The depletion of algal beds by geese: a predictive model and test

The short-term impact of herbivores on plant productivity has been well stu died. Demonstrating longterm effects of grazing on plant populations is muc h more difficult, but knowledge of such long-term effects is vital in under standing plant-herbivore interactions. We address this issue in a relativel y simple plant herbivore system involving the dark-bellied brent goose Bran ta bernicla bernicla and two marine macroalgae, Enteromorpha spp. and Ulva lactuca, on which the geese graze. In 3 years of sampling, goose grazing wa s responsible for depleting between 23% and 60% of the algae in the autumn, while wave damage caused between 15% and 70% depletion. The degree of depl etion in autumn had no effect on the biomass of algae present the following summer, suggesting no long-term consequences of grazing for the population dynamics of the algae. A model simulating the change in mean algal biomass over the autumn and winter, incorporating changes due to depletion by gees e, wave action and productivity, success fully described the date at which geese abandoned the algal bed in six different years. These years varied in numbers of geese, strength and timing of storms and initial biomass availa ble. The most important factor determining the date of abandonment of the a lgal bed was a tradeoff between the timing of storms and the numbers of gee se. When storms occur early, most depletion is due to wave action and geese must abandon the bed early, regardless of the numbers grazing there. As th e algae become depleted, the geese switch to feeding on saltmarsh, pastures or arable crops. The rate of depletion of algae affects the tinning of thi s movement, and the model presented here enables the switch to be predicted .