Composition and spatial distribution of subtidal Decapoda on the “ Reef Coast ” , northeastern Brazil , evaluated through a low-impact visual census technique

This study investigated the composition and spatial distribution of the sublittoral decapods on the reefs of Porto de Galinhas Beach, southern coast of Pernambuco, Brazil, through the Underwater Visual Census technique. Data were collected monthly, at night during full-moon tides in low tide periods from June 2004 to May 2005, using SCUBA diving and a visual census with a fixed belt transect (20 m long). Three sampling areas were defined: Confined Waters (low hydrodynamics) with shallow sites (up to 2.5 m deep); Semi-open Water (3 to 6 m deep), influenced by waves and tidal currents (moderate hydrodynamics); and Open Water (7 to 10 m deep), in the breaker zone (high hydrodynamics). A total of 6,287 individuals of 34 species belonging to the infraorders Brachyura (19 species), Achelata and Anomura (5 species each), Caridea (3 species), and Stenopodidea and Astacidea (1 species each) were collected. Two decapod assemblages were distinguished: in a habitat with low hydrodynamics and shallow (Confined) water; and in a habitat with moderate to high hydrodynamics and depths of 3 to 10 m (Semi-open and Open water). At the sites with high hydrodynamics, i.e., the Open-water Area in the breaker zone, decapod diversity was significantly lower than in the other, protected areas on the reef bench. These results suggest that the distribution of subtidal decapods on coastal reefs is influenced by depth and exposure to water stress caused by waves and currents (hydrodynamics). The visual census technique with SCUBA proved to be suitable for ecological studies on subtidal decapods.


Introduction
Among marine ecosystems, coral reefs show the highest known diversity of species (Dubinsky and Stambler, 2011;Huang et al., 2011).Brazil has the only coral reefs in the South Atlantic, and one of the bestdeveloped reef communities is the coastal reefs of northeastern Brazil (Leão et al., 2003).line, upward (toward light) to the upper limit of the subtidal zone (often protruding from the water) and expand laterally from the top, forming densely aggregated structures and an interconnected cave system beneath the reef surface.The depth of the reef structures rarely exceeds 10 m (Laborel, 1970;Dominguez et al., 1990;Maida and Ferreira, 1997).Similarly to reefs elsewhere, in Brazil these habitat is subject to intense anthropogenic pressure, aggravated by their proximity to shore (Castro and Pires, 2001;Feitosa et al., 2002;Fernandes et al., 2005;Barradas et al., 2010;Sarmento et al., 2011).
On hard bottoms such as coral reefs, the decapod crustaceans are one of the most important groups of benthic macrofauna (Abele, 1974;1976;Abele and Patton, 1976;Martínez-Iglesias and García-Raso, 1999;Alves et al., 2006).They range from inconspicuous to large forms, including shrimps, lobsters and crabs, and are distinguished by their high diversity and their importance in fisheries and trophic dynamics in these environments (Randall, 1967;Dubinsky and Stambler, 2011).The number of studies of diversity and distribution patterns of decapods on coral reefs is still relatively low, probably because these crustaceans have cryptic habits and live hidden during the day in burrows, crevices and caves (Barreto and Katsuragawa, 2008;Igarashi, 2010;Dubinsky and Stambler, 2011), limiting collection and observation.
This study investigated the composition and spatial distribution of the sublittoral decapods on the coastal reef of Porto de Galinhas Beach, northeastern Brazil, one of the main formations of the Reef Coast.The study was conducted by means of the Underwater Visual Census technique, and included areas with different depths and hydrodynamics.

