Enhancing rare plant population predictions through demographic modeling of seed predation, dispersal, and habitat suitability

Understanding the effects of seed predation, dispersal, and recruitment on the population dynamics of rare plant species is essential for generating effective management strategies. Unfortunately for most rare plants, the parameterization of these processes is limited and generally not included in demographic analyses. This exclusion can lead to biased estimates of vital rates and overall population growth rates, as well as limit inferences about inter-population processes like colonization and demographic rescue that can affect population viability. Based on previous empirical studies from Fort Liberty (formerly Fort Bragg) North Carolina (USA), we constructed a spatially explicit demographic model that accounts for pre-dispersal seed predation, dispersal, and habitat suitability for Lindera subcoriacea (bog spicebush), a rare shrub in the southeastern United States. We demographically modeled three scenarios: S1 did not include any of the three parameters; S2 accounted for seed predation and dispersal; and S3 included all three of the parameters. Results suggested that pre-dispersal seed predation, dispersal, and habitat suitability negatively impact the population growth rates of bog spicebush relative to the naïve demographic model. After 100 annual time steps, scenarios S1, S2, and S3 led to a 96%, 49%, and 1% increase in population size, respectively. In addition, over the course of 100 years, results of scenarios S2 and S3 demonstrated limited increases in site occupancy, with newly occupied areas located < 1 km from previously occupied habitat. Our results suggest additional parameterization of plant demographic models may be an informative endeavor and warranted, even in the absence of empirical data.