Conflicts of interest The authors declare no conflicts of interest.
Maria Cristina Follesa, Department of Animal Biology and Ecology, University of Cagliari, Via T. Fiorelli n.1, 09126 Cagliari, Italy. E-mail: firstname.lastname@example.org
Community structure and faunal composition of bathyal decapod crustaceans off South-Eastern Sardinian deep-waters (Central-Western Mediterranean) were investigated. Samples were collected during 32 hauls between 793 and 1598 m in depth over the 2003–2007 period. A total of 1900 decapod specimens belonging to 23 species were collected. Multivariate analysis revealed the occurrence of three faunistic assemblages related to depth: (i) an upper slope community at depths of 793–1002 m; (ii) a middle slope community at depths of 1007–1212 m and (iii) a lower slope community at depths greater 1420 m. In the upper and middle slopes the benthic (Polycheles typhlops) and epibenthic–endobenthic feeders (mainly Aristeus antennatus and Geryon longipes), which eat infaunal prey, were dominant, followed by the macroplankton–epibenthic feeders such as Acanthephyra eximia and Plesionika acanthonotus. In the deepest stratum, the most remarkable feature was the prevalence of macroplankton–epibenthic feeders (A. eximia and P. acanthonotus). A small percentage of the benthic deep-sea lobster Polycheles sculptus was also present. The biomass presented higher values in the middle slope and declined strongly in the lower slope. There was no general pattern of mean individual weight/size versus depth among decapods, and the changes seemed to be species-specific with different trends.
The study of the distribution of species along environmental gradients has traditionally been important for characterising the organisation of animal communities in aquatic systems (Wenner & Boesch 1979). In particular, in deep-sea marine environments, species and communities often change with increasing depth rather than along horizontal gradients (Gage & Tyler 1991; Cartes et al. 2004, 2007a), suggesting the idea of depth bands of high faunal homogeneity separated by boundaries of faunal renewal. Moreover, the structure of megafaunal assemblages in the continental slope is largely determined by a combination of both abiotic (the structure and type of the bottom and oceanographic conditions) and biotic (resource availability and food web structure) factors (Moranta et al. 1998, 2008), which may also play a fundamental role in the local zonation pattern (Maynou & Cartes 2000; Carbonell et al. 2003).
In Sardinian waters (Central-Western Mediterranean), the knowledge available on the deep-sea decapod crustaceans concerns the bio-ecology (mainly reproduction and trophism) of a few deep-sea species (Mura et al. 1993; Follesa et al. 2007; Cabiddu et al. 2008). In fact, the studies in this area are limited to the epibathyal and the mesobathyal levels (mostly at depths of less than 750 m) generally subject to intense trawl fishing. Mura (1987) and Mura & Cau (1992, 1994) described the faunistic composition and bathymetric distribution of the decapod crustaceans present in the lower part (down to 1050 m) of the mesobathyal zone of the Sardinian Channel.
The object of this paper was to update the data on bathyal decapod crustaceans (faunal composition, bathymetric distribution, zonation, abundance, biomass and length frequency distribution) in Sardinian waters (Central-Western Mediterranean). Despite the limitations of bathyal sampling, this study represents the first attempt to identify the deep-sea crustacean assemblages down to 1000 m and to improve knowledge of this area, considered the link between the Western and Eastern Mediterranean (Hopkins 1988).
The data analysed in the present work came from 32 hauls carried out during experimental trawl survey cruises intended to study the bottom-living community in the continental slope. Sampling was conducted on compact mud bottoms off South-Eastern Sardinian deep-waters (Sardinian Channel, Central-Western Mediterranean) (Fig. 1). All hauls were performed at main depths of 793–1598 m over the 2003–2007 period.
On board, trawl data (date, position and duration) were recorded (Table 1). The duration of each haul (bottom time) varied between 64 and 164 min. The towing speed was about 2.2 knots for all hauls. The otter trawl used was equipped with a 20-mm stretch mesh size cod end. Gear selectivity was assumed to be constant because the same fishing gear for each trawl was used.
Table 1. Characteristics of hauls (date, mean depth, position, time period of hauls) carried out off South-Eastern Sardinian deep-waters (Central-Western Mediterranean).
mean depth (meters)
haul duration (minutes)
Usually, the monitoring of crustacean assemblages is confined to the economically important depths (<750 m). For this reason, due to the lack of long-time series of data on deep-sea assemblages, catches from different years and seasons have been pooled in our analysis.
