Biodiversity and ecosystem processes in an experimental island system
von Hagen Andert
Datum der mündl. Prüfung:2017-11-15
Erschienen:2018-02-16
Betreuer:Prof. Dr. Teja Tscharntke
Gutachter:Prof. Dr. Teja Tscharntke
Gutachter:Prof. Dr. Christoph Scherber
Gutachter:Prof. Dr. Holger Kreft
Dateien
Name:Hagen Andert-Dissertation-complete-final.pdf
Size:4.78Mb
Format:PDF
Zusammenfassung
Englisch
The field of island research is one of the best researched topics in ecological and evolutionary research. As islands are spatially restricted and basic principles of island colonization are investigated, they display one of the best research patches to be based on. Islands are classified into oceanic islands, continental shelf islands, habitat islands or non-marine islands. Within the last decades, most studies focused on oceanic islands as they are good research models for habitats, which started to establish millions of years ago. In the context of global warming and rising sea levels, islands have also attracted public attention due to more frequent inundation and tsunami events in the last couple of years. Barrier islands are mostly located in the offshore region of mainland region with access to the open sea. As they are not too much isolated from mainland region, they are more or less densely populated. They are only built by offshore sedimentation processes. On the coastal line from the Netherlands op to southern Denmark, the East Frisian Islands belong to this worldwide distributed island type. Heavy thunderstorms and storm tide tremendously affect island maintenance and formation. Therefore, about 40 years ago, humans started to save these islands by constructing dykes and bulwarks to reduce wave energy and safe dunes from strong erosion processes. Our research area Spiekeroog belongs to the East Frisian Islands and is one of the larger islands of this barrier island chain in front of the German North Sea coastline. To generate knowledge, how barrier islands establish in the North Sea area, together with the University of Oldenburg, we constructed 12 experimental islands along the southern part of Spiekeroog to observe initial colonization processes of arthropods onto our artificial islands. We constructed six planted (filled with Wadden Sea sediment and Lower Salt Marsh sods) and six non-planted (Wadden Sea sediment filled only) experimental islands and related control plots in the salt marsh habitat of Spiekeroog. Our research aim was to (I) analyze general island specific characteristics (such as sedimentation and erosion rates) of all East Frisian Islands as main variables which influence colonization processes by different taxa, (II) observe the abundance and probability of taxon occurrence of different arthropod taxa colonizing our experimental islands and combine them regarding our experimental treatment and to (III) focus on animal decomposition in salt marsh habitats both in control and experimental island plots to generate information about ecosystem functions in these highly disturbed habitats. In the second chapter we analyzed more than 2,990 species across 36 taxon groups (including vertebrates, invertebrates and land plants) on German barrier islands, the East Frisian Islands. We tested for relationships between species richness and island area (SAR), for effects of island habitat heterogeneity (SHH) and further island parameters using binomial logistic regression modelling. The positive impact of annual sedimentation rates of barrier islands on rove beetles and flies illustrate the importance of considering ontogenetic island data in the biodiversity analysis of barrier islands. Four taxa such as spiders (173 species) and lichens (94 species) were primarily related to area. The heterogeneity of habitat types was a key predictor for the richness of 24 taxa, including bees (101 species), Auchenorrhyncha (131 species), Saltatoria (13 species) and further 21 taxa, whereas richness differences of 6 taxa could not be explained. Overall, richness of taxa differed greatly in their responses, with area (although varying from 0.1 to 38.9 km²) playing a minor and island heterogeneity a major role, while barrier island-specific sedimentation rates emerge as a new predictor variable in models explaining species richness (14 out of 31 taxa in total). In the third chapter we set up six planted (with salt-marsh vegetation) and six non-planted experimental islands at a distance of c. 500 m south of the North Sea island of Spiekeroog (Germany) to study colonization dynamics of mobile arthropods in a standardized way. We sampled c. 40,000 arthropods during one season (May-September) across eight taxa using storm-proof window, funnel and sticky traps, placed on each of the 12 islands and on six natural salt marsh plots of Spiekeroog. Seven of eight taxa (Aphididae, Araneae, Brachycera, Coleoptera, Hymenoptera, Nematocera and Thysanoptera, but not Auchenorrhyncha) had lower abundances on the experimental islands compared with the Spiekeroog control plots, the presumable source area of the arthropods colonizing the experimental islands. Taxon differed in time of maximal colonization (interaction of island abundance with month of sample) in seven of eight taxa (i.e. with the exception of Araneae). In contrast to expectations, colonization of planted vs non-planted islands did not show major differences: In five taxa, the non-planted islands showed significantly lower abundance than the Spiekeroog control plots, and in four taxa the planted islands. Our results showed that island colonization by these mobile arthropod taxa went quickly already in the first season after island establishment, while arthropod groups appeared to discriminate between planted and non-planted islands in contrasting ways. In the fourth chapter we performed an animal decomposition experiment using an experimental approach on the barrier island Spiekeroog, Germany. 264 pieces of cow meat were spread over 12 experimental islands and six control plots established in late 2014 to observe animal decay in a real-time experiment. Our two weeks experiment displayed the influence of time and experimental treatment (experimental island or control) on carcass biomass loss, whereas the habitat zonation on each of the 18 plots (pioneer zone, lower salt marsh and upper salt marsh) did not contribute to the explanation. Animal extraction from the carcass (a piece of cow meat) and the soil beneath the carcass highlighted maggots as the taxon with highest abundances, with Scatophaga stercoraria (n=670), Hydrotaea dentipes (n=590) and Spelobia luteilabris (n=235) playing a major role. Maggot community composition changed between marine pioneer zone up to terrestrial upper salt marsh zone. In pioneer zone plots, we recorded three maggot species, whereas in upper salt marshes, we found ten species. Interestingly, higher species richness of maggots was related to higher nutrient release during decomposition, resulting in a lower ratio of carbon and nitrogen in the soil under the carcass. Overall, we argue, that our barrier island system extend general rules and patterns of classical Island Biogeography and underlines the speciality of a highly disturbed island system, which is daily influenced by tides, wind and sedimentation processes. Our studies provide first evidence that experimentally simulated successional state (planted and non-planted experimental islands vs. transplanted salt marsh control plots) influences arthropod communities much more than habitat zone identity (pioneer zone, lower salt marsh, upper salt marsh).
Keywords: BEFmate; artificial islands; barrier islands; colonization; insects; island biogeography; sea level rise