Abstrakt
1. Estimation of responses of organisms to their environment using experimental manipulations, and comparison of such responses across sets of species, is one of the primary tools in ecology research.
The most common approach is to compare the response of a single life stage of species to an environmental factor and use this information to draw conclusions about population dynamics of these species. Such approach ignores the fact that interspecific fitness differences measured at a single life stage are not directly comparable and cannot be extrapolated to lifetime fitness of individuals and thus species' population dynamics.
Comparison of one life stage only while omitting demographic information can strongly bias conclusions, both in experimental studies with a few species and in large comparative studies. 2. We illustrate the effect of this omission using both an exaggerated fictitious example and biological data on congeneric species differing in their demography.
We are showing, taking simple assumptions, that different demography can completely revert conclusions reached by a comparison based on an experiment focusing on a single life stage. 3. We show that a 'demographic correction', namely translating observed effects into differences in outcomes of demographic models, is a solution to this problem.
It requires turning the detected effects from the experiment into changes of transition probabilities of projection matrix models. Although such solution is limited by the low number of species with demographic data available, we believe that existing data (and data likely to be collected in the near future) permit at least approximate handling of this problem.