Modelling and forecasting long-term dynamics of Western Baltic macrobenthic fauna in relation to climate signals and environmental change

Long-term macrobenthos data from Kiel Bight in the Western Baltic collected between 1968 and 2000 have been correlated with the winter NAO index (North Atlantic Oscillation Index) and other environmental data such as temperature, salinity and oxygen content in the bottom water in order to detect sys...

Full description

Bibliographic Details
Published in:Journal of Sea Research
Main Authors: Gröger, Joachim Paul, Rumohr, Heye
Format: Article in Journal/Newspaper
Language:English
Published: 2006
Subjects:
NAO
Online Access:https://doi.org/10.1016/j.seares.2005.11.005
https://www.openagrar.de/receive/timport_mods_00030630
https://www.openagrar.de/servlets/MCRFileNodeServlet/timport_derivate_00030630/dk039327.pdf
Description
Summary:Long-term macrobenthos data from Kiel Bight in the Western Baltic collected between 1968 and 2000 have been correlated with the winter NAO index (North Atlantic Oscillation Index) and other environmental data such as temperature, salinity and oxygen content in the bottom water in order to detect systematic patterns related to so far unexplained abiotic signals in the dynamics of zoobenthic species assemblages. The benthos data come from a cluster of five stations (Süderfahrt/ Millionenviertel) in Kiel Bay. Our investigations concentrated on the macrobenthic dynamics with a focus on the number of species m- 2 (species richness). Using logarithms and the time series analysis approach of Box/Jenkins (ARIMA modelling, transfer function modelling) it was shown that species richness was strongly influenced by the winter NAO (adjusted for a linear time trend within the 1968-2000 period) and salinity (with a shift/lag of four years). Bootstrapping experiments (i.e. sampling from the error process) and analysis of prediction power (by means of the one- or more-years leaving-out method) showed that the parameter estimates behaved in a stable way, leading to a relatively robust model.