A habitat classification scheme for seamount landscapes: assessing the functional role of deep-water corals as fish habitat

Abstract

Seamounts are drowned volcanoes rising from abyssal depths. Fishes on seamounts exploit a range of landscape features that likely enhance probabilities of prey capture and reduce predator success. The epifaunal community on seamounts is dominated by suspension-feeders of which deep-water corals are a dominant element. Such taxa are widespread components of seamount landscapes but their functional role in mediating the distribution and abundance of fi shes remains unknown. Here we propose a hierarchical habitat classifi cation matrix, which includes deep-water corals, as a foundation for partitioning seamount landscapes in which fi shes are observed. This scheme is based on our observations of fi sh distributions from the New England Seamounts, as well as literature review. Features of an idealized seamount landscape were divided at multiple spatial scales and included features at habitat class, subclass and microhabitat levels. Habitat classes were divided by major sediment types (i.e., basalt, fi ne grained sediments). Habitat subclasses included pavement, ridges, walls, ledges and tubes for basalt substrates and fl at sediment, ripples and waves for fi ne-grained sediments. Microhabitat features were classifi ed as fl ow related features, emergent structures (i.e., geologic and biologic including deep-water corals), and other biogenic structures (e.g., coral debris, depressions, burrows). Variations in the distribution of structures at multiple spatial scales can infl uence boundary fl ows and the ability of fi shes to search for prey (e.g., where active searching by swimming can occur, where pelagic prey delivery is suffi cient when station-keeping) and avoid predators (e.g., the ability to effi ciently exhibit various avoidance behaviors such as shelter seeking). Placing fi sh abundance data in such a matrix of habitat types enables a variety of statistical approaches for testing for non-random distributions of fi shes on seamounts and quantifying the functional role