Litter and soil respiration in a mixed hardwood forest in eastern Tennessee

Ongoing measurements of soil and litter respiration at the Chestnut Ridge Ameriflux site in Oak Ridge, TN were compared to empirical models. Soil respiration modeled using the 10 cm soil moisture and temperature captured daily changes in respiration well, but did not describe more rapid changes in half-hour soil respiration rates. Litter respiration was measured using a modified portable soil respiration chamber, and litter moisture was monitored manually using litter-basket weight. In addition, continuous automated litter moisture measurements were recorded using a new low-cost soil moisture probe. A minimum of 10 moisture probes were required for a representative litter moisture measurement, and a linear relationship between the average automated moisture measurements and the manual measurements was determined. Automated litter moisture and litter temperature measurements were used to model litter respiration. Despite the limited magnitude of the measured litter fluxes, which ranged from 0.1 to 1.4 µmol m-2 s-1, the litter respiration model fit the measurements well (R2 = 0.69). Half-hourly changes in litter temperature and litter moisture drove variability in the modeled litter respiration that may help describe the observed variability in soil respiration.