Study area
The study was carried out on the coral reefs of Porto de Galinhas (8°30'07" -8°30'54"S and 35°00'08" -34°59'47"W) (Fig. 1).These reefs have the typical characteristics described for the Reef Coast (Laborel, 1970;Dominguez et al., 1990;Maida and Ferreira, 1997) (Fig. 1), and are one of the main tourist attractions on the Brazilian coast (MMA/ SECTMA/CPRH, 2003).The Porto de Galinhas reef bench is in direct contact with the shore, which mixes beach and reef ecosystems during low tide when the reef bench emerges, exposing the intertidal zone.The outcropping reef acts as a barrier to the waves breaking in the outer areas.During low tide, the emerged reef bench forms the three areas sampled in this study: A) Confined Water (pools protected by the reefs); B) Semi-open Water (channels crossing the reef bench); and C) Open Water (in the breaker areas) (Fig. 2).
The Confined Waters area (A) is close to the beach and shallow (to 2.5 m deep), and has almost no current and waves (hydrodynamics).This area is exposed to weak currents and waves only during high tide and neap tides (Fig. 2).
The Semi-open Waters area (B) is mainly in the channels connecting the open sea with the protected water inside the reef bench.With depths between 3 to 6 m, this area is influenced by waves and tidal currents (moderate hydrodynamics).During neap high tide, the current is strong and the area is subject to indirect wave influence (Fig. 2).
The Open Water area (C) is in the outer areas of the reef bench, exactly in the breakingwave zone.Around 7 to 10 m deep, this area shows strong hydrodynamics during both spring and neap tides, and at high and low tide (Fig. 2).

Data Collection
The data were collected monthly, at night during the full-moon tide in low tide, from June 2004 through May 2005, by SCUBA diving for 3 h in each dive session (1.5 h before until 1.5 h after low tide), in three sampling areas A) Confined Waters, B) Semi-open Water, and C) Open Water.In each area, four collection points were sampled by counting all individuals by visual census, using fixed belt transects, each 20 m long, totaling 144 transects during the study period.
The chosen monitoring technique was the strip transect technique (STT) (Brock, 1954) with a 20 m-long transect.The area of this medium-scale sampling is based on the volume and size of the largest individuals covered (Hill and Wilkinson, 2004;Bakus, 2007), in this case the large reptant lobsters of the genus Panulirus White, 1847.
The STT is used by monitoring programs worldwide to study underwater communities (including decapods), such as the Australian Institute of Marine Science Longterm Monitoring Program (AIMS LTMP), the various Pacific monitoring programs (Lincoln-Smith transect), the Reef Check MAQTRAC (Marine Aquarium Trade Coral Reef Monitoring Protocol) Program (Hill and Wilkinson, 2004) and the Victorian Subtidal Reef Monitoring Program (Edmunds et al., 2003;2005;2007).
Decapods were visually identified in situ by trained divers.The training used the following protocol: 1) each species was identified in situ during night dives; 2) the species was recorded on an underwater clipboard; 3) the specimen was hand-collected (with tweezers and dip nets); and 4) the species identification was confirmed through appropriate references.This procedure was repeated until the diver was confident in identifying each species.Some decapods were not collected because their usual and known color pattern and morphological characteristics allow easy identification in situ.These included some commercially important species used for food (Cervigón et al., 1992) or the aquarium trade (Calado et al., 2003;Gasparini et al., 2005).

