Wild leek (Allium tricoccum) populations in the Southern Appalachians: 1999-2009

Background/Question/Methods Allium tricoccum, or wild leek, is a culturally important herb harvested by digging in the early spring. Population declines attributed to over-harvesting in its northern range along with anecdotal reports from harvesters in the southern range raised concern about the status of A. tricoccum in the southeastern U.S. In addition to the uncertainties of harvest, the southern Appalachian ecosystems are likely to be affected by changing temperature and rainfall predicted by climate models. Wild leek association with north and east aspects suggests that this species, along with other ephemeral spring wildflowers, may be extraordinarily susceptible to climate change. In 1999, we randomly selected 21 relatively high-density plots (>20 ramets/m²) in the upper Nantahala River watershed. Annually, we quantified leek density and cover, and soil disturbance along permanent transects. We address the following questions: Does abundance differ among plots and through time? Does disturbance affect mean density and temporal variability? Repeated measures analysis of variance (ANOVA) and linear models were used to test for population trends through time. Because there was a significant sample plot*year interaction (p<.0001), we used separate linear models for each plot to test for significant population trends. To investigate the effects of disturbance, we  classified plots as low or high disturbance and then used ANOVA to test for disturbance and year effects. 

Results/Conclusions Among environmental variables, only aspect was significant (p=0.004; r² = 0.3662), with greatest leek density on east through north aspects. All populations varied from year to year, but only six showed slopes significantly different from zero, five negative and one positive.  Negative trending plots all showed high soil disturbance, and the single increasing population was a high elevation undisturbed plot. Main effects of disturbance class (p=.0018) and time (p<.001) were significant.  Disturbed plots had lower densities in every year, but most appeared to be stable at the current disturbance levels.  Additional analysis will be conducted to clarify the confounding effects of disturbance and elevation that could signal climate effects.  The project has assembled a long-term record of wild leek population data that provide a basis for interpreting future dynamics.  Information about wild leek population biology will provide insight into the population dynamics of other spring ephemeral herbs and their patterns of recovery from disturbances that reduce population sizes.