For each haul, crustaceans were sorted by species and abundance (N) and biomass (g) data were noted and standardised to 1 h of haul (N·h−1). In the laboratory, the carapace length (CL, in mm, from the posterior margin of the eye socket to the posterior end of the carapace), individual weight (g) and sex of each species were also determined. Moreover, the crustaceans were classified as mesopelagic species, nektobenthic species or benthic species on the basis of their relative location in the water column (Maynou & Cartes 2000).
Standardised abundance data of decapod crustaceans were pooled in a matrix of species abundance with the primer (v6) package. Cluster analysis was performed using the Bray–Curtis similarity index and group linkage was used for the dendrogram (Bray & Curtis 1957; Field et al. 1982). Prior to analysis, standardisation and fourth-root transformation were applied.
Analysis of similitude (anosim) was used to test the significance of species assemblages between trawl surveys (Clarke 1993) and similarity percentage (simper) was also applied to detect bathymetric differences.
The ecological parameters such as mean abundance (N·h−1) and mean biomass (g·h−1), and the measures of species diversity, such as species richness (S) (DIVERSE routine), Shannon–Wiener index (H’) (Shannon & Weaver 1949) and evenness (J’) (Pielou 1977) were calculated for all hauls of the main groups obtained by prior classification.
Between-assemblage differences in species richness, diversity indexes, abundance and biomass spectra were tested using repeated anova measures (Zar 1999). Each haul was considered an observation and the data of abundance and biomass were normalized [ln(x + 1)] to adjust residuals to normality.
A bubble scatterplot was applied to the individual mean weight (g) of the most frequent species to detect the abundance tendencies with depth.
To show the bathymetric distribution of the main species, the overall length frequency distribution (2 mm size class) by sex for each depth interval was determined.
During the trawl surveys, a total of 1900 individuals (total weight 11,919 g), subdivided in 23 species of bathyal decapod crustaceans, were collected and are listed in Table 2. Within the investigated depth range, Plesionika acanthonotus was present in all hauls and Polycheles typhlops, Sergia robusta (97% frequency of occurrence) and Aristeus antennatus (91% frequency of occurrence) were present at nearly all stations. A high frequency of occurrence (%F > 50) of Acanthephyra eximia, Acanthephyra pelagica, Ponthophilus norvegicus and Geryon longipes was observed (Table 2). In terms of total abundance, the predominant species were found to be A. antennatus (23.6%) and P. typhlops (18.4%). In terms of biomass, A. antennatus (31.6%), and G. longipes (21.05%) were the species with the highest percentage of contribution (Table 2).
Table 2. Bathyal decapod crustacean species collected off South-Eastern Sardinian deep-waters (Central-Western Mediterranean) between 793 and 1598 m with their bathymetric range, occurrence’s frequency (%F), percentage of abundance (%Abundance) and biomass (%Biomass).
depth range (m)
Aristaeomorpha foliacea (Risso, 1827)
Aristeus antennatus (Risso, 1816)
Gennadas elegans (S. I. Smith, 1882)
Parapenaeus longirostris (H. Lucas, 1846)
Sergestes arcticus Krøyer, 1855
Sergia robusta (S.I. Smith, 1882)
Acanthephyra eximia S. I. Smith, 1884
Acanthephyra pelagica (Risso, 1816)
Pasiphaea multidentata Esmark, 1866
Plesionika acanthonotus (S. I. Smith, 1882)
Plesionika martia (A. Milne-Edwards, 1883)
Pontocaris lacazei (Gourret, 1887)
Pontophilus norvegicus (M. Sars, 1861)
Calocaris macandreae Bell, 1853
Polycheles sculptus S. I. Smith, 1880
Polycheles typhlops Heller, 1862
Pagurus alatus (Fabricius, 1775)
Munida tenuimana G. O. Sars, 1871
Monodaeus couchii (Couch, 1851)
Paromola cuvieri (Risso, 1816)
Geryon longipes A. Milne Edwards, 1881
Bathynectes maravigna (Prestandrea, 1839)
Macropipus tuberculatus (Roux, 1830)
The similarity dendrograms of the trawls revealed the presence of three main groups (Fig. 2), which can be clearly identified along the bathymetric gradient. The first group consisted of the deepest stations covering a depth range of 1420–1598 m (six hauls, mean depth 1475.5, SD 85.57) (lower slope) (Fig. 2). A second group was made up of stations investigated at 793–1002 m depth (eight hauls, mean depth 944.5, SD 80.48) (upper slope). The third aggregation consisted of six hauls (mean depth 1107, SD 64.06) carried out at middle depths of 1007–1212 m (middle slope).