Data analysis
The following estimated ecological indexes were based on Krebs (1994;1998) Dajoz (2005) ) and Odum and Barrett (2007).Abundance -total number of individuals and Frequency of occurrence (%) -percentage of occurrence of a given species in relation to the total number of transects or areas, calculated by Fa = (Pa x 100) / P, where: Fa = frequency of the species, Pa = number of transects in which the species is present, and P = total number of samples or stations.Based on the frequency value, the species were considered rare (Fa < 10%), common (10% ≤ Fa < 50%), or constant (50% ≤ Fa ≤ 100%).Dominance (%) -expresses the ratio between the number of individuals of a given species and the number of individuals of all species, calculated by the formula Da = (Na x 100) / Nt, where Da = Dominance of the given species, Na = number of individuals of the species, and Nt = number  of individuals of all species.Diversity (H') -Shannon-Wiener diversity index (Shannon and Weaver, 1949), considered slightly diverse (H' < 1 bits.ind-1 ), diverse (1 ≤ H' < 2 bits.ind -1 ), and very diverse (2 ≤ H' < 3 bits.ind-1 ). Equitability (J') -the Pielou equitability index (Pielou, 1966), ranging from 0 (lowest uniformity) to 1 (maximum uniformity).
To analyze the main groups of species according to area (Confined, Semi-open and Open), we performed a cluster analysis and generated a dendrogram with the species and the degree of similarity of each group, using the Bray-Curtis similarity coefficient (Bray and Curtis, 1957) based on the abundance of the most important species (excluding rare species with fewer than 10 individuals found).
To evaluate differences between the means for abundance, diversity and equitability in the study areas, the nonparametric Mann-Whitney test was used (Zar, 1999), with a significance level of 5%.To check the similarity in composition and abundance with respect to the above variables, a Non-Metric Multidimensional Scaling (NMDS) ordination analysis was performed (Clarke, 1993;Clarke and Warwick, 1994).
The similarity matrix was constructed using the Bray-Curtis similarity coefficient (Bray and Curtis, 1957) based on the abundance of the most important species (excluding rare species with fewer than five individuals found).The bifactorial Analysis of Similarity (ANOSIM) evaluated the existence of significant differences in the fauna composition according to area, through the NMDS analysis matrix.R values higher than 0.5 and with a significance level less than 5% were considered statistically significant.All calculations were performed using the statistical software package Primer® 6.0 (Clarke and Gorley, 2001).
In all, 29 species occurred in the Confined, 25 in the Semi-open, and 22 in the Open area.The species of infraorders Stenopodidea, Caridea, Achelata and Astacidea occurred in all three areas, but were more abundant, frequent, and dominant in the Semi-open and Open areas.Anomurans were best represented in the Confined area, with the exception of Cancellus ornatus Benedict, 1901, which occurred only in the Semi-open and Open areas.Brachyurans were also best represented in the Confined area (16 species); the number of species declined with the distance from shore, with 11 species in the Semi-open and 9 in the Open area.The caridean Cinetorhynchus rigens and the lobster Panulirus echinatus were the dominant and abundant species in the entire reef area (Table 1).

Multivariate analysis
Despite the high similarity (>70%) indicated by the cluster analysis, three distinct groups were apparent (Fig. 5).The A similar pattern was found in the MDS for the species distribution, with the formation of two statistically significant groups (Fig. 6 ).The southernmost point in the Confined Water (A4) was distinct; this point is deeper and less confined than the other points (Fig. 6).

Decapod composition
The brachyuran crabs are among the most abundant and diverse groups of species on coral reefs (Dubinsky and Stambler, 2011), and in this study comprised 55% of the total species on the reefs.The ornamental crab Mithraculus forceps was the most dominant and abundant crab, as also found on rocky shores of the southeastern and southern coast of Brazil by SCUBA divers (Gaeta et al., 2011) and manual sampling (Mantelatto et al., 2004;Bouzon and Freire, 2007).The dominance of M. forceps in shallow hard-bottom areas is probably due to: short larval stage, high survival rate of juveniles and rapid growth (Rhyne et al., 2005); high reproductive effort and growth stimulated by warm water (Mantelatto et al., 2004;Rhyne et al., 2005;Figueiredo et al., 2008); as well as herbivorous feeding habit (Figueiredo et al., 2008); and possibly effective defense strategies and an absence of large predators.
These species are commonly found associated with algae on coral reefs or rocky shores and coral reefs (Melo, 1996), and were also collected in the daytime by SCUBA divers in southeast and southern Brazil (Mantelatto et al., 2004;Bouzon and Freire, 2007).
Caridean shrimps are also an important group on coral-reef and other hard-bottom ecosystems, including marine and anchialine caves (Hobbs III, 1994;Clark et al., 2008).The caridean Cinetorhynchus rigens was the most abundant and dominant species on the Porto de Galinhas reef, comprising around 50% of all specimens found during the study.This shrimp is reported from cavernicolous marine environments (Okuno, 1997;Micael et al., 2006) and on the Porto de Galinhas reef was found in crevices in the walls, on cave roofs, and in reef grottos, moving out of the caves at night.Caillaux and Stotz (2003) found a similar pattern for the shrimp Rhynchocinetes typus H. Milne Edwards, 1937 in cavities and caves in Chile.Another very abundant caridean shrimp on the Porto de Galinhas reef was Janicea antiguenis, which is often found in underwater caves and tunnels (Manning and Hart, 1984;Hobbs III, 1994;Udekem d'Acoz, 2001;Wirtz, 2004).The high abundance of J. antiguensis and C.rigens in studied reef is probably due to: the studied area present many caves and a large crevicular system below the reef structures, typical habitat for these species (Manning and Hart, 1984;Okuno, 1997;Micael et al., 2006); J. antiguensis and C. rigens have nocturnal habits (Okuno, 1997).
Some species found on this reef deserve particular attention because of their economic and social importance.Brachycarpus biunguiculatus, Cinetorhynchus rigens, Stenopus hispidus, Enoplometopus antillensis, Palinurellus gundlachi, Calcinus tibicen, Domecia acanthophora, Stenorhynchus seticornis, Pelia rotunda, Platypodiella spectabilis, Dardanus venosus, Petrochirus diogenes and Mithraculus forceps are ornamental species popular in the aquarium trade (Calado et al., 2003;Gasparini et al., 2005;Balaji et al., 2009), while the lobsters Panulirus argus, P. laevicauda and P. echinatus are targets of artisanal fishing (Rocha et al., 1997).Many of these species were rare and in low abundance on the reefs, demonstrating the importance of developing local public policies for the management of their catches.
The portunid crab Charybdis hellerii was rare but present in the study area, especially in confined waters.An invasive species originally from the Indo-Pacific, C. hellerii is recorded from the Brazilian coast from Maranhão to Santa Catarina, in reefs and estuarine environments (Calado, 1996;Mantelatto and Dias, 1999;Bezerra and Almeida, 2005;Braga et al., 2005;Feres et al., 2007;Coelho et al., 2008;Loebmann et al., 2010;Silva and Barros, 2011).It competes for space and food with native species (Loebmann et al., 2010).

