A demographic study of deer browsing impacts on Trillium grandiflorum

When white-tailed deer populations reach high densities, they have negative and often dramatic effects on forest herb populations. However, it is not clear how deer affect the demographic processes of plant populations. We examined how the structure and dynamics of Trillium grandiflorum (Michx.) Salisb. populations are affected by deer browsing in the Upper Great Lakes region by sampling populations from nine study sites in a forested landscape in 1998 and 1999. We constructed a stage-based matrix population model for the regional population. Our model indicated that the long-term growth rate of the population to be - 3.56% per year (lambda = 0.965). Mortality rates were highest for seeds (97.5%) followed by seedlings (29.1%), and lower for all remaining stage classes (4.9 to 8.5%). The observed stage distribution significantly differed from the stable stage distribution, and the damping ratio (rho = 1.103) indicated the population would not reach its stable stage distribution anytime soon. In the absence of deer browsing, the long-term growth rate would improve to between - 3.46% and - 1.61% per year. A moderate drought during the study could account for the negative population growth rate, but deer browsing accelerates the rate of decline. Population growth is most sensitive to the proportion of plants remaining in the nonflowering stage, and deer browsing reduces this proportion. Browsing damage was relatively low in this study (5.4% of stems in 1998, 11.5% in 1999) compared to another study of browsing impacts on T. grandiflorum, indicating deer could have far more severe demographic consequences in populations subject to higher levels of browsing.