Introduction

Coral reefs are considered among the most bio-diverse marine ecosystems in the world (Reaka-Kudla 1997; Glynn and Enochs 2011), which is principally due to the high structural heterogeneity of their habitats. In the eastern tropical Pacific (ETP), the coral reef ecosystems are distributed from the Gulf of California to Ecuador, with Pocillopora and Porites being the main reef builders, followed by other species, such as Psammocora stellata, Pavona gigantea, Leptoseris papyracea, and Pavona clavus. These species build small fringing reefs, reef patches, and develop isolated coral colonies (Glynn and Ault 2000; Reyes-Bonilla et al. 2005). In western Mexico, on the coasts of Jalisco, Colima, and Michoacán, there are coral assemblages dominated mainly by the genera Pocillopora (Reyes-Bonilla and López-Pérez 1998; Reyes-Bonilla et al. 2013). Unfortunately, different human activities such as fishing, tourism, port industry, and sewage discharge, among others, lead to the fragmentation of these habitats with negative effects on recruitment, protection, and local migration of several benthic and pelagic species (Fahrig 2003; Gosset and Rives 2006; Mora et al. 2007).

Within coral ecosystems, the echinoderms constitute a relevant group because of their important role for maintaining ecosystem function and “stability”. Some echinoderms are considered as a keystone species in different marine ecosystems, because they control the abundance and distribution of other benthic species, such as fleshy macroalgae, molluscs, and corals (Reyes-Bonilla and Calderon-Aguilera 1999; Tuya et al. 2004), thus impacting the whole community structure (Paine 1969; Tuya et al. 2004; Gaymer and Himmelman 2008; Ortiz et al. 2013a). Likewise, recent studies based on network analysis show that the echinoderms conform keystone species complexes in different benthic marine ecosystems, controlling the dynamics of several species (Ortiz et al. 2013a, b). In coral reefs, sea urchin and sea star species drive the processes of bioerosion, coral recruitment, and transfer of energy in the ecosystem (Glynn et al. 1979; Carreiro-Silva and McClanahan 2001). Furthermore, several studies indicate that the spatial and seasonal variation of the echinoderms is related mostly to processes such as settlement, recruitment, food availability, depredation, competition, physical factors, and fishing (Hagen and Mann 1992; Tyler et al. 2000; Alves et al. 2001; Hasan 2005; Clemente et al. 2009), as well as to the structural properties of habitats (Entrambasaguas et al. 2008; Alvarado et al. 2012). Such processes operate at different spatial and temporal scales (Legendre and Legendre 1998; Chapman and Underwood 2008).

Although in the tropical Mexican Pacific coast, several studies have been conducted on the ecology and taxonomy of echinoderms that inhabit coral ecosystems (e.g., López-Uriarte et al. 2009; Granja-Fernández and López-Pérez 2012; Alvarado and Solís-Marín 2013; Ríos-Jara et al. 2013; Rodríguez-Troncoso et al. 2013; Hermosillo-Nuñez et al. 2015), few attempts have focused on the assessment of how human interventions change the influence of reef coral on echinoderms. Along the coast of Jalisco, Colima, and Michoacán, there are important coral assemblages with high richness and cover (Reyes-Bonilla et al. 2013), which could be impacted by the proximity to human settlements. Coral reefs have been subjected to a wide spectrum of human disturbances, such as eutrophication, pollution, and land-use changes that have deteriorated the coral ecosystems in terms of coral growth, reproduction and competition (e.g., Wilkinson 1999; Nyström et al. 2000; Koop et al. 2001; Szmant 2002; González-Lozano et al. 2006). In this case, the principal impacts on the marine ecosystems on the coast of Jalisco, Colima and Michoacán correspond to tourism, recreation, industrial activities, harbors, urban development, habitat destruction, and wastewater discharges (Ortiz-Lozano et al. 2005; Nava and Ramírez-Herrera 2012).

