Publication Details

Category Text Publication
Reference Category Qualification assignments
Title (Primary) Analysing effects of spatiotemporally distributed species interactions in Maculinea systems
Author Singer, A.
Year 2006
Department OESA
Page To 228
Language englisch

We analyse effects of species-interaction and spatiotemporal host distribution on persistence of Maculinea populations at isolated habitat sites. Maculinea butterflies are parasites of Myrmica ants. The butterfly caterpillars infest host ant nest in the vicinity of their initial oviposition plant. Thus, spatial distribution of initial host plants (oviposition plants), implies a spatial distribution of parasitism. We develop a generic spatially-explicit rule-based simulation model for a Maculinea systems, to analyse the influence of different spatially relevant parameters on the performance of a Maculinea population. Parameter variation considers spatial host plant distribution, initial spatial host ant distribution and budding-range, which defines the probability distribution of colonising distances of ants. In simulation runs, it can be seen that the spatial distribution of host ants adapts to the spatially distributed parasitism from Maculinea caterpillars. Areas without host plants (unexploited areas) are completely inhabited by host ants. In areas with host plants (exploited areas) mean density of host ants is lower. Ant nests show dynamics. They are abandoned or nest sites are recolonised. Because of spatial segregation of exploitable and non-exploitable areas and by small scale dispersal of host ants (budding), areas of different effectiveness for Maculinea population performance can be distinguished on a Maculinea habitat site: These are reproductive area, buffer, sink and non-contributing area. Consequently, number of host plants is not the only decisive factor for performance of a Maculinea population. As well it is important how well empty nest sites in the vicinity of host plants can be colonised by host ants. We find two different mechanisms, which limit re-growth of the Maculinea butterfly population and lead to different types of dynamics. Population viability analysis (PVA) is a standard method for single species systems. However, the Maculinea system consists of interacting species with more complex dynamics. There is still no standardised methodology for PVA of multi-species systems. We derive a phenomenological description of distributions of extinction times for populations in a large class of systems of interacting species. We find in analytical calculations that the long term behaviour of distributions of times to extinction in multi-species systems can be analysed with the ln(1-P_{0})-method suggested for single species systems. Thus, this method can be transferred. Comparing distributions of times to extinction of single and interacting species, we find that long term extinction of established populations follows a similar process. However, on short time-scales, population cycles modulate extinction risk. In systems of interacting species, an initial transient phase can have strong influence on persistence of the population. In simulations of the Macu model, we observe that it takes up to 50 generations until the system reaches its established state. This can lead to a bottleneck of the population. We develop a landscape analysis method for multi-species systems. For that purpose, suitability of different landscapes is assessed for their ability to sustain particular populations. Suitability is measured by population persistence. Therefore, dynamics of the system and requirements of the populations are taken into account when assessing suitability. This method is applied to assess effects of the influence of different spatially relevant parameters. No major effects of initial host ant distribution on persistence of Maculinea populations can be found. In contrast, spatial distribution of host plants has a main effect. Thus, the host plant distribution can be considered as landscape structure of the Maculinea system on the site. Rules of thumb for suitable structures of host plant distributions and a spatial index are given. In general heterogeneous host plant distributions of intermediate density are a good option.

Persistent UFZ Identifier
Singer, A. (2006):
Analysing effects of spatiotemporally distributed species interactions in Maculinea systems
Dissertation, Philipps-Universit├Ąt Marburg, Fachbereich Physik
228 pp.