Abstract:
Conditions for syntopic occurrence of the butterfly Maculinea arion and its host ant Myrmica sabuleti
Case study on endangered biodiversity in semiarid grassland of the Swabian Alb
The butterfly Maculinea arion is an extremely vulnerable species and endangered all across Europe due to its complex habitat requirements. Because of anthropogenic environmental changes, these demands can no longer be met. The species depends, on the one hand, on exclusive larval food plants, mainly Thymus pulegioides and, more rarely, Origanum vulgare. On the other hand the caterpillars of Maculinea arion need a syntopic presence of their only host species, the ant Myrmica sabuleti. In a three years field-study the habitat conditions required by both the butterflies and the ants were analysed in their South German range. The field work was designed for an evaluation of the factors enabling the syntopic occurrence. My corresponding hypotheses on the environmental situation have been tested and verified.
The results confirmed that Maculinea arion depends on a biotope with particularly unique habitat qualities. It was necessary to identify key parameters for the co-occurring of the butterfly and its host ant in order to protect the remaining populations in the Swabian Jura and to enable their recovery. After the quantification of single parameters and calculations with multivariate statistics I was able to establish a correct biotope model. Proposals for future practical habitat management could be deduced accordingly.
Results were obtained on the following topics.
1. Habitat requirements and competition of the host ant Myrmica sabuleti compared to several co-occurring Myrmica –species
At the research sites, beside the host ant M. sabuleti a number of congeneric species was found, in particular M. rubra and M. scabrinodis and, more rarely, M. schencki.
Host specifity of M. arion larvae: In fact only one M. arion larva was discovered in a nest of M. sabuleti. For the first time in Germany this confirms in that our region M. sabuleti really is the host ant of M. arion. To obtain this result, it was necessary to excavate 124 nests nearby 11 egg-laying spots of the butterflies. Unfortunately the appearance of an adult butterfly leaving a host ant nest could not be observed, despite 80 emerge traps which were placed over nest entrances in 15 egg-laying-spots marked during the previous year. A lot of field work had to be invested to get a few data.
Host capacity of M. sabuleti nests: In order to evaluate the capacity of nests to host a caterpillar of M. arion, the ant population of 18 nests was completely counted. Only 50% of the nests contained enough ant larvae representing food available for nutrition of one M. arion caterpillar to complete the development until diapauses. Consequently a high mortality of the caterpillars inside host nest due to lack of food has to be considered in this local community.
Temperature at the nest entrance: At the test sites an electronic recording system, especially designed for this purpose, was placed close to 64 entrances of M. sabuleti and M. rubra nests for long time registration of the temperature on the soil surface. The results indicate that M. sabuleti colonies were found in microhabitats warmer than the locations of M. rubra nests. The mean difference was 1.2 K over the whole recording period, suggesting specific nest-site preferences.
Habitat structure and requirements of the ant species: Habitat parameters such as bare ground, plant density, solar radiation and indicator values of occurring plant species like F – humidity, N – amount of nitrate in the soil, L – light were determined in two M. arion habitats close to 494 nest entrances of the four co-existing Myrmica species. The rank position of these biotic and abiotic parameters was calculated in relation to their importance for all the ant species occurring there. Significant differences between the different Myrmica taxa were proved for all parameters except temperature. Solar radiation was decisive for higher nest densities of M. sabuleti. The species does not tolerate high portions of open soil and avoids high nitrate content and humidity of the soil at their nest spots. The analysed habitat parameters have a significant quantitative influence especially on occurrence of the host ants their nesting density. By a multivariate regression model developed for this project both occurrence (R² = 0.4) and nest density (R² = 0.1) could be predicted partly.
Competition: The local distribution of nests of the butterfly’s host ant was monitored at the study sites and tested by ´Nearest-Neighbour-Analysis´. Inter- and intra-specific competition in the co-occurring Myrmica-species could not be proven. The nests of the different ant species were overall distributed randomly with only few exceptions. The conclusion is that the observed patterns are not due to intra- or inter-specific competition.
2. Analysis of the population and habitat preferences of the butterflies
Marc-recapture studies were performed at the two study sites during three years to obtain population parameters and data on the mobility of the butterflies. The flight period of M. arion lasted from mid June until end July, on the average, 31 days. The population at the Kirchenköpfle declined significantly and continuously from 792 to 284 adults, obviously due to reduced grazing intensity which caused deterioration of important habitat parameters. However, at the study area Kälberberg/Hochberg the population increased slightly. There the meadows were still grazed intensely by sheep.
Two marked females were recaptured at 1,525 m and respectively 1,875 m distances in neighbouring habitat areas which were partly separated by forests. These flight distances are remarkably longer as observed elsewhere. Gene flow and also the local foundation of new populations in habitats where the butterflies became extinct seem to be possible in a metapopulation range within a few kilometres of site distances.
Habitat preferences: In both study areas flowering plants of the family Lamiaceae, especially Thymus pulegioides, were the nectar source preferred by the butterflies.
Using the GIS-based digital landscape model, in the study areas Kälberberg/Hochberg and Kirchenköpfle the total space of habitats suitable for M. arion was calculated to comprise 14.3 ha and 3.9 ha, respectively. These are potential biotopes because both oviposition plants and host ants were abundant there.
The aim of my project, to specify key parameters in the habitat requirements of the Large Blue Butterfly, was successfully reached. Based on the established landscape model, now management measures can be scheduled for the protection of the endangered remaining populations. According to the calculated predictions, they will survive for the next 50 years in the study areas Kirchenköpfle and Kälberberg/Hochberg if a minimum area of 12 ha habitat with per ha at least 40 butterflies, 300 nests of the host ant M. sabuleti and a 20% coverage of the vegetation with Thymus plants is given. Because the structure of the environment is always subject to stochastic fluctuations, which might affect the vulnerable butterfly populations and possibly also the nest density of the host ant, a larger area should be included into protection. However, the recorded mobility of the butterflies will probably enable connections between the remaining populations of this rare butterfly species, thus preventing a further decline by forming a stable metapopulation of M. arion.
Conclusions
The hypothesis underlying this research project of a widely congruent ecological adaptation of the butterfly and its only host ant was confirmed by the results obtained. This enables and entails the syntopic occurrence of both species. Presumably through coadaptions, an obligate myrmecophily developed by coevolution. More recently this symbiosis switched into a parasitic status of the caterpillars during their intranidal phase. The abiotic and biotic factors deciding the suitability of a habitat for both butterfly and host ant could be quantified and weighted against each other. A GIS-based landscape model was designed for the two study areas. Predictions became possible which changes of environmental conditions would result in decrease or increase of the butterfly populations. The use of area-related planning, based on multi factorial analysis, is regarded a meaningful and solid requirement for the conservation of rare species.