Abstract
An important goal for ecosystem‐based management is to protect marine habitats and their associated fauna. Understanding the spatial structure and interrelationships of benthic communities and their underlying environmental drivers, therefore, is of great importance. The benthic community off the sub‐Arctic Lofoten‐Vesterålen islands was studied on multiple spatial scales, using the MEM (Moran's eigenvector maps) framework to identify spatial structure in broadscale (hundreds of kilometers), mesoscale (20 km), and fine‐scale (1.5 km) epifaunal and infaunal communities. A combination of eigenvector‐based multivariate analyses and variation partitioning on multiple sets of explanatory variables was used to identify characteristic species and environmental drivers that have a high importance in structuring the communities. Community structure of both epifauna and infauna strongly reflected the boundary of warm Atlantic (>0.5°C) and cold Arctic water masses (<0.5°C), which coincides approximately with the 800‐m isobath in this region. Apart from this dominant broadscale determinant, epifauna and infauna displayed different spatial patterns and drivers. Both components differ significantly across local marine landscapes; however, this characterization is not sufficient for infauna, where additional sediment characteristics play an important role in structuring the community across all spatial scales. Within marine landscapes (mesoscale), epifauna displays a spatial structure that, for the most part, could not be attributed to any of the included environmental drivers. The case of the Lofoten‐Vesterålen region demonstrated that the used analysis is a valuable tool in spatial planning, as it allows for comparison of the relative importance of individual environmental drivers for the studied faunal components, without losing the information about spatial patterns associated with individual drivers.