In the fragmented landscape of suburban central New Jersey, forests experience variable levels of deer pressure due to their landscape contexts and histories of deer management. Forests with lower deer pressure have more understory vegetation, but it includes abundant populations of invasives, including thorn-bearing plants. The invasive Asian species Rosa multiflora occurs at very high abundance in these forests relative to forests with lower deer pressure, while the increased abundance of native Rubus allegheniensis is not as notable. We hypothesized that deer browse may be a cause for this pattern, in two ways: 1) if deer browse is greater on R. multiflora, reduced browse may result in relatively greater increase in abundance; and 2) if R. multiflora has a more pronounced induced thorn defense strategy, it may be able to allocate more resources to growth under reduced browse conditions. To test these hypotheses, we measured both species in three forest preserves with lower and three with higher deer pressure, in fenced and unfenced 16 m2 plots (224 plots). We quantified deer browse presence/absence and abundance of both species over six years, and after four years measured their thorn:stem length ratio (including plants outside of plots) and herb layer cover.
Results/Conclusions
Herb layer cover and abundance of R. multiflora (ROMU) was less than R. allegheniensis (RUAL) in forests with higher deer pressure, but was much greater than RUAL in forests with lower deer pressure (species*deer F1,440=10.02, P<0.002). Their browse rates were similar in higher pressure forests (ROMU,13% of 64 plants observed; RUAL, 15% of 180 plants), but lower only for RUAL in lower pressure forests (ROMU,11% of 1988 plants; RUAL, 4% of 1297 plants; χ2=48.42, P<0.0001). Unfenced plants of both species exhibited greater thorn:stem length ratio in higher pressure forests (F1,258=28.8, P<0.0001). RUAL had lower thorn:stem ratio inside fenced plots in higher deer pressure forests (F1,206=9.37, P<0.01; too few ROMU plants for analysis). In the lower pressure forests, four years of fencing had no significant effect on the ratio of either species. That both species allocated more to growth vs. defense in the lower deer pressure forests, but only RUAL experienced significantly lower browse, suggests that even slightly lower deer browse over many years (more than the four years of fencing) may permit ROMU to reallocate to growth and be more competitive. This may partially explain why this invasive plant occurs at higher abundance in suburban forests under lower deer pressure.