Recovery of sponges after extreme mortality events: morphological and taxonomic patterns in regeneration versus recruitment.

Sponges on the Belize Barrier Reef were drastically diminished in 2 mortality events during a 6-year period. Patterns in mortality and recovery were revealed by detailed records of assemblage dynamics, in the form of 8 full censuses between June 2006 and July 2012. All sponges on a set of shallow patch reefs were mapped, identified, and measured for volume at yearly or more frequent intervals, allowing population dynamics of 54 sponge species to be quantified with respect to volume of live tissue and number of individuals. The degree to which sponges of different species suffered during each of the mortality events ranged from complete loss to no effect, resulting in immediate significant alterations in composition of the assemblage in addition to extreme losses (49% in 2008 and 71% in 2011) of the volume of living sponge. The repeated census data documented the early stages of recovery, both on the assemblage level and for every individual sponge. Groups of sponge species, defined by higher taxa or by growth form, not only experienced mortality very differently, but also recovered differently, with some showing efficient regeneration after partial mortality, others adding small individuals by recruitment, and still others not recovering at all. During the 2008 mortality event, losses of both volume and numbers of individuals were disproportionately heavy for members of the orders Aplysinida and especially Poecilosclerida, and for sponges of erect branching and semi-cryptic massive growth forms. Post-mortality recruitment was meager, resulting in no rebound in numbers of individuals; regain of volume lost was slowed by the extreme loss of many entire individual poecilosclerids. By contrast, during the 2011 mortality event, loss of volume was disproportionately heavy for members of the orders Hadromerida and Dictyoceratida, while numbers of individuals were lost from all orders in the same proportion. Among the growth forms, volume was disproportionately lost from massive and semi-cryptic massive sponges, while numbers were only disproportionately lost from the semi-cryptic massive species. During the first 7 months of recovery, a quarter of the lost volume and numbers were re-gained, by a combination of regeneration and recruitment. Relatively rapid early recovery stages reflected the high degree to which losses of volume were due to substantial partial mortality of massive sponges. Significant differences between the 2 mortality events in loss and recovery patterns among higher taxa, as well as among species within particular higher taxa, provide strong impetus for accurate identification to species of all sponges encountered in monitoring, and for continuing efforts aimed at understanding all levels of sponge systematics. In turn, patterns of mortality and recovery may be able to contribute additional characters for systematics.