Seafloor characterization and monitoring of the inner shelf in northern Israel using remote sensing

TitleSeafloor characterization and monitoring of the inner shelf in northern Israel using remote sensing
Publication TypeConference Abstract
AuthorsPe'eri, S, Tibor, G, Madore, B, Illin, V, Ben-Avraham, Z, Rilov, G, Ketter, T, Dijkstra, JA
Conference NameOcean Optics XXII
Conference DatesOct. 26-31, 2014

Coastal marine systems are experiencing rapid transformations as a result of multiple human-induced stressors. Many of these transformations occur at the level of a landscape and as such it is increasingly important to detect changes at these appropriate spatial scales. Remote sensing can be a powerful tool to investigate landscape level changes, but it still requires calibration and ground-truth via field observations. In this research, classification schemes using multispectral and hyperspectral imagery were developed in order to map and monitor temporal changes in the benthos over a fifteen-year period (1999-2014) along the Levant Mediterranean rocky reefs that are made of sandstone (known as Kurkar) and are found along the northern coast of Israel. The Levant Mediterranean rocky reef has been experiencing major temporal shifts in marine biodiversity. These shifts include, but are not limited to, the creation of reefs by large oysters that invaded from the Red Sea, the addition of non-native seaweed species, and over-grazing of seaweed beds by non-native herbivorous fish. Such changes can likely be observed at the landscape level.  The bathymetry of the reef is characterized by sandstone ridges that are parallel to the shoreline with emerging islands at their top and a long-shallow channel between the ridge and the shore. Using bathymetric models, a correction for the water attenuation was applied to the spectral dataset and an optical extinction depth was calculated. A recent field campaign (May, 2014) in the study area using acoustic (side scan sonar) and underwater optical measurements provided a ground truth dataset for the study. Decision trees used for the classification were developed based on in situ spectral measurements and underwater video imagery.  A time series of seafloor characterization maps was derived from imagery acquired using Itres CASI, Landsat 7 and Landsat 8.