HR: 10:20h
AN: B52A-01 INVITED     [Abstracts]
TI: Deconvolving Soil CO2 Efflux at Three Temporal Scales
AU: * Davidson, E A
EM: edavidson@whrc.org
AF: Woods Hole Research Center, P.O. Box 296, Woods Hole, MA 02543 United States
AU: Savage, K E
EM: savage@whrc.org
AF: Woods Hole Research Center, P.O. Box 296, Woods Hole, MA 02543 United States
AB: Soil respiration is a combination of plant and microbial processes that respond to climatic drivers at a variety of temporal and spatial scales. Deconvolving the soil respiration signal into several component processes and at different time scales may help us understand the controls on each. We used correspondence analysis and regression analysis of half-hourly soil CO₂ efflux measurements using automated chambers at the Harvard Forest of central Massachusetts to investigate the climatic drivers of soil respiration at three time scales: diel cycles, synoptic weather patterns, and seasonality. The diel cycle corresponded to diel variation in soil temperature. The amplitude of the diel cycle was proportional to the daily mean flux. In other words, the diel variation is greatest under warm and moist conditions. Synoptic weather patterns bring rainfall events that generally lead to a pulse of CO₂ production and efflux, which is revealed by correspondence between CO₂ efflux and soil water content at time scales of 4-14 days. The magnitude of the wet-up pulse was correlated with the number of preceding dry days and the size of the precipitation event. Once the effects of diel cycles and wetting events are partitioned out of the data, the remaining trend is seasonal and was correlated with soil temperature. However, this seasonal pattern was not symmetric, but rather showed a steeper incline during the spring than decline during the autumn. This seasonal hysteresis is consistent with a stronger contribution of respiration of growing roots during the spring than during the autumn. By assuming that wintertime CO₂ efflux is primarily from microbial respiration and that the microbial response to seasonal temperature variation follows an exponential fit with a Q10 of about 2.5, then the difference between this modeled seasonal pattern of microbial respiration and the deconvolved seasonal pattern of total CO₂ efflux yields a rough estimate of the seasonal pattern of root respiration, which was skewed toward the spring, as would be expected for temperate forests.
UR: http://whrc.org/new_england/Harvard_Forest/index.htm
DE: 0414 Biogeochemical cycles, processes, and modeling (0412, 0793, 1615, 4805, 4912)
DE: 0426 Biosphere/atmosphere interactions (0315)
DE: 0428 Carbon cycling (4806)
DE: 0439 Ecosystems, structure and dynamics (4815)
DE: 0486 Soils/pedology (1865)
SC: Biogeosciences [B]
MN: Fall Meeting 2005