Thursday, August 7, 2008 - 8:40 AM

COS 88-3: Exploring the spread of a bird-dispersed invasive plant using simulation modeling

Nancy LaFleur, Cory Merow, Margaret Rubega, and John A. Silander Jr. University of Connecticut

Background/Question/Methods

The spread of invasive plants threatens natural resources and environments, but in many cases, the mechanics of invasive species spread are not well understood. Understanding patterns of spread is particularly important for invasive plants with seeds that are moved by birds. Seeds of bird-dispersed plants have the potential to be moved long distances as the result of bird mobility, and may be distributed throughout the landscape in a non-random manner with respect to environmental factors as the result of bird behavior. Bird habitat preferences may result in seeds being delivered to germination sites where the biotic and abiotic conditions regulating germination are either enhanced or limited. Here, we develop cellular automata models simulating the spread of two fleshy-fruited invasive plants, autumn olive (Elaeagnus umbellata) and Oriental bittersweet (Celastrus orbiculatus) by an invasive bird, the European starling (Sturnus vulgaris). This modeling technique allows us to begin to understand how variation in habitat affects the ability of these bird-dispersed seeds to establish, by considering both the responses of seed-dispersing birds and the seeds they spread to variation in habitat. We used land-use-land-cover classes as a basis for our landscape, and examined the effects of a variety of seed dispersal distances and habitat preferences on plant spread. Plant preferences for habitat are estimated from survival and germination rates collected in the field. To determine the contribution of each of these factors to invasive plant spread, we compared this model to those in which 1) plant survival was equal across habitats, 2) all habitats were equally abundant in the landscape, and 3) birds traveled equally to all habitats.

Results/Conclusions

Preliminary results of our model suggest that variance in seed dispersal distances may be associated with increasing rates of spread. Additionally, plants spread equally quickly in landscapes where bird visits were in proportion to favorable habitat, or where birds more often visited habitat favored by plants. Where birds avoided favorable habitat, spread was slowed.