Spatial distribution
The abundance of decapods showed no statistically significant differences between the sampling areas.However, we expected that decapods found in the shallower and more easily accessed areas on a reef bench that is heavily used by tourists (MMA, SECTMA and CPRH, 2003) would be less abundant and less diverse, especially species with economic value.Surprisingly, these shallower areas showed the highest diversity and abundance, suggesting that the Confined-water areas provide shelter for juveniles and young decapods, as also recorded for the spiny lobsters Panulirus argus and Panulirus echinatus, which use the crevices of coastal reefs as refuges (Barreto and Katsuragawa, 2008;Igarashi, 2010).
Analysis of the faunal assemblage clearly showed the formation of three well-defined groups with different spatial distributions: 1) species found in shallow waters, with maximum depth about 2.5 m and less-intense hydrodynamics (Confined area); 2) species in deeper locations (3 to 10 m), which are subject to some form of wave and tidal stress (Semiopen and Open areas).These findings suggested that hydrodynamics (water confinement) and differences in depth on shallow coastal-reef benches influence the distribution of subtidal decapods, and agree with observations by Martínez-Iglesias and García-Raso (1999) and Huang et al. (2011), who included these among the main factors that influence the distribution of benthic invertebrates.The third group was formed by the crab Domecia acanthophora, the only species found only on the fire coral Millepora alcicornis Linnaeus, 1758, showing the importance of inter-specific associations existing on reef ecosystems (Patton, 1967;Criales, 1984;Wirtz and Udekem d'Acoz, 2008;Wirtz et al., 2009).
The assemblage in the Confined area was mainly composed of Brachyura and Anomura.The five most significant (dominant and frequent) species in this area, Calcinus tibicen, Pagurus provenzanoi, Mithraculus forceps, Microphrys bicornutus and Mithrax hemphilli are commonly associated with secondary substrates such as macrophytes, sponges and corals on reefs in northeastern Brazil, and can be collected manually during low tide (Melo, 1996;Almeida et al., 2008;Garcia et al., 2008;Batista et al., 2009).On these reefs, the Confined area, with its shallow, welllighted pools is favorable for the establishment of secondary substrates such as macrophytes, zooanthids and corals (Maida and Ferreira, 1997;Barradas et al., 2010) and decapod crustaceans are known to be associated with these living substrates (Stevens and Anderson, 2000;Batista et al., 2009;Wirtz et al., 2009).
In the Semi-open area, the decapod assemblage showed intermediate characteristics between the assemblage in the Confined area, composed mainly of brachyuran crabs, and in the Open area, composed mainly by caridean shrimps, suggesting a gradual change in decapod distribution according to depth and hydrodynamics.In the Semi-open area, Stenopus hispidus, Janicea antiguenis and Palinurellus gundlachi were prominent; these species are commonly recorded in caves, tunnels or inaccessible cavities on coastal reefs (Manning and Hart, 1984;Williams, 1984;Melo, 1999).The ornamental shrimp Stenopus hispidus is the most popular shrimp among aquarists (Calado et al., 2003), and probably finds refuge in the caves of the Semi-open area.
The Open-water area in the breaker zone is more exposed to wave action (with stronger hydrodynamics) and showed the lowest decapod diversity, significantly lower than the more-protected areas on the reef bench.The strong hydrodynamics limits the establishment of decapods and other benthic species such as seaweed, sponges and corals (Huang et al., 2011), which serve as living substrates for some decapod species.According to Thiel and Vásquez (2000), the branches and structures of algae are important microhabitats for decapods.
The characteristic species (highest dominance and frequency) for the Open area were caverniculous or crevicular species including Cinetorhynchus rigens, Brachycarpus biunguiculatus, Janicea antiguensis, Enoplometopus antillensis and Palinurellus gundlachi, with adaptations to cave life (winered color and/or long tactile structures), or species with strong structures to hold themselves in the reef structures under wave and current action, such as the ornamental crabs Stenorhynchus seticornis and Platypodiella spectabilis (Hart et al., 1985;Hobbs III, 1994;Iliffe and Bishop, 2007).
Stenorhynchus seticornis is commonly associated with sessile invertebrates (Hayes et al., 1998;2006;Wirtz et al., 2009), and at Porto de Galinhas, was found associated with the octocoral Carijoa riisei Duchassaing and Michelotti, 1860 in the Semi-open and Open areas.Stenhorhynchus seticornis has an elongated body, which provides perfect camouflage and symbiosis among the octocoral structures.The ornamental crab Platypodiella spectabilis was found on the zooanthid Palythoa caribaeorum Duchassaing and Michelotti, 1860 that commonly covers the flat reef surface in the breaker zone (Barradas et al., 2010), an association also reported by Gleibs et al. (1995) in the Colombian Caribbean Sea.Domecia acanthophora, which was found in this study only on the hydrocoral Millepora alcicornis Linnaeus, 1758, was also recorded on Millepora alcicornis at Parrachos de Maracajaú, northeastern Brazil (Garcia et al., 2008), and on three acroporid corals [Acropora cervicornis (Lamarck, 1816), A. palmata (Lamarck, 1816) and A. prolifera (Lamarck, 1816)] in Puerto Rico (Patton, 1967).This association with cnidarians provides the reef crabs protection against predators and the strong tidal currents typical of the Open and Semi-open areas (Williams, 1984).

