• regime shift; phase shift; canopy forming algae; C

Ecological systems can coexist, under similar environmental conditions and for an indefinite period of time, in alternative states characterized by different assemblage structure and population dominance. Many studies document the existence of transitions between alternative states (termed “regime shift” or “phase shift”) in both terrestrial and marine environments, in which small variations in the characteristics of the system generate drastic changes in the structure of populations, showing non-linear and hysteretic responses of the systems.
In temperate rocky subtidal environments, the benthic habitat may occur as assemblages dominated by erected macroalgae or alternatively as assemblages dominated by encrusting coralline algae, known as barrens. Barren habitats are characterized by a reduced complexity, biodiversity and productivity compared to adjacent vegetated areas, and are subjected to intense grazing by sea urchins: in the Mediterranean Sea, the two most common species of sea urchins are Paracentrotus lividus and Arbacia lixula.
The mechanisms promoting the transition from macroalgal forests to barrens are not entirely known: the present study aims to investigate experimentally which factors are involved in the formation of barren areas, in rocky subtidal environments. In particular, we tested the hypothesis that the presence of shelters or portions of substrate devoid of erect macroalgae can favor the persistence of sea urchins within vegetated areas, and then initiate the formation of barren grounds. In addition, the study tested the importance of identity and total density of the sea urchins assemblage in this process. The study was carried out along a rocky reef located on the southwestern side of the island of Capraia, which is part of the Tuscany Archipelago National Park. Between the end of June and the beginning of July 2013, 63 areas of 1 x 1 m characterized by an high cover of Cystoseira spp. were selected, in a depth range between 4 and 9 meters. The two species of sea urchin, P. lividus and A. lixula, were included, individually or in combination, at different levels of density, in patches with different types of substrate: vegetated areas (control), bare rock (simulating a natural mechanical disturbance) and areas in which an artificial shelter was set. Three areas were randomly assigned to each experimental condition.
Data were collected by means of visual sampling of the percent cover of algal assemblages, together with the abundance and species composition of sea urchins present in each area, in two successive dates, fifteen days and three months after the start of the experiment. The data obtained were analyzed using analysis of variance (ANOVA), according to an asymmetric design (BACI).
Results showed that the presence of shelters has a positive effect on the persistence and grazing activity of sea urchins, and that this is a sufficient condition promoting the formation of barren grounds. The presence of patches of substrate devoid of erect macroalgae within forested areas, however, does not seem to exert a positive influence on the formation of barren areas. These effects act independently of the identity and density of initial urchins assemblages, suggesting similar functional roles for the two species of sea urchin studied.
Given the increasing occurrence of barren areas around the Mediterranean Sea, these results can provide useful guidance in a perspective of management and conservation of shallow rocky reefs.