It has been widely accepted that habitat heterogeneity and habitat fragmentation promote opposite impacts on the structure, richness, spatial/temporal distribution, density, and coexistence of species (Wilcox and Murphy 1985; Frost et al. 1999; Attrill et al. 2000; Hovel and Lipcius 2001; Kelaher 2003). For example, Rodríguez-Zaragoza et al. (2011) showed that coral communities contribute to the heterogeneity of benthic habitats, promoting the diversity of reef fishes. Likewise, Hermosillo-Nuñez et al. (2015) concluded that the spatial distribution and abundance of echinoderms can be explained by habitat structure. Jiménez (2001) showed that reef areas with recreational activity and commercial diving reflect an equal level of damage to the corals. González-Lozano et al. (2006) determined that water quality in port areas exceeds the permitted limits of anthropogenic pollution, affecting the biota. It is therefore expected that the spatial distribution of echinoderms can be determined by habitat heterogeneity. The aim of this work was to assess how coral assemblages influence the spatial variation of echinoderms in coastal areas under different regimes of human interventions along the coast of Jalisco, Colima, and Michoacán (western Mexico). The information obtained in this study could be used when developing management strategies focused on coastal conservation.

Materials and methods

Study area

The Mexican Pacific coast between the 20° and 17°N includes the states of Jalisco, Colima, and Michoacán (Fig. 1). This area is characterized by an irregular coastline comprised principally of rocky, coral reefs, closed bays, sand beaches, and estuaries. Likewise, the coastal benthic systems are influenced by the California Current (between February and April), the North Equatorial Countercurrent (between August and January) and the convergence of both (between May and June) (Godínez-Domínguez and González-Sansón 1998).

Fig. 1
figure 1

Study area and sampling sites at Jalisco, Colima, and Michoacán in the western México

Full size image

Along this coastal area, it is possible to recognize sites with different levels of disturbances related to the distance to human settlements. The sites of Isla Cocina, Isla Pajarera, Isla Mamut, La Pajarera, Isla Pelicano, and Ensenada Cocinas are located in the state of Jalisco within the Santuario Islas e Islotes de Bahía Chamela, decreed as a natural protected area (NPA) for protection of biodiversity and richness (DOF 2002). This NPA is relevant because it protects unique environments relatively isolated from human activities. The Santuario Islas e Islotes de Bahía Chamela is exposed to lower levels of human intervention, since only recreational, research and education activities are permitted (CONANP 2008). The sites of La Virgencita and Las Monjas 2, located in Jalisco, Carrizales Norte, Carrizales Sur, Paraiso Montemar, L’Recif, El Elefante in Colima, and Barco Hundido, Isla Protegida, Isla Pajaros O, Isla Pajaros N, El Faro, El Relis, La Ensenada and El Teolán in Michoacán are coastal areas under medium disturbance because they are located at a distance from human settlements. Finally, Cuastecomatito, Las Monjas1, Punto B, La Boquita, El Muelle, and Manto de Carrizo are considered sites under higher disturbance because they are located close to human settlements.

Sampling procedure

The fieldwork was conducted during February and September of 2010. A total of 27 sites with coral assemblages (ten in Jalisco, seven in Colima and ten Michoacán) were chosen for sampling (Fig. 1). Each study site presented different geomorphological features, habitat heterogeneity, depth, and wave exposure. The coral assemblages were dominated mainly by the genera Pocillopora, Porites, and Pavona, which are distributed between a depth of 2 and 15 m. The abundance of echinoderms, corals, and other benthic groups was assessed by SCUBA diving at three transects per site, and the depth was also recorded. The transects were located parallel to the coast and different sampling strategies were applied per transect: (1) diurnal visual censuses of 20 × 5 m2 for sea stars, 20 × 2 m2 for sea cucumbers and 20 × 1 m2 for sea urchins; (2) video transects of 20 × 0.6 m2 were recorded (0.4 m above the bottom) for other benthic organisms. From each video, 40 frames were subsampled; 50 fixed points from each frame were used to estimate the hard-coral cover (Pocillopora, Porites, Pavona and Psammocora), as well as the cover of hydrocorals, bryozoans, octocorals, sponges, fleshy macroalgae, encrusting calcareous algae, articulated calcareous algae, turf, sessile organisms, mobile organisms, rocks, sand, and coral rubble (Rodríguez-Zaragoza et al. 2011; Hermosillo-Nuñez et al. 2015). Organisms were identified in situ to the taxonomic level of species, based on Brusca (1980), Hickman (1998) and using the photographic guides of Gotshall (1998), Kerstitch and Bertsch (2007). All census sampling was carried out by the same person.

Statistical analysis

The sampling effort for the study area was evaluated through the accumulation curves of the exponential function of the Shannon diversity index (eH′) and the reciprocal of the Simpson dominance (index) (1/D). All curves were built with 10,000 random combinations without replacement using the software Estimates V9.1 (Colwell 2009). The species richness (S) was defined as the total number of species recorded for each site. The abundance (N) was analyzed based on the density represented as the total number of individuals of each species per square meter (ind/m2). The heterogeneity of the species was evaluated with the Shannon diversity index (H′, nats) and the Simpson dominance index (D), which were estimated per transect level for comparison among sampling sites using their mean values (Magurran 2004; McCoy and Bell 1991). This was done with Primer V6.1 software (Clarke and Warwick 1994).

