Demographic heterogeneity among populations, that is, differences among populations in survival rates and fecundities, can result in differences among populations in population size, structure, dynamics, and persistence. Demographic heterogeneity among populations can also affect the dynamics of invasions. Centaurea stoebe (spotted knapweed; formerly C. maculosa; Asteraceae) is a non-native perennial forb that appears to be transitioning from naturalized to invasive status in New York State. We collected demographic data from five populations of marked individuals of C. stoebe on Long Island in 2012 and 2013. Plant size, reproductive status, and number of seedheads per plant were recorded. We also calculated seedbank dynamics and seeds per seedhead. We constructed an integral projection model (IPM) for each population separately; for example, each population could have a different relationship between plant size and plant survival. We calculated the population rate of increase (λ) and stable size distribution of each population and compared populations. We also calculated the elasticity of each demographic parameter and compared these among populations.
Results/Conclusions
Demographic parameters, particularly growth rates and fecundity-related parameters, differed significantly among the five populations. Because of these differences in demographic parameters, λ also differed widely among populations, taking the values 1.06, 3.19, 3.19, 12.58, and 25.19. Elasticities differed among demographic parameters and among populations. For example, in the population with the smallest λ, all elasticity values were small: no single demographic parameter has a disproportionate effect on λ. In two populations, λ was particularly sensitive to the size distribution of the plants that germinated during the winter and spring between the summer censuses; these were also the most rapidly growing populations. These two populations also had predicted stable size distributions with large proportions of large plants. All populations were sensitive to the probability of a seed germinating and to seeds/reproductive plant, especially populations with larger λs. The large differences in demographic parameters and λs among populations suggest that this invasion was not occurring uniformly, perhaps because these populations represented very different stages of the invasion.