The effects of climate variation on vertebrate population dynamics: a comparative approach

The aim with this project is to explore the interplay between vertebrate population dynamics and climatic conditions. The programme will quantitatively examine how an expected change in climate will affect population processes in space and time, using both terrestrial and marine examples.

Such information will be crucial when designing monitor-schemes or relating trends in population fluctuations to expected changes in climate.

Questions such as: whether the effects of climate are stronger on population dynamics in highly productive compared to poor environments; whether a stronger climate influence is found in species with a large clutch size and high specific growth rate compared to species with low reproductive rates and; whether the effects of climate on the spatial scaling of the synchrony in population fluctuations are dependent on interspecific differences in migration ability will be explored.

Despite the recent advances in our understanding of the impacts of climate factors on population fluctuations, few general hypotheses have appeared from the huge number of more or less phenological studies, perhaps because (1) The results from statistical analysis of time series have been difficult to interpret. For instance, the autoregessive coefficients from general stochastic age-structured models do not necessarily reveal the strength of density dependence because they are strongly confounded by life history variation. (2) Analysis of population fluctuations over time often depend on assumptions that may be difficult to justify biologically. For instance, the assumption of log-linearity is often not valid and may give biased population dynamics characteristics. (3) No demographic stochasticity is assumed to be present. (4) Sampling errors in population estimates are often neglected, even though they can be of considerable magnitude. Such sampling errors in population estimates will bias estimates of many population parameters and in particular overestimate the strength of density dependence. The two latter points are especially problematic when trying to make accurate estimates and model stochastic effects from traditional time series analysis.

The PIs of this project are Professors Bernt-Erik Sæther and Steinar Engen at the Norwegian University of Science and Technology.

Collaborating partners are NPI, the Institute of Marine Research (IMR), the Norwegian Institute if Nature Research (NINA), the CNRS Lyon, CNRS Chizé, and the Université Pierre et marie Curié, Paris.

This project is financially assisted by the Norwegian Research Council. The project period is 2003 – 2007.

Prosjektleder: Dr. Ronny Aanes