Harvard Forest >

Harvard Forest Symposium Abstract 2006

• Title: The Hemlock Removal Experiment: Initial changes in environmental variables following treatments
• Primary Author: Aaron Ellison (Harvard University)
• Abstract:

  The Hemlock Removal Experiment was established at the Simes Tract in 2003. This experiment consists of eight 90 × 90 m (0.81 ha) experimental plots grouped in two blocks, each consisting of three plots dominated by hemlock and one plot of mixed northern hardwoods. Block 1 (plots 1-3 and 8) is in undulating terrain bordered on its northern edge by a Sphagnum-dominated wetland. Block 2 (plots 4-7) is on a forested ridge. Each block consists of a hemlock control (unmanipulated) plot, a plot in which all hemlock trees were girdled in May 2005 to simulate death-by-adelgid, a plot in which all hemlocks >20 cm DBH and any merchantable hardwoods and white pine (Pinus strobus) were logged and mechanically removed from the site in February-April 2005, and a hardwood control plot. Dataloggers were installed in November 2004 to record air temperature (1 m above ground) and soil temperature (10 cm below ground) in each of the 3 hemlock plots in both blocks. Data are collected at 10-minute intervals and 60-minute averages, minima, and maxima are stored. Data are posted annually to the Harvard Forest Data Archive. Hemispherical canopy photographs are used to measure the light environment in each plot. Photographs are taken twice yearly (leaf-off in April/May; leaf-on in September) at 15-m intervals within each plot (25 photographs/plot). Photographs are analyzed using HemiView software. Film negatives and digital scans are stored in the Harvard Forest Archives.
  
  
  
  
  
  In the first year, temperature responses to treatments were apparent only in the logged plots (Figs. 1 and 2). Average soil temperatures in the logged plots were higher in the summer and lower in the winter than soil temperatures in the girdled or control hemlock plots (Fig. 1). Average air temperatures were comparable in all three treatments, but the diurnal variance in temperature is higher in the logged plots than the variance in the girdled or control hemlock plots (Fig. 2).
  
  
  
  
  
  Prior to treatments, light levels (expressed as global site factor, the percent of total direct and indirect radiation reaching the forest floor) in all hemlock plots was ~ 10% of full sun (Fig. 3). In the first year after treatment, no changes in light levels were observed in the girdled plots (Fig. 3). Logging increased light levels nearly 6-fold; the amount of light reaching the forest floor in the logged plots was comparable to that found in the hardwood plots in the winter (Fig. 3).
  
  
  
  
  
  One year after treatments, environmental changes in the experimental plots are already apparent. Data collection and analysis will continue, and data will be made available as it is collected for use in association with other studies ongoing at the Simes Tract.
  
  
  

• Research Category: Large Experiments and Permanent Plot Studies