COS 46-4 - Seed predation reduces the fecundity of a rare plant and leads to population decline

Tuesday, August 7, 2012: 9:00 AM
Portland Blrm 258, Oregon Convention Center
Helen M. Kurkjian, Sydney K. Carothers and Erik S. Jules, Department of Biological Sciences, Humboldt State University, Arcata, CA
Background/Question/Methods

The Lassics lupine (Lupinus constancei) is an extremely rare herb of limited distribution, restricted to a single square mile in the Lassics Geological and Botanical Area of the Six Rivers National Forest.  The continued existence of this species faces numerous challenges, including heavy seed predation by small mammals and encroachment of surrounding plant communities on its preferred habitat.  We used ten years of demographic monitoring data collected by the US Forest Service to build a stage-structured matrix model, project population growth, and estimate the probability of species extinction in the next 50 years. The model included vital rate estimates for each of three main sites, as well as correlations between and within sites and years.  We used a caging treatment and one-way life table response experiments to quantify the effects of seed predation on the population growth rate.  Finally, we used a regression life table response experiment to estimate the proportion of the population at each site that must be caged to avoid population decline.

Results/Conclusions

We found that in the absence of seed predation, the Lassics lupine population growth rate would be quite robust (λ = 1.17), but that without continued human intervention (i.e. caging), the current rate of seed predation is likely to drive the population to extinction (λ = 0.92).  A different proportion of reproductive plants must be caged at each site, in order to maximize the probability of avoiding extirpation.  Our work uses a wide array of population modelling tools that are likely to be useful in the study of many other rare plants.  It also provides an example of a relatively modest dataset from which useful management information was extracted.