| Species interactions A large part of the research in my group addresses questions associated with the ecological relationships between herbivorous insects and their host plants, their natural enemies (predators and parasitoids), and other organisms (competitors, mutualists). Aside from traditional two-species systems (herbivore-plant, predator-prey), multi-species systems became increasingly important in our research. In two-species systems we investigate, for example, how predator density and alternative prey shape the functional response of a generalist predator. In multi-species systems we explore the effect of third species on a two-way interaction. Questions we address here are how the population dynamics of a mutualistic interaction are modified by a shared resource, if apparent competition between two plant species can increase the herbivore pressure of a target plant, why an insect vector is necessary to transmit a rust fungus to its host plant, or how weed biocontrol is modified by the surrounding plant community. To answer these questions, we use classical manipulative field experiments, molecular methods (monoclonal antibodies, PCR) and population models. We are using these multitrophic systems as models for testing theories derived from behavioral, population, and community ecology. Results from these experiments are then used to develop a theoretical framework for biological control of pests and weeds. |
 Interactions between predators reduce their foraging efficiency |
Landscape ecology We also investigate how large-scale, landscape processes affect the composition and interactions in local communities. For example, we ask to which extent species diversity is determined by the local habitat, and what is the contribution of the surrounding landscape. While it has been shown that the landscape can have a tremendous effect on diversity in structurally poor habitats, it is at present unclear if this pattern holds for more diverse habitats. Such knowledge is indispensable for devising effective plans for landscape-level nature conservation. In this context, we also studied the effect of different agricultural farming practices (organic vs. conventional management, different weed and pest control techniques) on diversity and community composition at different trophic levels. Such studies reveal agricultural practices that can favor or prevent natural biological control. |
 Biodiversity on agricultural land can still be high, if the surrounding landscape is species-rich. |
Biological invasions Historical biological invasions are a cornucopia of “natural experiments” that provide unique opportunities for studying conditions under which species escape natural control. The theoretical framework of invasion biology is also highly relevant for biological control, but is as yet a largely untapped resource. In the past, several attempts to characterize traits of successful biocontrol agents failed due to inappropriate data and methods. We have compiled a database of worldwide species introductions (mammals, birds, insects) where we have information on the number of individuals released and the fate of the population, including unsuccessful introductions. This unique dataset enables us to answer questions about species traits that favor their invasiveness (e.g. behavioral flexibility, feeding niche), habitat characteristics responsible for their invasibility (e.g. biotic resistance), and their interactions (e.g. climate matching). Such studies may also shed light on the evolution of life-history traits in certain taxa (e.g. brain size in vertebrates). |
 Why are some parts of the world more affected by alien species than others? |
Biological control We also work on biocontrol solutions for specific weed and pest problems. For example, research in my group discovered that an insect vector is needed to transmit a rust fungus to creeping thistle (Cirsium arvense). The rust fungus kills infected plants before flowering and therefore is regarded as highly promising biocontrol agent. Currently we are investigating with molecular methods why an insect vector is needed to transmit the pathogen and if an application method can be developed to bypass the insect. The theoretical framework of invasion biology is highly relevant for biological control, and will become more so in the future. In order to minimize the risk associated with releases of exotic organisms and to enhance their effectiveness it is crucial to know which of the exotic enemies will be the most promising candidate and what are the pests most likely to be successfully controlled. We approach such questions by analysing large numbers of biological control programmes. |
 Can a rust fungus help to control one of the world's worst weeds, the creeping thistle (Cirsium arvense)? |