Distributions of two sea urchins, Paracentrotus lividus and Arbacia lixula, and their habitat associations in the northwestern Mediterranean Sea at STARESO, Corsica, France
T. McHugh and K. Elsmore
Department of Ecology and Evolutionary Biology, University of California, Santa Cruz 2012
ABSTRACT
Understanding the population structure of individual species within a community allows us to make predictions about the dynamics and the future health of an ecosystem. Studies on Paracentrotus lividus and Arbacia lixula conducted in the northwestern Mediterranean conflict in regards to diet, habitat occupation, and species interactions, leaving the questions of habitat associations unresolved. In this study, we provide an accurate and thorough assessment of Paracentrotus lividus and Arbacia lixula distributions and habitat associations for the fall season at STARESO, Corsica, France. We found statistically significant variations between the urchin species in density, spatial location, depth, substrate preference, and primary placeholder preference. Understanding the changes in the community surrounding the urchins is equally important, as it will show the effects of fluctuations in urchin abundance and distributions.
INTRODUCTION
A clear understanding ecological community structure is important when addressing interactions within an ecosystem. A community is bound together by the network of influences that species have on one another. Understanding the population structure of individual species within a community allows us to make predictions about the dynamics and the future health of an ecosystem. Habitats within the northwestern Mediterranean are especially vulnerable to degradationdue to tourism, global warming, and seafood collection (Chiantore et al., 2007). The niche theory generates a series of questions relevant to the distributions of a species, their habitat preferences, and how competition, predator presence, and resource partitioning influence their physical location within a system (Tilman et al., 2004). Comparing a set of species' fundamental niches to their realized niches andinvestigating the comparisons of available habitat to species distributions constitute potential reasons why a speciesflourishes and exceeds in a particular habitat (Tilman et al., 2004).
Ecologists have studied the cascading effects of sea urchins upon ecosystem dynamics in various ecological systems (Mills et al.,1993). The classic example of trophic cascades in kelp forests demonstrates the key role urchins play in a vegetative community. The removal of sea otters, a keystone predator, allows for urchins to overgraze the kelp, creating urchin barrens (Verlaque andNedelec, 1983, Verlaque 1987). These barrens are poor in biodiversity and ecosystem functions, which highlights the importance in understanding urchin densities and distributions (Privitera et al., 2011).
Two species of urchins, Paracentrotus lividus and Arbacia lixula, dominate the near shore habitat of the Northwestern Mediterranean. Paracentrotus lividus, the purple urchin, and A. lixula, the black urchin, coexist on hard substrate in shallow subtidal habitats (Bulleri et al., 1999). Studies have found that A. lixula is commonly found on vertical substrate and is considered a main grazer of coralline algae (Bulleri et al., 1999). Paracentrotus lividus is typically found in dense algal assemblages, rocky crevices, and under boulders (Tomas et al., 2004), grazing primarily on erect algae. Studies on P. lividus and A. lixula conducted in the Northwestern Mediterranean, however, conflict in regards to diet, habitat occupation, and species interactions, leaving the questions of habitat associations unresolved.
This study investigated the fundamental niche, potential habitat, and places where the two urchins can live, and compared it to the realized niche of the two urchin species. We examined habitat associations of P. lividus and A. lixula in the Northwestern Mediterranean in a nearshore habitat. Spatially accurate distributions and abundances of each species were collected to address the question of habitat associations and niche theory. Due to the variety in physical and biological habitat structure, STARESO harbor provided a good study system to address all habitat attributes discussed in the conflicting literature. In this study, we tested the hypotheses that two urchin species; Paracentrotus lividus and Arbacia lixula:
(1) occupy distinct spatial locations within STARESO harbor
(2) occupy distinct depth ranges
(3) have specific substrate associations
(4) have specific primary placeholder associations.
METHODS
Study System
This study was conducted in October 2012 at Station Research Sous-Marine et Oceanographique (STARESO), Point Revellata, Calvi, France. Due to the variety of substrate types and dense algal cover, STARESO harbor provided an ideal study system for understanding the spatial distributions of P. lividus and A. lixula. The diversity of primary placeholders and substrate, in STARESO harbor is extremely diverse in proportion to the small size of the study site. Substrates included sand, cobble, boulders, bedrock, large cement jacks, and cement vertical walls. The biotic system was composed of a wide collection/composition of algal and invertebrate species.
