Abstract
Loss of foundation tree species rapidly alters ecological processes in forested ecosystems. Tsuga canadensis, an hypothesized foundation species of eastern North American forests, is declining throughout much of its range due to infestation by the nonnative insect Adelges tsugae and by removal through pre-emptive salvage logging. In replicate 0.81-ha plots, T. canadensis was cut and removed, or killed in place by girdling to simulate adelgid damage. Control plots included undisturbed hemlock and mid-successional hardwood stands that represent expected forest composition in 50–100 years. Vegetation richness, understory vegetation cover, soil carbon flux, and nitrogen cycling were measured for two years prior to, and five years following, application of experimental treatments. Litterfall and coarse woody debris (CWD), including snags, stumps, and fallen logs and branches, have been measured since treatments were applied. Overstory basal area was reduced 60%–70% in girdled and logged plots. Mean cover and richness did not change in hardwood or hemlock control plots but increased rapidly in girdled and logged plots. Following logging, litterfall immediately decreased then slowly increased, whereas in girdled plots, there was a short pulse of hemlock litterfall as trees died. CWD volume remained relatively constant throughout but was 3–4× higher in logged plots. Logging and girdling resulted in small, short-term changes in ecosystem dynamics due to rapid regrowth of vegetation but in general, interannual variability exceeded differences among treatments. Soil carbon flux in girdled plots showed the strongest response: 35% lower than controls after three years and slowly increasing thereafter. Ammonium availability increased immediately after logging and two years after girdling, due to increased light and soil temperatures and nutrient pulses from leaf-fall and reduced uptake following tree death. The results from this study illuminate ecological processes underlying patterns observed consistently in region-wide studies of adelgid-infested hemlock stands. Mechanisms of T. canadensis loss determine rates, magnitudes, and trajectories of ecological changes in hemlock forests. Logging causes abrupt, large changes in vegetation structure whereas girdling (and by inference, A. tsugae) causes sustained, smaller changes. Ecosystem processes depend more on vegetation cover per se than on species composition. We conclude that the loss of this late-successional foundation species will have long-lasting impacts on forest structure but subtle impacts on ecosystem function.
Highlights
Changes in the distribution and abundance of canopy trees have system-wide impacts on ecological processes in forests (Lovett et al, 2006; Wardle et al, 2011; Hicke et al, 2012)
We examine and test three predictions that, relative to both hemlock and hardwood controls: (1) Vegetation structure – species richness and cover of understory herbs, and density and cover of tree seedlings and saplings – increases slowly following girdling but more rapidly following hemlock removal and soil scarification from logging; (2) Volume of standing dead wood and snags is highest in girdled plots, but downed coarse woody debris is higher in logged plots; (3) Core ecosystem functions – litterfall and soil carbon fluxes decline while rates of soil nitrogen mineralization and soil nitrogen availability increase decline slowly in girdled plots but rapidly in logged plots
T. canadensis was affected in the girdled treatment, but girdled individuals ranged from seedlings to canopy trees and they died within 2 years
Summary
Changes in the distribution and abundance of canopy trees have system-wide impacts on ecological processes in forests (Lovett et al, 2006; Wardle et al, 2011; Hicke et al, 2012). The vast majority of studies of the impacts of species loss on ecological processes in forests and other ecosystems have examined how changes in the absolute number (or percent) of species lost affects a wide range of ecosystem services (recently reviewed by Wardle et al, 2011; Cardinale et al, 2012; Hooper et al, 2012; Naeem, Duffy & Zavaleta, 2012). Moreau), respectively (McBride, 1973; Houston, 1975; Barnes, 1976; Huenneke, 1983; Twery & Patterson, 1984; Myers, Walck & Blum, 2004; Lovett et al, 2006); changes in faunal (Wills, 1993) and macrofungal diversity (Anderson et al, 2010), and functional diversity of soil bacteria involved in carbon and nitrogen cycling (Cai et al, 2010) following loss of Eucalyptus to Phytophthora outbreaks in Australia; bottom-up control by Populus
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
