The Spatial Distribution of Caprellidea (Crustacea: Amphipoda): A Stress Bioindicator in Ceuta (North Africa, Gibraltar Area)
Abstract
Abstract. The caprellid (Crustacea: Amphipoda) community associated with the alga Cystoseira usneoides (L.) Roberts 1967 was studied on a spatial scale in relation to the influence of environmental factors on the coast of Ceuta (North Africa) using multivariate analyses. Twenty-two stations were sampled and five environmental factors were tested (hydrodynamics, silting, suspended organic matter, organic matter in silt and suspended solids). The spatial distribution of the caprellid community reflected the physico-chemical conditions of the coast. Phtisica marina Slabber 1769 was the only species present at the stations of the harbour of Ceuta, characterised by the lowest values of hydrodynamics and the highest values of silting, suspended organic matter and suspended solids. The most exposed stations (high hydrodynamics and low values of silting, dissolved organic matter and suspension solids) were mainly dominated by Caprella danilevskii Czerniavskii 1868 and C. penantis Leach 1814. These two species have developed a “parallel” posture that enables them to attach strongly to the substratum and thus inhabit areas subjected to heavy exposure or strong currents. Caprella acanthifera Leach 1814 and Pseudoprotella phasma (Montagu 1804) preferred stations characterised by moderate values of silting and suspended solids, such as those located in the coastal areas in front of the city of Ceuta, but these species were not found at the most stressed harbour stations.
Problem
Although caprellids (Crustacea: Amphipoda: Caprellidea) inhabit most marine habitats, their value as a bioindicator of environmental conditions has been scarcely studied. Caprellids have been included in general pollution studies (marine ecosystems: Takahashi et al., 1999; arthropods: Sánchez-Moyano & García-Gómez, 1998; amphipods: Bellan-Santini, 1981; Conradi et al., 1997), but no studies focusing exclusively on the Caprellidea as bioindicators have yet been developed.
Only 33 of the more than 300 worldwide caprellid species have been found in Mediterranean waters (Krapp-Schickel, 1993; Sánchez-Moyano et al., 1995a, b; Krapp-Schickel & Vader, 1998). The ubiquity of the Caprellidea, which live on algae, hydroids, bryozoans, anthozoans and sediments along with the limited number of species inhabiting Mediterranean waters (which are relatively easy to identify) could turn them into a potentially very useful group in environmental studies.
Macrophytes constitute a very important substratum in benthic communities. In general, macrophytes have greater species richness, abundance and biomass of associated fauna than unvegetated habitats (Edgar, 1990). They provide protection against wave impact and hydrodynamic forces (Hicks, 1980; Buschmann, 1990) and give shelter to predators, especially fishes (Stoner, 1980; Edgar, 1990; Russo, 1991). The seaweed genus Cystoseira C. Agard 1820 is very important in Mediterranean waters from the floristic and ecological point of view. The dominant algal species of the investigated Mediterranean shallow benthic communities is Cystoseira usneoides (L.) Roberts 1967; it is the most abundant and widely distributed seaweed on the coastline of Ceuta (North Africa, Gibraltar area) (Romero, 1994).
The main objective of the present work was a spatial study of the caprellids associated with C. usneoides, analysing their value as bioindicators of environmental conditions (hydrodynamics, silting, suspended solids, suspended organic matter and organic matter in silt).
Material and Methods
Ceuta is a Spanish enclave located in North Africa, Gibraltar area (Fig. 1). The coastline, about 20-km long, is composed of heterogeneous environments imposed by the natural configuration of the coast as well as by intense port activity. At least 70 % of Ceuta’s coast is natural hard bottom.
Location of the sampling stations along the coast of Ceuta.
In a general study of the Caprellidea from Ceuta, 19 species were found living in various substrata along the coast (Guerra-García & Takeuchi, in press). Cystoseira usneoides is the most abundant and widely distributed seaweed here. Its population is present year-round throughout Ceuta, with maximum growth in summer. This algal species was selected as a substratum in order to eliminate the variation in community structure caused by the configuration of the different substrata (Sánchez-Moyano & García-Gómez, 1998).