The pair-wise test comparisons (anosim) showed, in terms of abundance, a low level of overlap between the hauls (R = 0.534, P < 0.01).
The results of the SIMPER routine showed a high percentage of similarity for the assemblages identified by clustering (Table 3). In the upper slope (793 and 1002 m), the species which took part in the assemblage appeared to be P. typhlops, A. antennatus and S. robusta (59.06%). Aristeus antennatus, P. typhlops, P. acanthonotus and S. robusta contributed 58.74% of the middle slope assemblage (1007–1212 m). The most typical species of the deepest bottoms (1420–1598 m) were A. eximia (19.10%), P. acanthonotus (14.83%) and A. antennatus (14.13%). Moreover, in the lower slope, Polycheles sculptus was a common species (13.65%).
Table 3. Results of the SIMPER routine to analyse the percentage contribution of typifying species (over 7%) to within-group similarity resulting from cluster analysis for crustacean samples during trawl surveys off the South-Eastern Sardinian deep-waters (Central-Western Mediterranean).
average similarity: 73.74
average similarity: 72.41
average similarity: 75.59
Mean values of the ecological parameters of each assemblage are reported in Table 4. Significant differences in species richness (S) were observed (F11.86; P < 0.05). The highest value (an average of 2.47 species) was found on the middle slope, followed by the lower (2.32 species) and the upper slope (1.79 species). The highest diversity (H’) was obtained for the lower (H’ = 1.89, J’ = 0.86) and the middle slope (H’ = 1.8, J’ = 0.81), with significant differences between assemblages (F5.25; P < 0.05).
Table 4. Some ecological parameters (mean and deviation standard) in the three groups resulting for the cluster analysis.
mean abundance (N·h−1)
24 ± 10.78
36 ± 13.31
37 ± 12.03
mean biomass (g·h−1)
205.48 ± 141.38
224.25 ± 106.27
121.81 ± 42.80
number of species
mean species richness (S)
1.79 ± 0.28
2.47 ± 0.34
2.32 ± 0.32
1.56 ± 0.26
1.81 ± 0.20
1.89 ± 0.20
0.83 ± 0.11
0.81 ± 0.08
0.86 ± 0.03
The mean values of abundance showed an increase with depth (24, 36, 37 individuals per hour, respectively), with significant differences between assemblages (F3.43; P < 0.05), whereas no significant difference (F1.95; P > 0.05) was observed between the mean biomass that showed a maximum value in the middle stratum.
The relative abundance analysis of the species caught in each identified group (Fig. 3A,C,E) generally highlighted a species’ dominance similar to that obtained from the simper analysis (Table 3). Moreover, in terms of relative biomass, A. antennatus, P. typhlops and G. longipes were the most abundant species in the upper and middle slopes (Fig. 3B and D). In the deepest strata, A. eximia and A. antennatus were the predominant species (Fig. 3F).
Regarding their relative depth distribution, in the shallowest stations the crustacean decapods could be characterised mainly by nektobenthic and benthic species (43% and 38% in number and 44% and 45% in weight, respectively) (Fig. 5). In the middle slope, nektobenthic species again dominated both in weight and number, while the benthic species were also present with a high percentage in biomass (44%), probably due to the occasional presence of large-bodied species such as the brachyuran crab G. longipes. The deepest range was inhabited by nektobenthic species (principally A. eximia and A. antennatus) but also by benthic decapods (29%), which became significant in number probably as consequence of the high abundance of the small deep-sea lobster P. sculptus.
The bubble plot for the bathyal decapod crustaceans showed different mean weight tendencies with depth depending on the species (Fig. 4). Aristeus antennatus, A. eximia, A. pelagica, P. typhlops and G. longipes showed a negative correlation between individual mean weight and depth, probably due to the recruitment of small individuals in the deepest waters. Otherwise, Munida tenuimana and S. robusta presented a ‘bigger-deeper’ trend, with mean weight increasing with depth. On the lower slope, the deepest species P. sculptus and P. norvegicus, which were captured for the first time below 1000 m, were characterized by small to medium sized individuals.