Sampling Design
Sampling was done entirely by SCUBA. A permanent transect line was set through the center of the harbor, with flagging tape set every 5 meters from 0m to 50 m. The line ran 150 degrees “east” from the permanent ladder in STARESO harbor. Eleven transects were run perpendicular to the main line. Total lengths varied per transect due to the shape of the harbor. Lengths ranged from 41m to 76m. To simplify sampling methods, the harbor was then separated into two distinctive regions; (+) 60 degrees and (-) 240 degrees from the main line (Fig. 1). A meter tape was reeled out along the bottom, along the substrate, until surfacing, or a physical barrier prevented the extension of the tape.
Habitat Data Collection
We characterized depth, substrate, and primary placeholders in STARESO harbor using uniform point contact (UPC) methods. For each of the eleven transects, we reeled out a meter tape flush along the bottom, following the contour until reaching the water line. The transect lengths were determined by the width of the harbor. We sampled every other meter along each transect, using 0.68m x 0.68m quadrats. Each quadrat contained 9 equidistant points, at which we recorded substrate and primary placeholders. Using an Aeris XR1 dive computer, we collected depth readings for every meter per transect. Depths were recorded and analyzed in feet to display a more drastic and tangible depiction of the distribution of urchins and bathymetry of the harbor. All other distance measurements will be referred to in meters. STARESO staff members identified five substrates and sixteen primary placeholders as the main biotic and abiotic features of the harbor worth assessing and quantifying for habitat characterization. Substrates and primary placeholder categories included bedrock, boulder, cobble, sand, cement, jacks, bushy, turf, film, encrusting algae, living Posidonia oceanica, dead Posidonia oceanica, colonial tunicates, solitary tunicates, bryozoans, octopus dens, polychaetes, black sponges, red sponges, hard coral, and anemones.
Urchin Data Collection
We used the swath method to measure urchin density and distribution along the eleven transects in STARESO harbor and the UPC method to characterize their substrate and habitat associations. Urchin data collection was restricted to the hours of 20:00:00 - 23:00:00 to account for the nocturnal feeding displayed by both urchin species (Guidetti, 2004). We swam along each transect, looking for both P. lividus and A. lixula. Surveys were restricted to the width of the quadrat (0.68 meters) and the west side of the meter tape. Upon finding an urchin, we placed the quadrat on the closest meter mark. Substrate and primary placeholder data was collected for each of the 8 point surrounding the urchin, in addition to the substrate and primary placeholder beneath the urchin. Species, depth, and coordinates were recorded for each urchin sampled.
Analysis
To determine if the spatial location of A. lixula and P. lividus differ, recorded depths were used to create a bathymetric map of the STARESO harbor in SYSTAT and SURFER programs. Paracentrotus lividus and A. lixula depths and exact locations were added to the habitat map to display their specific distributions within STARESO harbor. To determine if the depth ranges in which A. lixula and P. lividus inhabit differ, analysis of variance (ANOVA), followed by pairwise comparisons, were done to compare urchin distributions as a function of depth. An analysis of similarities (ANOSIM) was run to determine if substrate associations differ between the two urchin species. A similarity percentages (SIMPER) analysis was then used to determine the driving factors of substrate association differences. ANOSIM analyses were run to determine if primary placeholders associations differ between the two urchin species. SIMPER analyses were used to determine the driving factors of primary placeholders association differences.
RESULTS
According to our harbor map (Fig. 2, Fig. 4), P. lividus and A. lixula were found within distinct geographic locations. Paracentrotus lividus occupied a uniform region, gradually sloping with the contours of the seafloor. The abundance of P. lividus exceeded that of A. lixula within the STARESO harbor (Fig. 4). Arbacia lixula individuals were pressed along the shallower outer edges of the cove, towards the zero depth mark (Fig. 2).