A total of 22 stations distributed along the coast of Ceuta were chosen to encompass the broadest range of environmental conditions (Fig. 1). Stations were selected approximately at the same depth (between 5 and 10 m) in order to avoid depth-related effects. Cystoseira usneoides was collected by SCUBA diving in August 1999 (maximum growth) in all stations.
Each sample was placed in a plastic bag and fixed with a 10 % formalin-seawater solution. Four replicate samples (see Conradi et al., 1997; Sánchez-Moyano & García-Gómez, 1998) were taken at each station. All the replicate samples had approximately the same size (dry weight 35 – 50 g). All samples were sorted out under a dissecting microscope. The caprellids were separated, preserved in 70 % ethanol, identified to species and counted. The abundance data were expressed as number of individuals per 100 g dry weight algae.
The physico-chemical factors measured were hydrodynamics, silting, suspended solids, suspended organic matter and organic matter in the silt gathered in collecting bottles (Carballo et al., 1996; Sánchez-Moyano & García-Gómez, 1998). To measure silting (sedimentation rate) and hydrodynamics, a set of bottom-mounted frames was placed in the 22 stations, each one having 6 sediment-collecting bottles (Moore, 1972) for silting and 6 plaster spheres for hydrodynamics. The spheres were submerged for 72 h and the bottles for the whole month (August 1999). To estimate the hydrodynamics the method of plaster dissolution described by Muus (1968) and modified by Gambi et al. (1989) was used. The hydrodynamics was calculated in water speed equivalents (V) according to the formula suggested by Bailey-Brock (1979). The suspended solids, suspended organic matter and perecentage organic matter in the silt were measured using methods described by Strickland & Pearson (1960). Six samples were also analysed for suspended solids and suspended organic matter.
Multivariate analyses (classification and ordination) were carried out using the program PC-ORD v. 3.05 (McCune & Mefford, 1997). The affinities among stations were established through cluster analysis using the UPGMA method (unweighted pair-group method using arithmetic averages) (Sneath & Sokal, 1973), based on the Bray-Curtis similarity index for the species matrix and on the euclidean distance for the environmental matrix. Abundance data of the caprellid species were double square root transformed so that the ensuing classification and ordination were not determined only by the most dominant species (Clarke & Green, 1988). Environmental data were transformed with the log (x+1) (Estacio et al., 1997). In order to confirm the results of the cluster, an MDS analysis (non-metric multidimensional scaling) was used with the matrix of caprellid species. To test the ordination, the stress coefficient of Kruskal was employed (Kruskal & Wish, 1978). To ascertain whether environmental variables influence community composition, a canonical correspondence analysis (CCA) was applied. To avoid the distortion in the analysis caused by the rare species, these were down-weighted. The Monte Carlo test was applied to verify the statistical significance of the analysis.
Results
The Caprellidea fauna associated with the alga Cystoseira usneoides is shown in Table 1 and the general trend of abiotic variables along the coast of Ceuta is shown in Fig. 2. The cluster analysis using the environmental matrix of physico-chemical parameters shows that two station groups can be differentiated (Fig. 3A). The first includes stations E8, E9 and E10, located inside Ceuta’s harbour and characterised by the lowest values of hydrodynamics and the highest values of silting, suspension solids and organic matter. The second group includes the remaining stations. This group can be split into two subgroups. The stations E1, E2 (Benzú), and E13, E14, E15, E16 (Rompecala-Desnarigado), located at the most exposed littoral areas, are characterised by clean waters and strong currents (high hydrodynamics). The remaining stations, most of them adjacent to the city of Ceuta, present intermediate characteristics between the harbour and Benzú as well as Rompecala-Desnarigado.
