Modeling responses of American ginseng (Panax quinquefolius L.) populations to different levels of simulated harvest

Background/Question/Methods

Wild harvested American ginseng (Panax quinquefoliusL.) is a medicinal herb traded internationally, and North Carolina is one of the United States’ leading exporters of wild American ginseng. States regulate ginseng harvest and must demonstrate that harvests are not detrimental to the long-term viability of populations. Studies have shown that harvest is not sufficiently regulated in many states. We examined the demographic characteristics of five ginseng populations in Western North Carolina to determine if state regulations adequately protected local populations. We conducted an annual census for two years, tracking each plant individually and recording the size class based on leaf number and reproductive status. In the second year, we recorded fruit production of each plant and estimated seed production using a published value of average seeds per fruit. Using MATLAB (Mathworks; Natick, MA), we constructed a population matrix model using these data, and estimated growth rate and stable size distribution for each population. 

Results/Conclusions

Estimations of growth rate for all five populations varied substantially, with the finite rate of increase (λ) ranging from 0.9 to 1.2. Three of the five populations were declining (λ<1) in the absence of any harvest, so for these we projected a time to extinction of ten to fifteen years. We also projected the effects of increasing harvest rates on growth rates of populations with λ>1. These two populations could sustain harvest rates of 20 and 70 percent annually while maintaining a stable population size.  Stable size distributions of the five classes indicated that four of the five populations did not include any of the largest class of plants and that the two largest, reproductive classes comprised less than ten percent of all plants in any of the populations. These results indicate that even geographically close populations may have very different population dynamics, that North Carolina populations cannot be considered to have identical demographic parameters, and that populations should be regulated and monitored individually. Longer-term data collection will allow us to describe the demographic properties of local populations more accurately and, in partnership with land managers, use these data to inform effective harvest regulations.