Geomorphological and ecological characterization of a hard-substrate complex in a temperate shallow shelf sea (SE North Sea) using hydroacoustic sensors and machine learning algorithms

Abstract

<p>The North Sea is a shallow marine environment. The sediment distribution of the seabed is dominated by sand-sized material. Hard-substrate areas are a relatively rare, but important habitat for sessile and mobile species. This habitat type forms island-like geomorphic features owing to the presence of glacial deposits in the shallow subsurface. While their ecological importance is widely acknowledged, hard-substrate areas are characterized by a large degree of spatial heterogeneity and an unaccounted high local diversity in physical surface properties, sediment composition and temporal change.</p><p>The aim of this study is the detailed investigation into the spatial characteristics and temporal variability of an exemplary hard-substrate complex located 10 km offshore the island of Sylt (N-Germany). The area has a size of c. 3 km<sup>2</sup>and was investigated between 2008 and 2019 using a range of hydroacoustic and optical sensors (multibeam echosounder, sidescan sonar, sub-bottom profiler, acoustic ground discrimination system, underwater videos) and machine learning algorithms (haar-like features) to track the changes in the number and local distribution of exposed stones.</p><p>The maximum water depth in the area is 16 m and a linear arrangement of hard substrates emerges up to 4 m from the seabed. A layer of fine sand with a thickness of 0.5 m overlays the more planarly deposited coarse sediments in the proximity of the stony outcrop. This layer of fine sand is relatively mobile and leads to a frequent temporal change of the distribution of sediment on the seabed, whilst the stony outcrop is only marginally affected by the spatial dislocation of sediments. The spatial extent of hard substrates is variable due to the presence of a mobile sand cover on the seabed.</p><p>This study emphasizes the need for quick and automated object classification routines to be integrated in monitoring approaches in the highly dynamic coastal zone. It has shown that the geomorphological diversity and interannual variability of hard-substrate areas can be captured using the presented approach. Detailed studies and monitoring tools are important to better understand the interrelation of geomorphological and sedimentary processes at the seabed with the ecology of epibenthic organisms.</p><p><strong>Keywords: </strong>North Sea; hard-substrate habitats; mobile sediments; hydroacoustic; haar-like features</p>