Harvard Forest >

Harvard Forest Symposium Abstract 2005

• Title: Landscape-level effects of forest harvesting on vascular plant composition in Massachusetts
• Primary Author: Michael Bank (Harvard School of Public Health)
• Additional Authors: Robert McDonald (Harvard Forest)
• Abstract:

  The effects of land use and natural disturbance on plant species distributions and community and landscape dynamics are important to understanding the long-term legacies of landscape history on modern forested ecosystems. However, little attention has been focused on evaluating the interactive effects between historical legacies, modern disturbance regimes, and environmental variation on current vegetation composition. Specific objectives of this investigation were to: (1) identify historical and environmental controls on vegetation patterns and species distributions, including invasive plants; (2) compare modern anthropogenic disturbance patterns with natural disturbance regimes, historical land-use activity, and environmental and biotic variation; and (3) provide management guidelines for forest conservation and natural resource planning.
  
  
  For this investigation we randomly selected ~500 (5%) Forest Cutting Plans (FCP’s, >10 ha) from the Massachusetts statewide database using ArcGIS 9.0, stratified by ecoregion and 19th century land-cover. During pilot work in summer 2004 we sampled 82 sites. Sampling proceeded at three spatial scales at each site. First, at the broadest scale, we carefully searched the polygons for invasive plant species and recorded the number of observer-hours to estimate search intensity. Second, within the largest harvested stand in each polygon, stand characteristics, tree basal area and density were measured using the point-quarter-center method (modified to record 2 size classes of trees: 5-20 cm DBH and >20 cm DBH) at 10 sampling points located at 50-100 m intervals along a randomly-oriented transect, along with seedling and sapling density and the number of stumps within an 11.m fixed plot. Finally, quantitative data on vascular plant composition was recorded in an intensive 20 m x 20 m plot within the dominant stand. Each intensive plot was carefully searched for evidence of past disturbance (e.g., plow horizons, stonewalls, wind throw mounds, soil scarification), aspect, slope, and canopy cover, and we categorized each plot as either 1830 wooded or 1830 open habitat.
  
  
  Our pilot field data show that 1830 open habitat had a significantly higher percent observation of non-native plant species in contrast to 1830 woodland sites (Figure 1). These data suggest that historical land use may be more important than modern disturbances such as forest harvesting activities.
  
  
  A set of parallel analyses of the entire geospatial database of forest harvesting plans have been conducted to quantify the harvesting regime throughout Massachusetts (Figure 2). Mean harvesting intensity varies from 28.3 m3 ha-1 in the Taconics to 50.3 m3 ha-1 in the coastal plain, with higher harvesting intensities generally occurring in the eastern portion of the state where forests are already more fragmented and threatened by development. However, harvesting plans occur over a bigger area in western Massachusetts, and there are many more harvesting events in western Massachusetts, with the net effect that the highest total volume removed came from the upper and lower Worcester Plateau (2.8 million m3year-1, or 42% of the state’s total harvest). Somewhere around 0.5% of forests in the coastal plain are logged each year, while 1.4% of forests are logged each year in the Worcester Plateau. Finally, there appears to be little significant temporal variation in the total volume harvested in Massachusetts or in mean harvest intensity, although there has been an increase over time in the amount of softwood cut for cordwood.
  
  
  

• Research Category: