Hemlock Woolly Adelgid Research in the Southern Appalachians

The Problem

Hemlock woolly adelgid (HWA) is a non-native, invasive pest infesting eastern hemlock (Tsuga canadensis) and Carolina hemlock (Tsuga caroliniana). First reported in the 1950s, HWA has spread to the southern Appalachian region in the early 2000s. Without control, hemlock trees die within 5 to 7 years after infestation. Hemlock trees serve important ecological roles in the southern Appalachians. They are a keystone species in near-stream areas, providing critical habitat for birds and other animals, and shading streams to maintain cool water temperatures required by trout and other aquatic organisms. Hemlock trees are also prized for their visual beauty in both forest and urban settings.

A comprehensive research program is underway to address four key elements of the HWA problem:

  1. Evaluating control strategies to reduce or eliminate the spread of HWA,
  2. Understanding the impacts of hemlock mortality on ecosystem resources,
  3. Developing monitoring techniques that determine and predict HWA spread, and
  4. Developing and evaluating restoration techniques to mitigate the impacts of hemlock mortality on ecosystem resources.

The research is being led by the USDA Forest Service's Southern Research Station and is being coordinated with the Northeastern Research Station, university cooperators throughout the Eastern U.S., and USDA Forest Service, State and Private Forestry's Forest Health Protection through participation in the HWA Research Coordinating Committee and the HWA Steering Committee. Most of the work has been centered at the Coweeta Hydrologic Laboratory in western North Carolina. The 5400 acre watershed has conducted watershed ecosystem research for 75 years and has long-term data on climate, soils, vegetation, and water quality. Research on problem areas 1–3 initiated in 2003 and is nearing completion. Research on problem area 4 is nascent, and began in 2014.

Evaluating Control Strategies

While biological control offers the best hope for long-term management of HWA in forests, chemical control is quite effective on small spatial scales. Application methods include foliar sprays and soil and stem application of systemic insecticides (e.g., imidacloprid, a chloronicotinyl (1-[(6-chloro-3-pyridinyl) methyl]-N-nitro-2-imidazolidinimine) insecticide). The effectiveness of soil and stem systemic imidacloprid applications, however, has varied with injection method, timing, and other tree-specific characteristics. Because, uptake and transport of systemic insecticides occurs via mass flow in the transpiration stream, our research has focused on testing whether treatment efficacy is linked to the volume and velocity of xylem water movement, and thus the concentration of the insecticide in the xylem sap, as well as the timing of delivery. (Ford et al. 2010). Other lines of research have focused on the fate and transport of the chemical insecticides in the soil (Knoepp et al. 2012) and stream (Churchel et al. 2011).

University of Georgia, Virginia Tech, SRS-4505 (Athens), University of North Carolina Asheville

Understanding Impacts

More than a decade of research has been published to determine the impacts of HWA mortality on key ecosystem processes. In 2003, twelve study plots were located in areas with and without evidence of HWA infestation. These plots are being used to experimentally isolate the effects of hemlock loss on nutrient, carbon, and water pools and cycling rates. Measures have included: nitrogen mineralization, soil CO2 flux, litterfall and coarse wood debris inputs, forest floor mass and decomposition, microenvironmental changes, overstory and understory species composition, evapotranspiration and stream habitat and water quality. Our research has shown that the rapid loss of eastern hemlock from southern Appalachian riparian forests (Ford et al. 2012; Elliott and Vose 2011) due to hemlock wooly adelgid invasion has altered carbon (Nuckolls, et al. 2008), water (Ford and Vose 2007; Brantley et al. 2013), and nutrient cycling (Knoepp et al. 2011; Block et al. 2012; Block et al. 2013); and has affected terrestrial (Ball et al. 2008) and aquatic litter decomposition (Kominoski et al. 2009; Kominoski and Pringle 2009; Kominoski et al. 2007), stream temperature and trophic processing (Webster et al. 2011), and shifted stream metabolism more towards heterotrophy compared to autotrophy (Northington et al. 2012).

University of Georgia, Virginia Tech, SRS-4505 (Athens), University of Illinois, University of Minnesota

Monitoring HWA Spread

In 2004, a network of monitoring plots was established to determine rates and patterns of HWA spread. This network utilizes the approximately 800 long-term vegetation plots that were established at Coweeta in the 1930s. The long-term plots span gradients in soil, climate, slope, aspect and elevation comparable to those observed across the southern Appalachian region. A subset of plots have been surveyed every two years since 2004 to determine HWA presence, eastern hemlock crown loss and eastern hemlock mortality.

University of Minnesota


The final phase of the HWA research program began in 2014. Two projects are underway that will examine silvicultural treatments to restore structure and function of southern Appalachian riparian forests. With the loss of eastern hemlock in riparian areas, rhododendron has experienced three times greater growth than expected compared to rhododendron growing in non-hemlock areas (Ford et al. 2012).

Rhododendron is an important species in these forests as it is highly shade tolerant, forms a dense shrub layer that strongly attenuates light incident on the forest floor (Clinton 2003), has little to no herbaceous or seedling (henceforth understory) cover below its canopy (Beckage et al. 2000; Clinton et al. 1994), negatively affects tree seedling densities where they can establish (Lambers and Clark 2003), and decreases nitrogen availability in the soil and litter layer to non-ericaceous species (Wurzburger and Hendrick 2009; Wurzburger and Hendrick 2006). Riparian forests affect stream organisms and in-stream processes (Webster et al. 2012); and loss of riparian eastern hemlock with a resulting community dominated by rhododendron could be expected to affect streams in several ways, particularly stream chemistry (e.g., pH) and temperature having potential to change the most. With climate change, we expect rhododendron to respond more than other species to increased CO2 concentrations in the atmosphere due to characteristics associated with its leaf structure. Thus, removal of this species may be critical to restoring riparian areas affected by eastern hemlock loss. We’ve initiated field experiments with the Nantahala Ranger District at the plot and stream reach scales to remove rhododendron and the rhododendron litter layer and monitor microenvironment, vegetation dynamics, soil physical and chemical changes, stream water quality (sediment and chemistry), stream channel morphology, and biological integrity along 300 m of each stream reach.

Region 8 National Forest System managers and hydrologists; State and Private Forestry’s Forest Health Protection unit; University of Georgia LTER