Markov chain analysis of sessile community dynamics in a degraded Philippine reef to support restoration of coral populations

Abstract

Quantitative approaches are needed for assessing assisted recovery potential in degraded coral reefs. A lagoonal community (Bolinao, Philippines) that has experienced severe disturbances (overfishing, blast fishing and coral bleaching) in the past was observed for 26 months. Benthic state sequences in 4 × 4 cm patches were used to quantify monthly probabilities of transitions between reef states. Benthic cover distribution was consistent with a spatially heterogeneous Markov process, with high variability in transition probabilities within a subset of four states: crustose algae with cropped turf, the codominant sponge Callyspongia samarensis, articulated coralline algae and fleshy macroalgae (MA). Once a patch is dominated by any of these states, there is high likelihood of cycling within the set before escaping to the rarer invertebrate groups. The assemblage is unlikely to recover naturally given prevailing conditions. Patches dominated by juvenile and adult corals have mean turnover times of about 3 and 5 months, respectively, due to (partial) mortality and competition. Asymmetry was detected for coral–macroalgal competition, despite low fleshy algal cover (9.5%), that was more adverse for coral juveniles than adults. While competition between coral and the mat-forming C. samarensis was symmetrical, loss of coral cover through this path is relatively higher as a result of the higher interaction frequency. Articulated coralline algae do not appear as a constraint. Complementary strategies for assisted recovery were inferred from successional indices as well as the sensitivity of stationary coral cover to changes in transition probabilities. Results demonstrate that short-term, fine-scale observations of state variables can be used to resolve biotic constraints to inform restoration initiatives.