Figure 5 shows the size frequencies and sex distribution by depth interval of the main bathyal decapod species. Juveniles of A. antennatus (CL < 20 mm; Sardàet al. 2004) were most representative in the deepest part of range (below 1420 m), with a high percentage of females; between 793 and 1212 m (middle slope), the adults appeared well represented, with an elevated proportion of males. On the upper and middle slopes, A. eximia showed a range in size of 20–38 mm CL, with males mainly represented only by the smallest size class, whereas on the lower slope, juveniles (CL 14–18 mm) were also present. Plesionika acanthonotus showed a range in size of between 4 and 22 mm CL and a sex-ratio in favour of females increasing with depth. An inverse sex-ratio was observed in S. robusta (range 8–26 mm CL). It was difficult to find a clear pattern for the population structure of P. norvegicus because the individuals were present exclusively below 1000 m and were relatively scarce in all depth intervals.
Our results confirm the importance of decapod crustaceans in Mediterranean deep-sea benthic communities because they may be more competitive than other invertebrate groups, in contrast to more productive oceans like the Atlantic (Tyler & Zibrowius 1992).
The bathyal decapod crustacean community of the South-Eastern Sardinian deep-waters (Central-Western Mediterranean) presented a clear zonation effect, with a series of well-defined bathymetric boundaries that seemed to be connected to depth-related factors. Three faunistic assemblages along the continental slope were identified by means of cluster analysis. The lowest values of species richness (S) were found in the upper and lower slope, which could be explained best by a decrease in food supply enhancing competitive exclusion of the species; the highest mean value of species richness was found in the mid-bathyal interval (1002–1212 m), probably implying low trophic pressure or diminished competition with fish (Maynou & Cartes 2000). This phenomenon, also reported in various taxa among macrofauna (e.g. gastropods; Rex 1973), generates a typical bell-shaped response along depth or other environmentally mediated gradients (Gage & Tyler 1991).
Following the feeding classification of Cartes (1998) for the main bathyal decapod species in the Catalan Sea, our study area showed in the upper and middle slope (between 793 and 1212 m), a prevalence of benthic (P. typhlops) and epibenthic–endobenthic feeders (mainly A. antennatus and G. longipes) that eat infaunal prey, with a low percentage of macroplankton–epibenthic feeders such as A. eximia and P. acanthonotus. In contrast, in the deepest stratum [the main boundary similar to the lower subzone defined by Pérès (1985), Cartes & Sardà (1993) and Stefanescu et al. (1993)] the most remarkable feature was the prevalence of macroplankton–epibenthic feeders (A. eximia and P. acanthonotus), followed by the benthic deep-sea lobster P. sculptus. A similar distribution of feeders in the continental slope was found by Maynou & Cartes (2000) off the South-West Balearic Islands (Western Mediterranean), where the distance from the mainland and the absence of submarine canyons justify the low values of superficial primary production and the consequent dominance of the macroplankton–epibenthic feeders. This result differs greatly from what was registered in the Catalan Sea (Cartes et al. 1994), where advective inputs of organic carbon via submarine canyons represent an additional contribution to deposit feeders and epibenthic–endobenthic feeders (Maynou & Cartes 2000). This phenomenon, supporting the food availability and local geographic conditions as responsible for species distribution, also highlighted the presence of a remarkable west–east productivity gradient in the Mediterranean, probably mainly due to the variability in the vertical fluxes of organic carbon to the sea floor (Danovaro et al. 1999). Many studies have compared the phytoplankton pigment concentrations (PPC) in the Mediterranean Sea, highlighting differences of an order of magnitude between the west and eastern basin, and confirming the increase of oligotrophy in the west versus east (Maynou & Cartes 2000; Cartes et al. 2004; Company et al. 2004; Tselepides et al. 2004; Politou et al. 2005).
The decrease of total decapod biomass with depth has been established in oceans worldwide (Haedrich et al. 1980; Lampitt et al. 1986). The data available in the Mediterranean are consistent with the general decrease down to 2200 m in the Western Mediterranean (Cartes & Sardà 1992) and down to 1000 m in the Cretan Sea (Kallianotis et al. 2000). On the whole, this trend was confirmed for the South-Eastern Sardinian deep-waters, where the biomass showed a strong decrease in the deepest slope (1420–1598 m), probably due to the small size of specimens caught (principally A. eximia, P. acanthonotus and P. sculptus). The highest value of biomass was found between 1007 and 1212 m, due to the presence of big size species (A. antennatus and G. longipes).
In conclusion, this study provides useful information about the composition, distribution and structure of bathyal decapod crustaceans in the Central-Western Mediterranean, considered the link between the western and eastern basin. Further investigations should be devoted to increasing the bathymetric range of the research to improve knowledge of the Mediterranean Sea fauna.