Determining Coral Density Thresholds for Identifying Structurally Complex Vulnerable Marine Ecosystems in the Deep Sea

Vulnerable marine ecosystems (VMEs) are at risk from the impacts of deep-sea trawling. Identifying the presence of VMEs in high seas fisheries management areas has to date relied mainly on presence records, or on habitat suitability models of VME indicator taxa (e.g., the stony coral species Solenosmilia variabilis Duncan, 1873) as proxies for the occurrence of VMEs (e.g., cold-water coral reefs). However, the presence or predicted presence of indicator taxa does not necessarily equate to the occurrence of a VME. There have been very few attempts to determine density thresholds of VME indicator taxa that relate to a “significant concentration” which supports a “high diversity” of associated taxa, as per the current criterion for identifying structurally complex VMEs (FAO 2009). Without knowing such thresholds, identifications of VMEs will continue to be subjective, impeding efforts to design effective spatial management measures for VMEs. To address this issue, we used seafloor video and still image data from the Louisville Seamount Chain off New Zealand to model relationships between the densities of live Solenosmilia variabilis coral heads, as well as percent cover of live and dead coral matrix, and the number of other epifauna taxa present. Analyses were conducted at three spatial scales; 50 m2 and 25m2 for video, and 2 m2 for stills. Model curves exhibited initial steep positive responses reaching thresholds for the number of live coral heads at 0.11 m-2 (50 m2), 0.14 m-2 (25 m2), and 0.85 m-2 (2 m2). Both live and dead coral cover were positively correlated with the number of associated taxa up to about 30% cover, for all spatial scales (24.5 to 28%). We discuss the results in the context of past and future efforts to develop criteria for identifying VMEs.