An ANOVA analysis, followed by pairwise comparisons, indicates all comparisons of available depths and actual depths on which urchins were found to be highly significant (Table 1). Confirming our harbor map, A. lixula occupied an average depth of 7.2 feet (Table 2, Fig. 3). Alternatively P. lividus inhabited an average depth of 10.6 feet (Table 2, Fig. 3). Arbacia lixula populated a depth range higher in the water column and at a steeper depth gradient than that of P. lividus (Fig. 4).
The p-values yielded from the substrate PERMANOVA analysis indicate that there are significant differences between each of the pairwise comparisons (Table 3). Boulder and cobble are the major components responsible for the statistical difference between available substrate and the substrate P. lividus inhabits (Table 4a). Boulders and cement are the major components responsible for the statistical difference between available substrate and the substrate A. lixula inhabits (Table 4b). Boulders cement, and cobble are the major components responsible for the statistical difference between the substrate P. lividus inhabits and the substrate A. lixula inhabits (Table 4c).
The p-values yielded from the primary placeholder PERMANOVA analysis indicate that there are significant differences between each of the pairwise comparisons (Table 5). Bushy algae, turf algae and encrusting algae are the major components responsible for the statistical difference between available primary placeholders and those associated with P. lividus (Table 6a). Bushy algae, film, and encrusting algae are the major components responsible for the statistical difference between available primary placeholders and those associated with A. lixula (Table 6b). Bushy algae, film, and encrusting algae are the major components responsible for the statistical difference betweenprimary placeholders associated with P. lividus andA. lixula, respectively (Table 6c).
DISCUSSION
Our graphical depiction of the urchins in STARESO harbor indicates that both species of urchins occupy the perimeters of the north and south edges of the harbor. Urchins were absent from the harbor’s central sand and P. oceanica channel. This implies that the deeper, flat channels of the harbor are not ideal conditions for the urchins, contrary to what some literature states (Kempf, 1962; Regis, 1978). Our results provide support for the idea that morpho-functional features such as attachment tenacity, spine length, and test robustness of sea urchins could be involved in affecting their distributions (Guidetti & Mori, 2005). Support for this hypothesis, spatial locations of A. lixula and P. lividus differ, was the motivation for looking at possible mechanisms of habitat occupation.
One probable mechanism driving the difference in habitat occupation between the two species is depth. Arbacia lixula dominates the shallower regions of the harbor at an average depth of 7.2 feet, while P. lividus is found in deeper regions at an average depth of 10.6 feet. One possible mechanism could be due to wave action exposure. Organisms found at shallower depths are more exposed to wave action and stresses from water movement. Hydrodynamics appear to play a key role in affecting the distribution across depths in several species of sea urchins (Ebert, 1968; Littler, 1980; Monteiro Marques, 1982; Denny and Gaylord, 1996). Despite being subject to dislodgement by high magnitude waves, A. lixula is found most commonly on vertical walls at shallower depths (Kempf, 1962; Regis, 1978). Morphological studies of A. lixula have concluded that strength and arrangement of calcareous plate connections composing the test increase tolerance to physical stresses caused by wave action (Regis 1978; Chelazzi et al., 1997). Alternatively, P. lividus shows a distinct behavioral adaptation to wave action, by preferring deeper waters and occupying crevices and self-burrowed refuges (Chelazzi et al., 1997).
Deeper waters also provide refuge from seasonal, annual, and storm-related water condition changes (Fernandez et al., 2001). Urchins are generally considered to be stenohaline organisms, unable to tolerate fluctuations in salinity (Roller and Stickle, 1993). Bressan et al. (1995) suggests that the distribution of P. lividus may be due to the water quality tolerances of adults. Urchins found at shallower depths are more vulnerable to salinity decreases (Fernandes et al., 2001), suggesting that A. lixula may be more robust in tolerating salinity and temperature fluctuations.
Another possible mechanism driving the difference in habitat occupation is substrate. The substrate preference of an urchin influences its ability to hide from predators, forage, and maintain homeostasis. Morphological limitations may be another factor in substrate choice. Test robustness and attachment tenacity were significantly greater for A. lixula than those for P. lividus, and the difference decreased in relation to smaller sea urchin size. These morphological adaptations and limitations confine the two urchin species to exploit two very contrasting substrates within the STARESO harbor.
