You are here

Harvard Forest >

Harvard Forest Symposium Abstract 2008

  • Title: Interannual, seasonal, and diel variation in soil respiration along a wetland to upland slope
  • Primary Author: Steve Phillips (University of New Hampshire - Main Campus)
  • Additional Authors: Jill Bubier (Mount Holyoke College); Patrick Crill (University of New Hampshire - Main Campus); Steve Frolking (University of New Hampshire - Main Campus); J. William Munger (Harvard University); Ruth Varner (University of New Hampshire - Main Campus); Steven Wofsy (Harvard University)
  • Abstract:

    Soil carbon dioxide flux (soil respiration or SR) was measured using automatic chambers at Harvard Forest over a five-year period, 2003-2007. Eight autochambers were placed along a slope from upland to wetland soils. Soil profiles of soil moisture and temperature were measured during the years 2004, 2005, and 2007, at three locations along the slope. Four of the chambers on the wetland edge are located in soils with a deep organic layer (18 cm) and high soil water content. The other four chambers are located on drier soils with a thinner organic layer (5 cm). The two sets of chambers are referred to in the analysis as the 'wetland margin' and 'upland' chambers. An additional set of two clear flux chambers measured net ecosystem exchange (NEE) in the wetland during 2006 and 2007.



    In general the wetland margin and upland chambers follow a similar seasonal pattern, but the magnitude of upland chambers' SR exceeds that of the wetland margin during the late summer/early fall (Fig. 1). The timing of when the two slope locations SR diverges varies by year. Wetland night NEE or ecosystem respiration (ER) was typically lower than SR at any chamber location. There is a seasonal hysteresis of SR with respect to temperature- higher SR in fall than spring at a given 2 cm or 5 cm soil temperature. The magnitude of the hysteresis varies between years, and the hysteresis is more pronounced in the upland chambers. There is a distinct diel hysteresis of SR in which SR peaks at night. The pattern of this hysteresis varies by location, season, and year. The seasonal changes in the temperature dependence of SR suggest that the seasonal pattern is influenced by temperature gradients in the soil or the phenology of plant or microbial biomass.



    SR was compared to measurements of ER measured at the eddy covariance flux tower for the years 2003-2006. SR reached a maximum later in the year than total ER measured at the eddy covariance flux tower (Fig. 2), and their seasonal hystereses were roughly opposite when plotted against soil temperature. SR exceeds ER during the late summer/early fall which suggests that some soil-respired carbon is advected away from the tower and vented through the canopy at another location, or the tower is preferentially measuring areas of lower SR such as the wetland. The difference in the timing and hystereses of the SR and ER measurement is consistent with the notion that the seasonal patterns of vegetation and soil are not temporally aligned.

  • Research Category: Forest-Atmosphere Exchange

  • Figures:
  • flux_time_series.jpg
    ratio.jpg