Diploma Thesis
Verschiebung der Umweltnische im Invasionsprozess – das Beispiel der Tigermücke Aedes albopictus
Nils Tjaden (07/2010)
Support: Björn Reineking, Carl Beierkuhnlein
Summary
Niche-based models are increasingly used to predict species’ dispersal in space and time based on information about their preferred environmental conditions. So far, the assumption of niche-conservatism is crucial to most modelling techniques and the possibility of niche-shift occuring during the process of invasion is not being regarded. However, current publications show that niche-shift can influence the explanatory power of niche-based models (Medley 2010, Rödder & Lötters 2009, Broennimann et al. 2007). Using the Asian tiger mosquito Aedes albopictus (Skuse 1894) as an example, I analyse how differences in the environmental conditions of different ranges of distribution can contribute to those shifts.The modelling software Maxent (Phillips et al. 2006, Phillips & Dudík 2008) was used to generate models for occurence-data from different continents, which were then projected to each of the other continents respectively. A comparison of the prognosticated probabilities of occurence revealed distinct differences between the combination of environmental variables favoured by Ae. albopictus in the different continents. Human Influence, represented by the Human Influence Index (Sanderson et al. 2002) was of huge importance for the distribution of Ae. albopictus, leaving the more »conventional« environmental variables used far behind.
Furthermore, it is shown that the meaning of single environmental variables may vary between different continents, which may cause models trained on the data of one continent to fail in predicting the distribution of Ae. albopictus in other continents.
A principal component analysis (PCA) was calculated for all environmental variables used in the models. The results show, that significant differences exist in the availability of environmental conditions relevant for the occurence of Ae. albopictus. When occurence data was projected into environmental space, it could be shown that the direction of niche-shift in environmental space is associated with the local availability of preferred environmental conditions.