PARENT SESSION
Contributed Oral Session 122: Invasive Species: Pollination - Dispersal
Thursday, August 11, 8:00 AM - 11:30 AM, Meeting Room 521 A, Level 5, Palais des congrès de Montréal
Dispersal measurement and modeling for invasive species: Sampling affects estimates of spatial spread.
Skarpaas, Olav*,1, 2, Shea, Katriona1, 1 The Pennsylvania State University, State College, PA2 University of Oslo, Oslo, Norway
ABSTRACT- Theoretical models predict very different rates and patterns of species invasions depending on the dispersal kernel, i.e. the distribution of dispersal distances of propagules. This suggests that empirical studies need to obtain accurate measurements of dispersal kernels, in particular their tails, for predictions of species invasions. We investigated the effect of alternative seed trapping designs on estimates of dispersal kernels and spatial spread in empirical and simulation studies of two invasive thistles in the US; Carduus nutans and C. acanthoides. Using models, we found that increased sampling effort at the tails increased the chances of recovering the shape of the dispersal kernel even when only a few seeds were trapped. For parametric dispersal kernels (probability density functions) a low effort at the tail implied lower precision in estimates of dispersal parameters and rates of spread. For empirical dispersal kernels (empirical distribution functions truncated at the observed maximum) low sampling effort at the tail tended to lead to underprediction of spatial spread rates.
Key words: dispersal, sampling, spatial spread, invasive species