Impacts of Hemlock Woolly Adelgid at the Arnold Arboretum

• Investigators: Robert Cook, Peter Del Tredici, David Foster, Heidi Lux, David Orwig, Richard Schulhof
• Contact: David Orwig
• Start date: 2004-04-26
• End date: ongoing
• Location: Arnold Arboretun (Jamaica Plain MA)
• Latitude: +42.29754
• Longitude: -71.12413
• Elevation: 52 meters
• Taxa: Adelges tsugae (hemlock woolly adelgid), Betula lenta (black birch), Tsuga canadensis (eastern hemlock)
• Keywords: logging, pest infestation, successional dynamics, tree mortality, Tsuga canadensis, urban infestation
• Abstract:

  Hemlock Hill at Harvard University's Arnold Arboretum is located in Jamaica Plain and has long been considered a remnant of the forest primeval in the heart of Boston. Eastern hemlocks have dominated the flanks and crest of this hill for centuries, despite experiencing many past disturbances. Following the 1997 snowstorm, the introduced hemlock woolly adelgid (HWA) was discovered on downed branches. Since the discovery of HWA at the Arboretum, an action committee was formed to develop plans for managing Hemlock Hill to meet the new threat. Over 1800 hemlock trees were tagged, measured for diameter, assigned a crown health rating, and mapped with GPS coordinates. Various chemical and biological control treatments were attempted to control the spread of HWA at the arboretum, with mixed success. From 1998 to 2002, 263 trees have died or were removed due to poor health. Of the remaining 1600+ trees, 70% are rated as being in poor condition. Since this is a heavily used portion of the Arboretum, which is part of the Boston Parks Department, the decision has been made to remove many of the dead and dying trees to reduce risks posed to the general public by falling limbs. This unfortunate turn of events provides us with an unusual opportunity to examine the environmental impacts of hemlock death and hemlock removal in an urban environment, including soil nutrient cycling, microclimate changes, and vegetation succession, especially the spread of invasive species. During 2004, six 15 x 15 m plots were fenced off, and baseline data on soil nutrient cycling, microclimate, and vegetation information was collected from each. Analyses of these baseline data are underway. Hemlocks are currently being removed from 4 plots, and 2 will remain untouched as control plots. Slash in the cut plots is either being chipped and left on site or removed so that we can investigate what impact these post-cut inputs have on ecosystem and vegetation trajectories. We will continue with data collection on vegetation and ecosystem parameters following harvesting, especially the spread of invasive plant species. Data from this project will provide a nice comparison to the rural cutting and girdling that is ongoing at HF this winter (see Barker-Plotkin et al. abstract). Simultaneously, we are investigating the resistance of Chinese hemlock (Tsuga chinensis) to HWA. This species, which was planted on Hemlock Hill in 1999, continues to demonstrate resistance to HWA and plantings have begun on some sections of the hill.

• Methods:

  Nitrogen mineralization was measured using 4 closed-topped cores at each of six 15 x 15 m plots. Sampling occurred at 5-week intervals during each growing season with one set of cores left over winter. An additional assessment of forest floor N availability and mobility was determined from mixed-bed cation + anion resin bags that were buried for 6-month intervals at the organic layer-mineral layer interface (ca. 5 cm) in each sampling plot. Soil temperature at 1 and 5 cm depths were measured at each soil sampling plot every 30 minutes throughout the growing season using iButton temperature data loggers.

  Soil samples were returned to the laboratory on ice and processed the next day. Organic and mineral soils were passed through a 5.0 mm mesh screen, weighed for total mass, and subsampled for gravimetric moisture and inorganic N. NH4-N and NO3-N were extracted from all soils using 1M KCl and from resin bags using 2M KCl and concentrations were determined colorimetrically with a Lachat flow-injection 8500 autoanalyzer. We determined the following properties of soils collected during our initial sampling at each site: organic matter by loss on ignition (5.5 hrs at 550 deg C), pH in a soil and CaCl2 slurry (1:10 organic soil:solution; 1:4 mineral soil:solution), and total C and N content by dry combustion with our Fisons CHN autoanalyzer. Soil texture texture was determined on mineral soils in each subplot. Bulk density of organic and mineral horizons was determined from dried samples from each subplot using a soil sampling corer of known volume. Remaining organic and mineral soils were air-dried and stored in our sample archive.

  Vegetation was sampled in six 15 x 15 m plots systematically located across the northern slope of Hemlock Hill. Prior to cutting, all trees (stems at least 8 cm dbh) were tagged, measured for diameter, assigned a crown health rating, and mapped with GPS coordinates. The presence of all vascular plant species in each plot were recorded, along with slope, aspect, and topographic position, and saplings (less than 8 cm dbh and more than 1.4 m tall) were tallied by species. Ten 1-m2 subplots were established at randomly selected subplot (5 x 5 m) corners set up for soils analyses (see below) to assess understory vegetation. In each subplot, percent cover of seedling, herb, and shrub species were estimated using a Braun-Blanquet scale.

• Related datasets: HF054 HF083