Skip to main content
SLU publication database (SLUpub)
Doctoral thesis, 2012

Population viability analysis under environmental change : development of Bayesian tools

Ruete, Alejandro

Abstract

An understanding of the links between population dynamics and environmental variability, combined with information on how these factors change over time, is necessary to understand and predict population dynamics and viability in changing environments. Scientists need also to acknowledge uncertainties in their understanding of systems, which is straightforward using Bayesian statistics. This allows us to know with more certainty, although it sounds contradictory, how a biological system works. First, the hierarchical model developed in paper I illustrates how conclusions and decisions to be made based on population viability analysis could be dangerously misleading if uncertainties are not taken into account. The probabilistic long-term growth rate parameter, log λS, is estimated for the first time, and I discuss a new way to interpret this parameter. Based on simulations done with this model, we stress in paper II that ignoring relevant uncertainty sources generally gives an unwarranted impression of confidence in the results. The procedure used in this work increased our understanding of the relative importance of different uncertainty sources, and helps choosing which sources to include when evaluating the impact of climate change. Second, the modelling approach developed in paper III allows us to estimate colonization rates of non-equilibrium metapopulations. It reconstructs a time series of the most likely colonization events leading to the observed pattern of occupied and non-occupied patches. It requires only snapshot data on the occurrence pattern, as well as data on patch ages and on the landscape history. In this case I stress how the choice of a modelling approach has important implications on metapopulation viability analysis. I finally draw conclusions on the methodological advances achieved, and on the implications for the conservation of the study species. Using Bayesian statistics both process uncertainty, and parameter uncertainty and variability are captured, and predictions are turned into a probabilistic statement that is useful for management. Uncertainties are no longer an obstacle, but a mandatory aspect to include in population viability analysis.

Keywords

climate change; epiphyte; epixylic; fragmentation; habitat loss; hierarchical model; lichen; metapopulation dynamics; moss; population dynamics; PVA; uncertainty

Published in

Acta Universitatis Agriculturae Sueciae
2012, number: 2012:22
ISBN: 978-91-576-7658-0
Publisher: Department of Ecology, Swedish University of Agricultural Sciences

    UKÄ Subject classification

    Ecology
    Climate Research
    Forest Science

    Permanent link to this page (URI)

    https://res.slu.se/id/publ/79033