In many organisms dispersal is restricted to specific life history stages, thus the stage distributions of founder populations are not stable and we expect significant transient deviations from long-term dynamics. The transient population growth rate following a dispersal event is a critical component for the establishment success because propagules typically arrive in new patches in small numbers and small populations inherently have a high extinction risk as a result of Allee effects or demographic stochasticity. The population growth rate after arrival determines how long populations remain in the state of “dangerously low numbers”. A high colonization success is especially important for organisms exploiting ephemeral resources, such as pioneer plant species. In this paper we use density dependent integral projection models on the endangered pioneer species, Penstemon haydenii to explore the role of transient dynamics in population establishment. We parameterized the models with data collected by Kottas (PhD thesis, 2008) on size dependent growth, recruitment, and survival of penstemon in eleven spatially separated populations in the Sandhills of Nebraska in 2005-2007. We used Monte Carlo analysis and Partial Rank Correlation Coefficients to evaluate the relative importance of model parameters for transient and asymptotic dynamics.
Results/Conclusions There is large spatial variation in the predicted asymptotic population size (N*); the observed population density in the eleven blowout populations falls within the 95% bootstrapped confidence intervals of N*. Our modes predict that transient dynamics can cause colonizing (seedling) populations to crash to very low numbers (transient dip) and it may take > 40 years to reach numbers that are close to N*. In contrast, if the colonizing plants are the size of reproducing adults (a possible option in habitat restoration projects) the model predicts transient amplification, where population size temporarily exceeds N*. Our Monte Carlo analysis suggests that survival is by far the most important process influencing the magnitude of the transient dip as well as N* for P. haydenii. Moreover, our analysis implies that model parameter values resulting in a low N* also cause a large transient dip.
