Identification of biogeochemical provinces in the Southern Ocean: spatial modeling of biological, geochemical and sedimentological data
Final Report Abstract
Biogeochemical provinces are regions with relatively consistent or characteristic biogeochemical conditions, e.g. oxygen conditions or carbon fluxes. This project identifies those provinces by means of geostatistical and multivariate modeling based on geo-referenced data. They allow us to better detect biological and geochemical processes and to quantify and analyze their future change - e.g. due to climate change - in models. The better understanding of Antarctic ecosystem resilience and stability under changing conditions is one of the issues that play a prominent role in the recent research interests. The extensive interdisciplinary data sets, which were acquired and managed in a geodatabase within the project, are a valuable result that can serve as a profound basis for many future scientific questions inter alia focusing on the variability of ecosystem function in space and time. They have been acquired, processed, made visible via web map browsers of the AWI. Furthermore, results have been published in PANGAEA including data sets. Methodologically, several interpolation methods, species distribution models and cluster analyses have been accomplished and applied to two regions: Weddell Sea and Potter Cove, West Antarctic Peninsula, caused by the bundled project with Katharina Zacher and the excellent data availability in Potter Cove for testing the methods. Based on the scientific question of the project and the availability of the data the following subjects have been chosen for successful processing: Weddell Sea: • A geomorphological seabed classification for the Weddell Sea, Antarctica. • Sediment texture of the Weddell Sea, Antarctica. • Habitat modelling of crabeater seals (Lobodon carcinophaga) in the Weddell Sea using the multivariate approach Maxent. • Present and future cold water coral species distribution modelling in the Weddell Sea – identification of relevant environmental drivers. • Comparing the surface and the bottom of the Southern Ocean using multivariate cluster analysis: regional effects of environmental parameters. Potter Cove: • Habitat modeling as a predictive tool for analyzing spatial shifts in Antarctic benthic macroalgae communities due to global climate change. • Explanation of the spatial distribution of physiochemical properties of Potter Cove, Antarctica, by classification of Potter Cove, Antarctica, via k means clustering, canonicalcorrelation analysis and multidimensional scaling. Conclusions are: • data homogenization of interdisciplinary data sets is challenging and time-consuming. • GIS techniques combined with R software are a powerful combination of tools for spatial analyses. • Models require quality assessment and presentation of strength and weakness by providing e.g. the standard deviation of occurrence probability. • SDMs are still too over-simplified to account for the complexity of factor interactions in coastal climate change, but despite missing environmental factors the results certainly improve the understanding of the spatial distribution and the prediction of the possible changes in the community of Antarctic species and assemblages due to a changing environment; • the iterative exclusion of available environmental variables based on collinearity, ecological expert knowledge and the identification of variable importance provide the selection of minimum environmental parameters to establish comparable studies in other coves of the WAP, Arctic and temperate sites; • multivariate regionalization by cluster analysis provides a basis for the calculation of sparsely spread parameters as well as indication for a future reasonable sample site management improving spatial analyses.
Publications
- 2016. A geomorphological seabed classification for the Weddell Sea, Antarctica. Marine Geophysical Research 37(2): 127-141
Jerosch, K., Kuhn, G., Krajnik, I., Scharf, F.K., Dorschel, B.
(See online at https://doi.org/10.1007/s11001-015-9256-x) - Habitat modeling as a predictive tool for analyzing spatial shifts in Antarctic benthic communities due to global change. 2017 (epic.awi.de, AWI Org. > Biosciences > Functional Ecology) GeoHab 2017 – 17th International Symposium, Nova Scotia Community College (NSCC), Dartmouth, Canada, 1 May 2017 - 5 May 2017
Jerosch, K., Scharf, F.K., Deregibus, D., Campana, G.I., Zacher, K., Pehlke, H., Falk, U., Hass, C., Quartino, M.L., Abele, D.
- Habitat modelling of crabeater seals (Lobodon carcinophaga) in the Weddell Sea using the multivariate approach Maxent. Polar Biology May 2017, Volume 40, Issue 5, pp 961–976
Nachtsheim, D.A., Jerosch, K., Hagen, W., Plötz, J., Bornemann, H.
(See online at https://doi.org/10.1007/s00300-016-2020-0)