The abundance of P. lividus found on boulders is more than double what would be expected, given the available boulder habitat (Table 4a). Due to the morphological constraints and predator defense weakness of P. lividus, boulders provide large stable surface area on which to attach, as well as crevasses to hide from predators (Guidetti and Mori, 2005). Boulders maintain a surface area to provide sufficient lighting for algae to grow. Additionally, A. lixula were found in high quantities within the boulder substrate (Table 4b). The size and orientation of boulders provide the vertical landscape and refuge that A. lixula seek (Kempf, 1962; Regis, 1978).
Despite the abundance of cobble in the harbor, this substrate contained fewer urchins than expected (Table 4). Contrasting boulders, cobble provides less surface area to hang onto, which is a stress for P. lividus. As previously stated, P. lividus attachment tenacity is extremely weak, and individuals can be dislodged easily when force is applied. Cobble within the harbor was found in extremely shallow depths and can move around during rapid water flow (Guidetti and Mori, 2005). Additionally, the small surface area, and shallow depth does not sustain the algae to which P. lividus individuals have been known to graze on (Chiantore et al., 2008). Although A. lixula is morphologically more adept at clinging onto a substrate with more force, they were not found on cobble in high abundances. The minimum amount of food and horizontal orientation that cobble provides may be the limiting factor that restricts A. lixula from this habitat.
Previous studies within the Northwestern Mediterranean have associated P. lividus with patches of P. oceanica (Thomas et al., 2004). Contrary to these results, our studies did not record any P. lividus or A. lixula individuals in P. oceanica meadows, which were located within the sand substrate (Table 4a, 4b). Sand does not support the food types commonly consumed by either species in STARESO harbor, nor does it provide protection. In comparison to sand, boulders are a more appealing and viable substrate providing the protection and food P. lividus needs to maximize growth.
Historically, there is a strong association between A. lixula, and anthropogenically created cement walls (Sala et al., 1998). Our studies found that A. lixula was found on cement over three times as often as what we expected (Table 4b). The solid, vertical orientation of cement walls provides the perfect substrate on which A. lixula to cling onto. Their strong spines and their attachment tenacity allow the urchin to be exposed without risk of predation or removal. The algal assemblage on cement walls was diverse, consisting of erect, film, and encrusting algae, which are ideal food types for A. lixula (Privitera et al., 2008). When comparing P. lividus to A. lixula, there were zero P. lividus individuals found on cement substrates (Table 4c). The exposure to predators and their weak attachment strength may limit their ability to utilize this area. Each of the substrates harbors different types of food, refuge, and attachment surfaces. The characteristics of STARESO’s substrates limit P. lividus and A. lixula to inhabit these areas based on their morphological and physiological constraints.
Another possible mechanism driving the difference in habitat occupation is primary placeholders. When looking at P. lividus, it was found in association with bushy algae nearly twice as often as what we would have expected, based on amounts available in the habitat (Table 6a). The most plausible reason is that bushy algae is their preferred food source. Studies on diet have confirmed that P. lividus primarily consumes erect algae such as Dictyotales and Padina (Chiantore et al., 2008, Privitera et al., 2011), which are algae that collectively compose our bushy algae category. Another possible reason for this is because bushy algae is found most commonly on their preferred substrate, boulders, which provides shelter essential for survival, as discussed above. Turf algae was found associated with P. lividus far less that what we had expected (Table 6a). The only factor distinguishing bushy and turf algae is their height from the substrate, so we predicted that P. lividus would be consuming both bushy and turf algae. Perhaps due to the abundance of erect algae available in STARESO harbor, P. lividus was able to demonstrate a preference toward the larger, bushy algae. Paracentrotus lividus was found in association with encrusting algae almost twice as often as we expected (Table 6a). While encrusting algae could be a food source for P. lividus, literature more strongly suggests that encrusting algae provides a better attachment surface for refuge than other algae (Guidetti and Mori, 2005; Barrios et al., 2010). Most of the encrusting algae was found in the crevices (where P. lividus resides for protection from predators) as bushy and turf algae dominated the completely exposed surface areas (personal observation).