The analyses of variances (ANOVA) were performed using permutations to evaluate the changes of the mean values from the species richness, abundance, Shannon diversity and Simpson dominance. These unrestricted analyses were performed because the data did not meet with parametric statistical assumptions. The ANOVAs were constructed using Euclidean distance matrices following the criteria of Anderson et al. (2008). Unrestricted ANOVAs generate a variance partitioning through of the estimation of a statistical Pseudo-F following a design similar to a general linear model (Anderson et al. 2008). The experimental design analysis was based on three nested factors (state, area, and site with 3, 6, and 27 levels, respectively), using a type II model with random factors. The spatial variation of the composition and abundance of the species of echinoderms was evaluated through a permutational multidimensional analysis of variance (PERMANOVA), which was based on the same ANOVA design. These data were pretreated with Log (X+1) to reduce the contribution of the most abundant species and increase the contribution of those with very low abundance. Subsequently, a matrix of Bray-Curtis similarity was constructed. All ANOVA and PERMANOVA tests were conducted using the software PRIMER V6.1+PERMANOVA (Clarke and Gorley 2006; Anderson et al. 2008). Several similarity percentage analyses (SIMPER) were performed to compare the contributions of the species within and among sampling sites. SIMPER is based on the Bray-Curtis index for estimating the average dissimilarity between pairs of sample groups and determining the contributions to the average similarity within each group (Clarke and Warwick 2001).

A canonical additive partitioning of the spatial variation of the echinoderms was done between environmental and spatial predictive variables based on canonical redundancy analyses (RDA) (Legendre and Legendre 1998). The spatial variation of the echinoderm assemblage was represented by Y vectors (species richness, Shannon diversity index, Simpson dominance index and total species abundance) or the Y matrix (echinoderm species composition and abundance). Environmental variables (Env. Var.) were included in the X matrices using the mean values of the coverage of the benthic organisms and the substrate types estimated from the videotransects. Additionally, the depth was considered in this matrix. The W matrices correspond to spatial variables (Spat. Var.), where x, y, and y 2 were derived from third-order polynomial terms of the geographic coordinates represented as distances in kilometers (Legendre and Legendre 1998). The Trace statistic indicated the variation of Y explained by X and Y. Stepwise forward selection was conducted to identify the environmental variables that best explained the variation in Y. The RDA ordinations were developed using the software CANOCO v4.5 (ter Braak and Šmilauer 2002), assuming a linear relationship between the biological and the spatial-environmental components. Multicollinearity was evaluated among the environmental variables because it could modify the RDA model outputs. All RDA analyses were built using the variance inflation factor (VIF) lower than ten to avoid severe multicollinearity (Chatterjee et al. 2000; Graham 2003).

Results

A total of 3754 individuals belonging to 14 echinoderm species (two sea stars, seven sea urchins, and five sea cucumbers) were registered. The accumulation curves of the Shannon diversity index and Simpson dominance index showed an asymptotic behavior (Online Resource 1), indicating that the abundant and very abundant species, which are those that contribute most to the assemblage structure, were well represented. The most abundant species were Diadema mexicanum and Phataria unifascialis, with 2520 and 477 individuals, respectively.

There was significant spatial variation in the echinoderm assemblage (i.e., community attributes, and composition and abundance) associated with the coral ecosystems (Table 1). The outcomes of the three-way fully nested ANOVA showed that the average values of the species richness, echinoderms abundance, and Simpson dominance index had a significant spatial variation with the area and site levels (Table 1). The sites with the highest average richness were Isla Pájaros Norte, Isla Protegida, La Boquita, and Punto B (Fig. 2). Meanwhile, the sites Isla Cocinas, Ensenada Cocinas, La Boquita, Las Monjas 1, and El Faro showed the highest magnitudes of dominant species (Fig. 2). On the other hand, the Shannon diversity index showed a significant variation at the site scale, where Paraíso Montemar, Las Monjas 2, L’Recif, Isla Pájaros Norte, El Relis, La Ensenada, Carrizales Sur, and Isla Pelicano had the greatest magnitudes of diversity (Fig. 2). Similarly, the three-way fully nested PERMANOVA showed that the echinoderm composition and abundance varied significantly only at the site level (Table 1).

