Background/Question/Methods Understanding how ecological communities change over time is critical for biodiversity conservation. However, there are few long-term studies that directly address decadal-scale changes in the ecological communities of protected areas. In this study, we take advantage of a network of permanent forest plots established in Great Smoky Mountains National Park in 1978 to investigate temporal changes in plant communities with different disturbance histories and environmental conditions. In 2007, we revisited 15 permanent plots that were logged in the late 1920s, prior to the Park establishment, and 15 permanent plots that have no documented history of human disturbance. In addition to differences in disturbance history, these plots varied in elevation and a variety of soil nutrient parameters. We measured plant species presence and abundance in both the understory (herbs, shrubs, and trees <1 m tall) and overstory (trees ≥12 cm dbh) communities in these plots. Results/Conclusions We found that understory species richness decreased by an average of 4.3 species over 30 years in the logged plots, while understory richness remained unchanged in the unlogged plots. Overstory richness did not change during the study period, but overstory density decreased by an average of 219 stems/ha in the logged plots and 68 stems/ha in the unlogged plots. Historic logging had no effect on within-site understory or overstory compositional turnover. However, environmental site factors were correlated with temporal turnover. Sites at lower elevations had higher understory turnover than sites at higher elevations, and sites with higher soil pH had higher understory turnover than sites with lower soil pH. In addition, sites with lower initial understory productivity had larger temporal changes in understory composition than sites with higher understory productivity. High overstory compositional turnover was correlated with low soil cation exchange capacity and low initial overstory productivity, but was not related to elevation. Taken together, our results indicate that human disturbance can affect plant diversity for decades following the disturbance event, but change in community composition over time may depend more on environmental conditions than disturbance history.