© Marcel Burkhardt
Integrated population models
We study the dynamics of populations using novel statistical methods. This allows understanding why population numbers change over time.
A demographic analysis is an established method to study fluctuations in population numbers. Usually, demographic data like capture-recapture data or reproductive data are analysed separately and the estimated demographic rates, like survival or fecundity, are merged afterwards in a population model. In addition, survey data on population numbers are often available, yet cannot be directly used in the population model, leaving thus this information unexploited.
An integrated population model is perfectly suited for this situation, because it integrates all available demographic data into one model. A great advantage of integrated population models is that otherwise inestimable demographic parameters (such as immigration) can be estimated. Moreover, because all demographic processes are included in one model, deeper insights into population dynamics are possible than by traditional modelling approaches. However, integrated population models are not well understood yet.
The goal of this project is to further develop and test integrated population models. In particular we aim to develop and test models that allow the estimation of immigration and of density dependence. The models will be applied in a variety of case studies.
The development of integrated population models and tests of their performance were achieved through simulations conducted within the PhD thesis of Fitsum Abadi at the division of Conservation Biology, University Bern. The work will be continued through international collaborations.
The data for the case studies originate from Switzerland (hoopoe, wryneck, eagle owl, peregrine, red kite), from Germany (little owl, red-backed shrike, common tern), from France (black-legged wittiwake) and from the USA (lesser scaup). In each of these cases studies, different topics have been or will be addressed.
The simulation studies have shown that integrated population models are indeed very useful: they allow the estimation of demographic parameters that are not estimable otherwise and all parameters are estimated with increased precision.
The case studies performed so far showed that immigration often contributes significantly to the dynamics of local populations. If immigration is absent in local populations, most of them decline strongly. The reason is that most locally born young disperse to other populations, and thus recruitment is not sufficient if there is no immigration. The case study on the little owl demonstrated that immigration depended on food availability: if voles were abundant, immigration of owls to the local population was stronger. Another study showed that the small eagle owl population in the Valais only remained constant because of massive immigration which counter-balanced high adult mortality induced by electrocution at power lines.
Immigrants are attracted by local pre-breeders and recruits in a seabird colony.
Strong contribution of immigration to local population regulation: evidence from a migratory passerine.
Estimating the strength of density dependence in the presence of observation errors using integrated population models.
The demographic drivers of local population dynamics in two rare migratory birds.
Integrated population models: a novel analysis framework for deeper insights into population dynamics.
Estimation of immigration rate using integrated population models.
An assessment of integrated population models: bias, accuracy, and violation of the assumption of independence.
Massive immigration balances high anthropogenic mortality in a stable eagle owl population: Lessons for conservation.
Helfen Sie mit, die Vogelwarte-Webseite zu verbessern.
Nehmen Sie an unserer Online-Umfrage teil.
Wir würden uns freuen, wenn Sie sich ein paar Minuten Zeit für die
Beantwortung der Umfrage nehmen könnten. Wir werden Ihr Feedback nutzen, um
unsere Website für Sie zu verbessern.
Herzlichen Dank für Ihre Mithilfe.
Die Teilnahme erfolgt anonym. Es werden keine persönlichen Daten gespeichert.