Table 1 ANOVA and PERMANOVA outputs showing the spatial variation analyses of echinoderm assemblage
Full size table
Fig. 2
figure 2

Species richness (a), abundance (b), Shannon diversity (c), and Simpson dominance (d) by sampling sites at Jalisco, Colima, and Michoacán. Sites: Jalisco, La Virgencita (LVI), Cuastecomatito (CUA), Las Monjas 1 (LMO1), Las Monjas 2 (LMO2), Isla Pajarera (IPA), Isla Pelicano (IPE), Isla Cocina (ICO), Isla Mamut (IMA), Ensenada Cocinas (ECO), La Palma (LPA). Colima, Elefante (EEL), Carrizales Norte (CAN), Carrizales Sur (CAS), La Boquita (LBO), Punto B (PUB), Paraíso Montemar (PMO), L’ Recif (L´RE). Michoacán, El Relis (ERE), Manto de Carrizo (MCA), El Muelle (EMU), El Teolan (ETE), La Ensenada (LEN), El Faro (EFA), Isla Pájaros Norte (IPN), Isla Pájaros Oeste (IPO), Barco Hundido (BHU), Isla Protegida (IPR)

Full size image

The SIMPER results showed that D. mexicanum, Eucidaris thouarsii, P. unifascialis, Centrostephanus coronatus, Toxopneustes roseus, Holothuria fuscocinerea, Cucumaria flamma, and Echinometra vanbrunti contributed principally to the dissimilarities among the sites (Online Resource 2). In general terms, these outcomes showed that the site explains the major spatial variation of the echinoderm assemblage in the eastern tropical Pacific.

The canonical additive partitioning showed that the accounted variation [a + b + c] in these models was from 52.8 to 84.1 %, where the Shannon diversity index and Simpson dominance index had the highest values, while the species richness showed the least explanatory power (Table 2). The variations of the abundance, Shannon diversity, Simpson dominance, and the composition and abundance of the echinoderm species were explained mostly by pure environmental component—fraction [a]—with values between 42.1 and 48.7 %. In turn, the spatially structured environmental component—fraction [b]—explained from 17.6 to 33.6 % of the variation of the community attributes and echinoderm assemblage. On the other hand, variation in the species richness was mainly explained (24.6 %) by the spatially structured environmental component [b] and (20.7 %) the pure environmental component [a]. However, spatial variables—fraction [c]—accounted for less than 8 % of the total variation analyzed in all partitions, while the unexplained variation—fraction [d]—was between 15.9 and 47.2 % (Table 2). These results suggest that the spatial variation of the echinoderm assemblage related to the coral ecosystems was mainly explained by the environmental variables, that is, by the reef habitat structure. Nevertheless, the echinoderm species richness depends also on a set of environmental variables, but these covary with the spatial configuration of the sampling sites along the western coast of Mexico.

Table 2 Canonical additive partition of the echinoderm assemblage’s spatial variation at Western Mexican Pacific coral communities
Full size table

The environmental variables that contributed to the species richness of the echinoderms were the Porites cover, coral rubble, turf, and articulated calcareous algae, while the spatial variables’ relationship was x and y 2, representative of the longitudinal and latitudinal variation, respectively (Table 2). Likewise, the environmental variables that explained the greatest variation of the abundance, Shannon diversity, and Simpson dominance were the coral cover of Porites, rocky cover, encrusting calcareous algae, fleshy macroalgae, Balanus, Pavona coral, and sponges. On the other hand, the environmental variables that explained the greatest variation of the composition and abundance of the echinoderm species were the rocky cover, Balanus, soft coral, Pavona coral, fleshy macroalgae, coral rubble, Porites coral, dead coral, encrusting calcareous algae (ECA), Pocillopora coral and sponges, as well as the depth (Table 2). The RDA ordination showed that E. thouarsii and P. unifascialis had a positive relationship with articulated calcareous algae, and deep and soft coral; however, P. pyramidatus and Euapta godeffroyi presented a positive relationship with Porites cover. The echinoids D. mexicanum and T. roseus were related with rocky and Balanus cover, while A. pulvinata, T. depresus and Isostichopus fuscus presented a positive relationship with coral rubble and dead coral. Meanwhile, C. flama was correlated with the coverage of Pavona, Pocillopora, coral rubble and dead coral. Finally, H. fuscocinerea, E. vanbrunti, C. coronatus, and H. arenicola showed a positive relationship with the sponges and fleshy macroalgae. Likewise, it is important to mention that the coverage of Porites, Pavona, and Pocillopora corals, soft coral, rock, and Balanus shows a positive relationship with the sampling sites included within the natural protected area (black circles) and with the sites with medium human disturbances (white circles). In contrast, fleshy macroalgae, sponges, and soft coral show a positive relationship with higher disturbance sites (gray circles) (Fig. 3).