| E1 | E2 | E3 | E4 | E5 | E6 | E7 | E8 | E9 | E10 | E11 | E12 | E13 | E14 | E15 | E16 | E17 | E18 | E19 | E20 | E21 | E22 | |
| Caprella acanthifera Leach 1814 | 1.1 | – | 5.5 | 22 | 4.0 | 15.0 | 9.0 | – | – | – | 29.0 | 16.0 | – | 1.6 | – | – | 29.0 | 22.0 | 9.0 | 32.0 | 47.0 | 5.3 |
| (0.3) | (0.7) | (9.1) | (0.9) | (7.5) | (1.7) | (4.9) | (3.1) | (0.2) | (8.8) | (9.8) | (1.7) | (6.3) | (7.0) | (0.2) | ||||||||
| C. danilevskii Czerniavskii 1868 | 22.0 | 20.0 | 12.0 | 8.5 | – | – | – | – | – | – | – | – | – | 1.3 | – | – | – | – | – | – | – | – |
| (6.3) | (2.7) | (3.5) | (2.0) | (0.3) | ||||||||||||||||||
| C. fretensis Stebbing 1878 | – | 0.8 | – | – | – | – | – | – | – | – | – | – | – | – | – | – | – | – | – | – | – | – |
| (0.5) | ||||||||||||||||||||||
| C. liparotensis Haller 1879 | – | – | – | – | – | – | – | – | – | – | – | – | – | – | – | 3.2 | – | – | – | – | – | – |
| (0.9) | ||||||||||||||||||||||
| C. penantis Leach 1814 | 1.7 | 4.9 | 2.5 | – | – | – | – | – | – | – | – | – | 20.0 | 36.0 | 28.0 | 46.0 | – | – | – | – | – | – |
| (0.2) | (2.6) | (0.4) | (3.8) | (9.1) | (2.9) | (8.0) | ||||||||||||||||
| C. santosrosai Sánchez-Moyano et al. 1995b | – | – | – | – | – | – | – | – | – | – | – | – | – | – | 0.9 | 11.0 | – | – | – | – | – | – |
| (0.8) | (5.9) | |||||||||||||||||||||
| Phtisica marina Slabber 1769 | 11.0 | 13.0 | 26.0 | – | 12.0 | – | 32.0 | 86.0 | 68.0 | 97.0 | 12.0 | 16.0 | 6.0 | 2.0 | 1.7 | 0.1 | – | 17.0 | 41.0 | 36.0 | 45.0 | 24.0 |
| (6.9) | (6.8) | (7.3) | (9.5) | (5.3) | (9.9) | (9.7) | (5.6) | (2.6) | (3.9) | (4.9) | (1.8) | (1.0) | (0.1) | (3.9) | (8.7) | (2.9) | (9.0) | (6.6) | ||||
| Pseudoprotella phasma (Montagu 1804) | – | – | – | 0.7 | 49.0 | 0.1 | 6.9 | – | – | – | 1.1 | 3.5 | – | – | – | – | – | – | 19.0 | 19.0 | 12.0 | 30.0 |
| (0.1) | (7.7) | (0.1) | (5.4) | (1.1) | (2.9) | (2.5) | (3.7) | (8.1) | (8.5) |
Multiple ‘box-and-whisker’ plot for each abiotic variable at the 22 sampling stations. Median values are included; the rectangles contain values between the first and the third quartiles; the bars connect the extreme values. Six data of each variable were obtained.
A. Dendrogram of similarity between the stations based on the environmental data. B. Dendrogram of similarity between the stations based on the average abundance of the caprellid species.
The cluster analysis (Fig. 3B) based on the Caprellidea associated with Cystoseira usneoides (Table 1) was very similar to that obtained with the physico-chemical parameters. The two-dimensional representation of the ordination analysis MDS (Fig. 4) confirmed the groups and the affinities among stations based on the abundance data.
MDS ordination of the stations according to the abundance of the caprellid species (stress: 0.035).
The relation between the Caprellidea and the environmental variables is evident in the canonical correspondence analysis (CCA) (Fig. 5 and Table 2). The first axis is practically determined by the hydrodynamics and the sedimentation rate. The most exposed stations are mainly dominated by Caprella danilevskii and C. penantis. Caprella fretensis, C. santosrosai and C. liparotensis, although very infrequent species, were also found at stations with high hydrodynamics. On the other hand, C. acanthifera, Phtisica marina and Pseudoprotella phasma inhabited the stations characterised by lower hydrodynamism. The second axis is determined basically by the dissolved organic matter, but also by the suspension solids and organic matter in the silt. According to these water characteristics P. marina can be distinguished from P. phasma and C. acanthifera. While P. marina showed the highest densities inside the harbour and was the only caprellid species there, P. phasma and C. acanthifera seemed to prefer stations less enriched in organic matter and suspended solids than the harbour stations.
