Changes in vegetation structure and composition of urban and rural forest patches in Baltimore from 1998 to 2015

Abstract

Urban forests often occur as highly fragmented patches with many non-native plant species, altered disturbance regimes, environmental pollutants, and uncertain trajectories of plant community composition. In 1998, the Baltimore Ecosystem Study, a U.S. National Science Foundation-funded Long-Term Ecological Research (LTER) program, established eight forested plots to investigate long-term impacts of urbanization on natural ecosystems. All plots were located in the Baltimore metropolitan area with four in urban forest patches and four in a rural forest. In 1998, these forest patches had closed canopies with well-developed tree, shrub and vine layers, and extensive herbaceous cover. The 1998 alpha diversity (species richness) was higher in urban plots than in rural plots, whereas both plot types had similar forest structure and a relatively small number of non-native species. In 2015, we resampled these plots to investigate changes in plant structure, composition, and diversity based on the abundance, cover, and size of plant species. Trees and vines experienced minimal structural changes in all the plots. Sapling, shrub, and herbaceous abundances all declined over time in the rural plots. In the urban plots, however, only the sapling and herbaceous layers experienced declines. Despite having fewer structural changes, urban plots showed a greater shift in species composition than did rural plots. As in 1998, alpha diversity was lower in the rural plots. Beta diversity (community dissimilarity) decreased among rural plots but remained nearly unchanged in the urban plots, whereas beta turnover (species turnover) was much higher in the urban plots. These data suggest that the urban plots may have divergent compositional trajectories from the rural plots, which may help urban forests retain structural similarities through functional redundancy.