Fig. 3
figure 3

RDA triplot of spatial variation of echinoderm species at Jalisco, Colima, and Michoacán, México. Ordination of echinoderm species, habitat variables, and sites. Triangle represents the echinoderm species, thick line arrows show habitat variables, and circles represent the sites. Black circles represent the sites included within the natural protected area with low human disturbance, white circles represent the sites with middle human disturbances and gray circles represent the sites with higher human disturbance. Species codes: D. mex is Diadema mexicanum, T. ros is Toxopneustes roseus, A. pul is Astropyga pulvinata, I. fus is Isostichopus fuscus, C. fla is Cucumaria flamma, T. dep is Tripneustes depresus, C. cor is Centrostephanus coronatus, E. van is Echinometra vanbrunti, H. fus is Holothuria fuscoscinerea, H. are is Holothuria arenicola, E. tho is Eucidaris thouarsii, P. uni is Phataria unifascialis, E. god is Euapta godeffroyi, P. pyr is Pharia pyramidatus, Aca is articulate calcareous algae. Site codes are shown in Fig. 2

Full size image

Discussion

Fourteen echinoderm species were found associated with the coral, which is fewer than the number of echinoderms recorded in previous studies of comparable areas (Rodríguez-Troncoso et al. 2013; Ríos-Jara et al. 2013). The lower species richness obtained in this work could be explained as a consequence of different sampling techniques and effort, as well as different aims. Despite the lower species richness, our work shows that D. mexicanum, P. unifascialis, T. roseus, E. thouarsii, and C. coronatus were conspicuous species, which are also representative species of comparable ecosystems in other places of the eastern tropical Pacific (Alvarado and Fernández 2005; Holguín-Quiñones et al. 2008; Alvarado et al. 2012; Hermosillo-Nuñez et al. 2015).

Our results suggest that the structural elements of the habitat play an important role in the spatial variation of the echinoderms inhabiting the coral ecosystems. This latter could correspond to the widely accepted fact that corals are niche constructers (sensu Lewontin 1978) or bio-engineering species that increase the heterogeneity of benthic habitats and in turn provide shelter, foraging, and reproduction zones (sensu Jones et al. 1994). This is corroborated by Hermosillo-Nuñez et al. (2015) who showed that different coral species mainly determine the environmental condition of the habitat for the dominant echinoderm species. The coverage of the different types of algae associations (i.e., turf, encrusting calcareous algae, articulated calcareous algae, fleshy macroalgae), sponges, and bryozoans showed an influence over the spatial variation of the echinoids. This could be explained by the omnivorous diet and grazing behavior of the sea urchins, such as D. mexicanum, T. depresus, and E. vanbrunti, which feed on various species of algae, sponges, and benthic invertebrates (Tuya et al. 2004; Sonnenholzner et al. 2013). Similarly, Vance (1979) described the sea urchin C. coronatus as omnivorous, feeding on fleshy macroalgae, tunicates, and sponges. In addition, E. thouarsii feeds on coral polyps, coralline algae, and even juvenile molluscs (Glynn and Wellington 1983). It has also been reported that bryozoans are massively depredated by sea urchins and sea stars (Lidgard 2008; Figuerola et al. 2013).

Other environmental variables, such as rocky cover, coral rubble, and dead coral, also contributed to the spatial variation of the echinoderms species. Many of the species studied in this work were sampled in such substrata. For example, most of the species of sea cucumber detected (I. fuscus, E. godeffroyi, H. arenicola, and H. fuscocinerea) were recorded over rocky-sandy bottoms, perhaps because of their detritivore feeding behavior (Roberts et al. 2003; Granja-Fernández et al. 2013). Similarly, in this study, C. flamma was observed in the cracks of rocks and coral, which would be explained by this species’ feeding behavior of trapping particulate organic matter with its tentacles. Nevertheless, more studies are required for determining the food spectrum of the sea cucumbers because it is not precisely known (Calva 2003). Similarly, the sea urchin T. roseus was dominant on the rock and rubble coral, which could be explained by the cover of macroalgae, shells and other rubble, serving as camouflage against predators (Richner and Milinski 2000) and protection against excessive solar radiation (Sigg et al. 2007) and strong bottom currents (Dumont et al. 2007; Amato et al. 2008). Additionally, A. pulvinata was observed on the sand and rock coral, which could be related to its ingestion of deposit material from the bottom (as foraminifera and small gastropods) (De Ridder and Lawrence 1982).

