2012
JOURNAL OF BIOGEOGRAPHY

Global distribution and habitat preference of deep-sea octocorals

C Yesson, M Taylor, DP Tittensor, A Davies, J Guinotte, A Baco-Taylor, J Black, J Hall-Spencer, AD Rogers.

Abstract

Three-quarters of Octocorallia species are found in deep waters. These coldwater octocoral colonies can form a major constituent of structurally complex
habitats. The global distribution and the habitat requirements of deep-sea
octocorals are poorly understood given the expense and difficulties of sampling at
depth. Habitat suitability models are useful tools to extrapolate distributions and
provide an understanding of ecological requirements. Here, we present global
habitat suitability models and distribution maps for seven suborders of
Octocorallia: Alcyoniina, Calcaxonia, Holaxonia, Scleraxonia, Sessiliflorae,
Stolonifera and Subselliflorae.
Location Global.
Methods We use maximum entropy modelling to predict octocoral distribution
using a database of 12,508 geolocated octocoral specimens and 32 environmental
grids resampled to 30 arc-second (approximately 1 km2
) resolution. Additionally,
a meta-analysis determined habitat preferences and niche overlap between the
different suborders of octocorals.
Results Suborder Sessiliflorae had the widest potential habitat range, but all
records for all suborders implied a habitat preference for continental shelves and
margins, particularly the North and West Atlantic and Western Pacific Rim.
Temperature, salinity, broad scale slope, productivity, oxygen and calcite
saturation state were identified as important factors for determining habitat
suitability. Less than 3% of octocoral records were found in waters
undersaturated for calcite, but this result is affected by a shallow-water
sampling bias.
Main conclusions The logistical difficulties, expense and vast areas associated
with deep-sea sampling leads to a gap in the knowledge of faunal distributions
that is difficult to fill without predictive modelling. Global distribution estimates
are presented, highlighting many suitable areas which have yet to be studied. We
suggest that approximately 17% of oceans are suitable for at least one suborder
but 3.5% may be suitable for all seven. This is the first global habitat suitability
modelling study on the distribution of octocorals and forms a useful resource for
researchers, managers and conservationists.