Figure 1 .
Figure 1.Reef area studied on Porto de Galinhas Beach, state of Pernambuco, northeastern Brazil.

Figure 3 .
Figure 3. Abundance box plot of decapod species in Porto de Galinhas, Pernambuco, Brazil, showing the values for the three sampling areas: Confined Water (A), Semi-open Water (B) and Open Water (C).

Figure 4 .
Figure 4. Diversity and equitability box plot of decapod species in Porto de Galinhas, Pernambuco, Brazil, showing the values for the three sampling areas: Confined Water (A), Semi-open Water (B) and Open Water (C).

Figure 5 .
Figure 5. Dendrogram of similarities for the most numerous decapod species in Porto de Galinhas, Pernambuco, Brazil, based on analysis of the three sampling areas: Confined Water (A), Semi-open Water (B) and Open Water (C).

Table 1 -
Abundance, Dominance and Frequency of each species found in the subtidal of the Porto de Galinhas reef bench, Pernambuco, northeastern Brazil, by sampling areas: Confined Water (A), Semi-open Water (B), and Open Water (C).Relative abundance (Ab) with the mean number of individuals per transect.Dominance(Do)with the dominance percentage.Frequency of occurrence indicated by the color of the cells: Rare (white), Common (gray), and Constant (dark gray).*indicates absence of the species.