The sea stars P. pyramidatus and P. unifascialis are described as herbivorous species (Luna-Salguero and Reyes-Bonilla 2010), and they are principally recorded on the rocks and the coral. This could be a consequence of the cavities in the coral providing refuge against depredation, and food is available on the rocks due to the algal beds that grow on them. On the other hand, the outcomes show that the soft corals and Tubastrea coccinea (hard coral) influence the spatial variation of the echinoderm assemblages, which agrees with other studies that indicate that hydrocorals and soft corals confer suitable habitats for several species of ophiuroids (Castro et al. 2006; Granja-Fernández and López-Pérez 2011). Similarly, it remains unclear how the coral T. coccinea contributes to other species. Despite the insufficient information about the soft-corals, they are a part of the benthic habitat heterogeneity, facilitating the echinoderm spatial distribution.

Water depth and Balanus spp. were also determined as important variables in the present study. It has been reported that the differences in depth change the abundance and size of mobile intertidal and subtidal invertebrates, which is attributed to biotic and abiotic components (Larsson 1968; Gaymer et al. 2001; Siddon and Witman 2003). Several authors have concluded that the highest abundance and species richness of echinoderms occurs mainly at the first 18 m due to the availability of food (Hooker et al. 2005; Tuya and Duarte 2012). On the other hand, barnacles may be part of the diet of sea urchins, as echinoids are known for having a wide-range feeding spectrum. Although to date the contribution of Balanus spp. to the spatial variation of the echinoderms studied is unclear, D. mexicanum presented the highest average density in Chamela coinciding with Balanus spp, which was recorded exclusively in this site.

Edgar et al. (2011) showed that protected marine areas with high and low levels of protection present higher coral cover and total densities of mobile macroinvertebrates, respectively, compared to exploited zones. In our study, the sites belonging to the natural protected area and those with medium human perturbations showed a positive relationship with the cover of soft and hard corals. Similarly, thirteen species of echinoderms showed a positive relationship with these sites, and these sites simultaneously presented the highest cover of hard coral and density of echinoderms, which would be explained by the reduced human interventions. The highly perturbed sites were dominated by fleshy macroalgae and sponges while the holothuria C. flamma presented a relationship with one these sites. These places are close to human settlements, harbors, and tourist development and are thus frequently impacted by fisheries, tourism, agriculture, and sewage waters, producing a decline in the coral assemblage and associated organisms (Chavez-Comparan and Macias-Zamora 2006; Nava and Ramírez-Herrera 2012). These negative human impacts could be amplified since the sediment in suspension and temperature stresses facilitate the growth of sponges and macroalgae, which constrain the growth of coral reefs (Rützler 2002; Fabricius et al. 2005). However, further studies are required in order to assess the health conditions of the coral assemblages and associated fauna that inhabit the protected and unprotected marine areas along the coast of Mexico.

Conclusions

The outcomes obtained in this work help us to gain a better understanding of the dynamics of the echinoderm assemblage in the coral assemblages located in the coastal ecosystems of Jalisco, Colima, and Michoacán in western Mexico, especially when the echinoderms perform an important functional role in the rocky and coral reefs. This study showed that the habitat structural elements of the coral ecosystems are the most relevant environmental variables for explaining the spatial variation of the echinoderm assemblage. Likewise, the highest cover of hard corals is positively related to highest abundance of echinoderms, especially in locations with low and medium levels of human perturbations. The findings highlight the importance of preserving the structural heterogeneity of the coral ecosystem habitats and the need to protect and create marine-protected area in order to improve biodiversity conservation. In this regard, the information obtained by the current study can contribute to the design of conservation strategies for this kind of marine ecosystem because some study sites show well-developed reefs with high values of richness and abundance compared to other areas of the Mexican Pacific (Reyes-Bonilla et al. 2013). While our results should be considered as an improvement in the knowledge of coral ecosystems, it remains necessary to conduct studies with an ecosystem ecology perspective, constructing trophic networks, and analyzing the macroscopic or emergent properties.