SEARCH

SEARCH BY CITATION

Long-term conservation of forested areas requires an estimate of nonnative species’ impact on a scale of centuries. However, long-term changes in forest structure have not been included in previous succession studies, which have typically covered only 20–50 yr. To estimate persistence of nonnative plant species on a longer time scale, and to examine the development of spatial structure in their distributions, we selected thirty-seven second-growth forest sites in southeastern Ohio, USA, to form a chronosequence spanning 160 yr. At each site nonnative species were surveyed in 100 contiguous plots forming a 50-m belt transect. Nineteen nonnative species were encountered, largely consisting of shade-intolerant herbs typical of abandoned agricultural land. Nonnative species richness, abundance, and frequency declined through the chronosequence with most species dropping out after the first twenty years, although three persisted 120–140 yr. Most species appeared to be residual populations from the open-habitat stage. Presence of Rosa multiflora and Polygonum hydropiper in stands older than their date of arrival in the region indicate their capacity to colonize long-established forest. Distributions of four species showed dependence on micro-environmental gradients. All four were positively autocorrelated at scales up to 2–8 m, but pattern did not vary among age classes. We conclude that pattern in nonnative species’ distributions arises very early, possibly in the open-habitat stage. Most species encountered in these sites are not likely to become long-term components of the forest community and do not require active management. Small-gap disturbances have not allowed colonization of long-established forest stands, nor released pre-existing populations from competitive suppression. Management should focus on the small number of shade-tolerant species able to persist in long-established forest.