@phdthesis{Gogina2010,
  author    = {Mayya Gogina},
  title     = {Investigation of interrelations between sediment and near-bottom environmental parameters and macrozoobenthic distribution patterns for the Baltic Sea},
  journal    = {Untersuchung von Wechselbeziehungen zwischen Bodensatz und Bodenumweltparameters und makrozoobenthik Verteilungsmustern f{\"u}r die Ostsee},
  url       = {https://nbn-resolving.org/urn:nbn:de:gbv:9-000803-1},
  year      = {2010},
  abstract  = {The objectives of the present work are to relate the spatial distribution of benthic macrofauna in the Baltic Sea to patterns in environmental variables describing near-bottom hydrographical conditions and sediment characteristics, analyzing the data for two various spatial extents. The first case study is devoted to an exploratory statistical description of the prevailing ecological structure within the limited area attached to the region of the Mecklenburg Bight. Key environmental descriptors of spatial distribution of macrofaunal communities were disclosed within the area of investigation: water depth, regarded as a proxy for other environmental factors, and total organic content. Distinct benthic assemblages that are discriminated by particular species (Hydrobia ulvae–Scoloplos armiger, Lagis koreni–Mysella bidentata and Capitella capitata–Halicryptus spinulosus) were defined. Each assemblage is related to different spatial subarea and is characterized by a certain variability of environmental factors. This study represented the basis for the predictive modelling of species distribution in the selected investigation area, which constituted the next part of the investigation. Species-specific models predicting the probability of occurrence relative to environmental and sedimentological characteristics were developed for 29 representative macrofaunal species using a logistic regression modelling approach. Subsequently, the technique for a predictive modelling of species distributions in response to abiotic parameters based on single-factor logistic regression models, utilizing Akaike’s information criterion (AIC) and Akaike weights for multimodel inference, was used. Thus, probabilities of occurrence for selected exemplary species (Arctica islandica, Hediste diversicolor, Pygospio elegans, Tubificoides benedii and Scoloplos armiger) were modelled and mapped. Finally, the investigation proceeded on a large spatial scale. The discriminating ability of such factors as salinity, bathymetry, and sediment characteristics (considered only generally due to the lack of more detailed data) to explain the occurrence of typical macrozoobenthic species on the Baltic Sea-wide extend was tested. Full coverage macrofauna distribution maps, though being increasingly demanded, are generally lacking, with information being merely restricted to point observations. In contrast to spatial interpolation, periled by presence of short distance changes in community structure and dependence of the result on density of the samples, predictive habitat suitability modelling allows to objectively produce distribution maps at a level of detail limited only by the availability and resolution of the environmental data. Various literature sources and available databases were analyzed in respect to the information on macrozoobenthos distribution in the Baltic Sea, resulting in the compilation of an extensive list of taxa and an inventory dataset on species distribution for the whole Baltic Sea. The study demonstrates the need to analyze species’ relationships in gradient systems such as the Baltic Sea and provides a basis for a tool to predict natural and anthropogenic forced changes in species distribution.},
  language  = {en}
}