Date Published: April 26, 2017
Publisher: Public Library of Science
Author(s): Jesús Ernesto Arias-González, Tak Fung, Robert M. Seymour, Joaquín Rodrigo Garza-Pérez, Gilberto Acosta-González, Yves-Marie Bozec, Craig R. Johnson, Giacomo Bernardi.
Coral-algal phase shifts in which coral cover declines to low levels and is replaced by algae have often been documented on coral reefs worldwide. This has motivated coral reef management responses that include restriction and regulation of fishing, e.g. herbivorous fish species. However, there is evidence that eutrophication and sedimentation can be at least as important as a reduction in herbivory in causing phase shifts. These threats arise from coastal development leading to increased nutrient and sediment loads, which stimulate algal growth and negatively impact corals respectively. Here, we first present results of a dynamic process-based model demonstrating that in addition to overharvesting of herbivorous fish, bottom-up processes have the potential to precipitate coral-algal phase shifts on Mesoamerican reefs. We then provide an empirical example that exemplifies this on coral reefs off Mahahual in Mexico, where a shift from coral to algal dominance occurred over 14 years, during which there was little change in herbivore biomass but considerable development of tourist infrastructure. Our results indicate that coastal development can compromise the resilience of coral reefs and that watershed and coastal zone management together with the maintenance of functional levels of fish herbivory are critical for the persistence of coral reefs in Mesoamerica.
Coral-algal phase shifts in which coral cover declines to low levels and is replaced by algae [1,2,3] challenge the management of coral reefs worldwide [4,5], including in Mesoamerica . Phase shifts may be caused by many factors, encompassing both episodic pressures of short duration and chronic pressures of long duration . Following the Reefs At Risk assessment , among the most important local and global threats identified are overfishing, pollution, coastal development and climate change. Climate change combined with local stressors was identified as threatening 75% of the world’s coral reefs, and overfishing was identified as the most prevalent local threat, affecting around 55% of the world’s reefs . In the scientific literature on coral-algal phase shifts, overfishing is commonly cited as a key underlying driver (e.g., [1,3,8]). This has motivated coral reef management responses that include restriction or prohibition of fishing effort in designated areas  and efforts to limit consumer demand for ecologically important species (e.g., [10,11]).
This study does not require an ethics statement as we did not manipulate any animal or plant.
The parameterization of our Mesoamerican benthic model resulted in the same parameter ranges as in Fung and others (2011) , except that the upper limit of the exogenous coral recruitment rate (arising from recruits produced by spawning corals on non-modeled reef areas), lCs, is now 0.0002 yr-1 instead of 0.01 yr-1; the upper limit of the endogenous coral recruitment rate, lCb, is now 0.05 yr-1 instead of 0.5 yr-1; and the lower limit of the parameter measuring the negative effects of macroalgae on coral growth, βM, is now 0.4 instead of 0.2. Random sampling of the parameter space resulted in a set of 10,000 pristine model reefs with an average equilibrium coral cover of 45% (Fig 3A). This is near the upper end of the observed range of 20–40% coral cover considered to be representative of healthy reefs in Mesoamerica . When we applied the stressors of fishing, nutrification and sedimentation in isolation and in all combinations to the pristine model reefs, average equilibrium coral covers declined under each scenario, with concomitant increases in average equilibrium algal covers (Fig 3A). The results clearly indicate that a coral-algal phase shift can occur without fishing; under sedimentation alone, average coral cover decreased to 14% and average total algal cover (turf algae plus macroalgal cover) increased from 18% to 27%, and under the combined effects of nutrification and sedimentation, average coral cover decreased further to 11% and average total algal cover increased further to 43%. While average macroalgal cover was low under each scenario (<10% for the pristine scenario and seven scenarios with added stress), macroalgal cover values for individual runs under each scenario always encompassed a wide range of at least 0–81% macroalgal cover, reflecting strong high non-linearity in macroalgal dynamics with changing parameter values. Importantly, this represents the range of macroalgal covers found on real reefs in Mesoamerica [1,28,59]. Under each stressor scenario examined, the percentage of parameter sets yielding multiple stable equilibria was <1%, supporting the view that discontinuous phase shifts are rare relative to continuous ones (Fig 3B; ). This bodes well for management intervention because in the case of a continuous phase shift, all else being equal the affected reef would return to high coral cover once issues related to water quality and overfishing are rectified. Nevertheless, our results indicate that discontinuous phase shifts are possible and nutrification in particular was found to increase the probability of multiple stable equilibria and thus hysteresis by a factor of around five, when acting in isolation, or a factor of around six, when acting together with fishing (Fig 3B). Source: http://doi.org/10.1371/journal.pone.0174855