Graph representation of the stations and species with respect to the first two axes of the canonical correspondence analysis (CCA).
| axis 1 | axis 2 | axis 3 | |
| species-environment correlation | 0.86 | 0.79 | 0.55 |
| percentage of species variance | 26.90 | 9.30 | 3.00 |
| correlation with environmental variables: | |||
| hydrodynamics | – 0.79*** | – | – |
| silting | 0.44* | – | – |
| suspended solids | – | 0.62** | – |
| suspended organic matter | – | 0.73*** | – |
| organic matter in silt | – | 0.53** | – |
Discussion
Although the number of studies on the ecological and behavioural aspects of the Caprellidea are increasing (e.g., density fluctuation and population ecology: Bynum, 1978; Takeuchi et al., 1990; Sconfietti & Luparia, 1995; life-histories under laboratory conditions: Takeuchi & Hirano, 1991, 1992; clinging behaviour: Takeuchi & Hirano, 1995; Guerra-García, unpublished data) no work has yet been published on the relation between physico-chemical parameters and the specific community of the Caprellidea.
The present study suggests that the caprellid community can be used as a bioindicator of environmental conditions. The number of caprellid species inhabiting the seaweed Cystoseira usneoides along the coast of Ceuta is low; nevertheless, the ecological requirements of these species are very different.
Phtisica marina was the most abundant and frequent species here. Although it was present in almost all stations, the highest densities were recorded inside the harbour. Furthermore, P. marina was the only caprellid able to resist the stress conditions inside the harbour and to reduce competition with other species by increasing its own density. Phtisica marina is an almost cosmopolitan caprellid (Bellan-Santini & Ruffo, 1998) with a relatively wide ecological distribution (Bellan-Santini, 1998; Guerra-García, unpublished data). Sánchez-Moyano & García-Gómez (1998), in a general study of the arthropod community as a bioindicator in Algeciras bay, found P. marina at all stations; it was especially abundant at the calm internal zones of the bay. Conradi et al. (1997) also reported P. marina and Pseudoprotella phasma as species typical of environments with low hydrodynamics and high sedimentation, such as harbours. Although we found no P. phasma inside the harbour in our study, this species, along with C. acanthifera and P. marina, was dominant at the stations surrounding the city of Ceuta. These are not as ‘perturbed’ as inside the harbour, but have moderate values of suspended organic matter, suspended solids and sedimentation rate. Caprella acanthifera has been associated with areas characterised by low water movement and rich detritus (Krapp-Schickel & Vader, 1998).
On the other hand, we found C. danilevskii and C. penantis living in the most ‘unperturbed’ areas with high hydrodynamics and low values of organic matter and suspended solids. Recent studies on clinging behaviour (Takeuchi & Hirano, 1995; Guerra-García, unpublished data) pointed out that these two species have developed a “parallel” posture with the body pereonites straight and parallel to the substratum. These caprellids spent most of the time grasping the substratum with gnathopod 1 and pereopods 5 – 7. These “parallel” caprellids have acquired morphological characteristics (elongate somites, short basis of gnathopod 2, gills bent basally etc.) that specifically enable them to attach strongly to the substratum and thus inhabit areas subjected to heavy wave exposure and/or strong currents (Takeuchi & Hirano, 1995). Our finding of these caprellids at the stations with high hydrodynamics is in agreement with the above behavioural observations.
Conclusions
Today, society is increasingly demanding quick and effective environmental studies in coastal areas. This precludes developping long temporal series of data with complex matrices of various taxonomic groups, whose identification requires great effort. The study of the caprellid community associated with a macrophytic species is presented here as an alternative strategy. A short-term spatial study dealing with only several easy-to-identify species can yield results similar to those obtained with costly physico-chemical analysis.
Acknowledgements
We express our thanks to Compañía del Mar and Club Calypso for assistance in the field. Our gratitude to Asamblea de Ceuta and a grant “Programa de Formación de Profesorado Universitario AP98/28 617 065” from the Ministry of Education and Culture of Spain for financial support.
