell 1995; Weisshappel and Svavarsson 1998; Cunha et al. 1999). Grain size is, however, one of the most often reported (Robertson et al. 1989).

Our results suggest that sedimentary composition and a number of environmental gradients existing in the ria related to depth such as hydrodynamism, sedimentation, carbonates, organic matter and the presence of seaweeds are the major factors controlling peracarid spatial distribution. The number of species tended to be higher in fine sand and coarse sand assemblages than in muddy sediments. Similarly, Biernbaum (1979) reported higher number of species in coarse sand and Dauvin et al. (1994) found a high peracarid diversity in circalittoral coarse sands from the English channel. In the Ría de Aldán, total number of individuals is higher in muddy sand sediments than in other sediments. This fact can be due to the great variety of habitats present in sampling stations close to the river mouth which are, in addition, colonized by a number seaweeds. The presence of seaweeds increases the number of microhabitats and there are more ecological niches. Furthermore, seaweeds contribute to stabilize the sediment, food availability is greater than in naked sediments, and seaweeds give protection against a number of predators.