Dokument

Summary:
A major goal of ecological research is to understand how plant populations respond to spatial and temporal variation in environmental conditions. Short-lived species rapidly respond to changes in habitat conditions, but little is known about the effects on the population dynamics of long-lived plants. I analyzed the effects of varying habitat conditions on the population dynamics of two herbaceous perennials. One of the plants, Trifolium montanum, grows in nutrient-poor grasslands in a semi-natural landscape of central Europe, while the other plant, Heliconia metallica, grows in lowland rainforests in Amazonian Peru. During the last decades, changes in land use have resulted in a considerable loss of nutrient-poor grasslands in central Europe. I studied the effects of habitat degradation and fragmentation on the dynamics of the populations of the declining perennial plant Trifolium montanum L. in central Germany using matrix models. To assess the habitat quality in a site, I measured the leaf area index (LAI). The finite growth rate of a population strongly decreased with LAI. In unmanaged sites, population growth was <1, which was mostly due to lower survival and flowering of large plants. Management by clipping rapidly increased population growth because of a higher flowering probability of large plants. The number of seeds per fruithead was not related to LAI, but increased with local density, suggesting pollinator limitation in populations with a low density of flowering plants. In a common garden, the seed production of the offspring decreased with isolation, and in contrast to previous studies, also decreased with size and density of the population of origin. This might be due to increased inbreeding because of pollination between closely related neighboring plants in dense and large populations. In stochastic simulations, the median time to extinction of unmanaged populations of 100 flowering plants was about 50 years. However, small populations though managed were threatened by stochastic extinction and required a minimum number of 60 flowering plants to survive with 95% probability over the next 100 years. However, most populations of T. montanum in central Germany consist of less than 50 flowering plants. Many populations of other perennial plants of semi-natural grasslands might also be doomed because of the low quality of their habitats and their small size. However, the extinction process may take a long time in perennial plants, resulting in an extinction debt. The actual plant diversity of the remnant sites might thus be a misleading indicator of their conservation status. Large areas of tropical rainforests are still shaped by natural processes. Clonal herbs are an important component of the understory of these forests. However, very little is known about the population dynamics of these herbs. In a Peruvian floodplain forest, I analyzed the influence of seasonal and spatial environmental variation on the populations of the clonal herb Heliconia metallica Planchon & Linden ex Hooker over two wet and two dry seasons, using periodic matrix models. All populations increased in size during the wet and decreased during the dry season. Finite annual growth rates of non-flooded populations were <1 both in gaps and non-gaps. In contrast, growth rates of flooded populations were close to 1 in non-gaps and significantly >1 in recently formed gaps. The lower growth rate at non-flooded sites was due to lower survival of ramets during the dry season. In stochastic simulations, the extinction risk of the non-flooded populations of H. metallica was high, whereas that of flooded populations was low and decreased with the frequency of gap formation. Seed addition in unoccupied sites increased seedling recruitment, in particular in experimentally disturbed sites. The results indicate that the population dynamics of H. metallica are shaped by the interacting effects of seasonal flooding and of transient canopy gaps. Ramet survival and clonal growth are crucial for the persistence of populations of H. metallica in the forest understory, while the colonization of unoccupied sites is limited by insufficient dispersal of seeds to disturbed sites. The environmental differences between the wet and dry season affected the populations of H. metallica, and all populations decreased in size during the dry season. However, an increase in the length of the dry season has been observed in Amazon rainforests during the last decades. The results of this study suggest that the clonal life form of H. metallica enables populations to persist under variable light conditions, but might not be able to buffer the effects of drought stress during longer dry seasons. Climate change might cause a decrease in the regional abundance of the widespread clonal understory herbs of Amazon rainforests.

Das Dokument ist im Internet frei zugänglich - Hinweise zu den Nutzungsrechten