Harvard Forest >

Harvard Forest Symposium Abstract 2017

  • Title: Soil carbon stocks and fluxes in a hemlock stand infested by the hemlock woolly adelgid
  • Primary Author: Marc-Andre Giasson (Boston University)
  • Additional Authors: Adrien Finzi (Boston University)
  • Abstract:

    The introduction of the hemlock woolly adelgid (HWA) has resulted in widespread hemlock mortality. The impact of tree mortality on carbon fluxes remains poorly constrained. Exactly how much C is lost and how long it would take to recover those C stocks following succession is unknown. The HWA is now widespread at Harvard Forest and the objective of this research is to quantify the impact of hemlock mortality on components of the forest C budget. Here we present data on soil respiration and soil C stocks.

    From April to December 2016, we measured soil respiration (Rsoil) using an automated soil respiration chamber system based on the original design by Jim Tang. Six chambers (#1-6) were deployed near the Hemlock tower in an area visibly affected by the hemlock woolly adelgid. Six more chambers (#7-12) were located downstream from the Bigelow Brook weir where hemlock trees appeared to be healthy. The auto-chambers run continuously and are activated sequentially resulting in 119,133 valid soil CO2 efflux measurements over the season.

    Rsoil was significantly higher at the weir (healthy) site than the tower (infested) site (Fig. 1). It also varied greatly between chambers (Fig. 2). For example, at the infested site Rsoil was higher in microsites where black birch seedlings or ground cover vegetation was present. At the weir site topographical position influenced fluxes with downslope chamber locations (n=2) having 40% higher fluxes than the upslope chambers (n=4).

    Changes in soil organic C (SOC) content and root biomass are notoriously difficult to measure because of great spatial variability and high labor cost. At present it is still not well understood exactly how many samples would be needed to detect a change in SOC or root biomass if in fact it were occurring. To address this uncertainty for the Harvard Forest we estimated soil carbon stocks and fine root biomass using an expanded setup of “Conant” plots. The plots are 2 x 5m and six subplots were sampled in each of 16 plots located in different areas in the hemlock stand surrounding the eddy-covariance tower site. In each subplot, we collected a 10 cm × 20 cm (plots 1-6) or 10 cm × 10 cm (plots 7-16) organic horizon monolith and sampled the underlying mineral soil to a depth of 30 cm in 10 cm increments. Soil carbon and fine root biomass (diameter < 2mm) were measured on each sample.

    Both soil C (Fig. 3) and fine root biomass (Fig. 4) varied widely between samples. On average there was about 43% more C and 67% more fine roots in the first 30cm of the mineral soil than in the organic horizon (Fig. 5). We used a large number of samples so we could determine how many should be resampled in the future to be able to determine if any difference in soil C content or fine root biomass is statistically significant. We developed look-up tables (available upon request) that can be used to estimate the number of samples needed to reach a given precision level (in g/m2) in soil C and fine root biomass depending on different probability levels (α).

    Continued Rsoil measurements and soil sampling in future years will help us determine temporal trends in soil carbon stocks and fluxes during and after hemlock woolly adelgid infestation.

    Figure captions:
    Fig. 1:
    (a) Mean daily respiration and (b) total soil respiration from 28 May to 5 Dec, 2016 at the tower (infested) and weir (healthy) sites.

    Fig. 2:
    Total soil respiration from 28 May to 5 Dec, 2016 for each chamber.

    Fig. 3:
    (a) Total soil organic C content from the soil surface to 30cm depth in the mineral soil for each subplot and (b) running average of 100 randomized repetitions of the samples presented in (a).

    Fig. 4:
    (a) Total fine root biomass from the soil surface to 30cm depth in the mineral soil for each subplot and (b) running average of 100 randomized repetitions of the samples presented in (a).

    Fig. 5:
    (a) Mean soil C content and (b) root biomass in the organic horizon and mineral soil.

  • Research Category: Forest-Atmosphere Exchange; Invasive Plants, Pests & Pathogens; Soil Carbon and Nitrogen Dynamics

  • Figures:
  • Giasson_Finzi_Fig1.JPG