Factors controlling soil respiration in a deciduous mixedwood forest: Evaluations at daily and sub-hourly time steps for land surface modelling

The release of carbon dioxide from the soil occurs as a result of the decay of organic matter (heterotrophic respiration, R_h) as well as growth and maintenance respiration from roots (autotrophic respiration, R_a,root and R_m,root). In the field, measurements with chambers capture the total soil respiration (R_s). Typically, the largest control on soil respiration is soil temperature, which governs the potential rate of microbial activity. The relationship between R_s and temperature is usually quite well defined over a range of soil moisture, and is represented well using an exponential model. At near-saturated conditions and as the soil dries beyond a certain threshold moisture value, R_s loses its sensitivity to temperature, and decreases as microbial activity slows.

In land surface models, R_h, which usually comprises more than half of R_s, is typically modelled using an exponential function based on soil temperature in one or more soil layers, and sometimes in a separate litter layer. The sensitivity of R_s to temperature is usually represented using a fixed Q₁₀ value, representing the sensitivity of R_s to a 10°C change in soil temperature, and some models include a relationship with soil moisture or matric potential. The Canadian Terrestrial Ecosystem Model (CTEM) models daily heterotrophic respiration in the litter and soil carbon pools separately, using Q₁₀ values that vary with temperature. An optimal matric potential (suction) of 0.04 to 0.06 MPa is assumed with R_s decreasing at larger and smaller matric potentials.

In this study, we examine R_s measured over six growing seasons at the Borden Forest Research Station, a deciduous mixedwood forest in southern Ontario (44°19'N, 79°56'W). An automated single-chamber system (LI-COR Inc., model LI-8100) was employed in a primarily deciduous area, with occasional manual spatial measurements in deciduous and coniferous areas of the forest. The behaviour of soil respiration is evaluated with respect to soil temperature and soil moisture at various depths. The analysis is shown for both daily and half-hourly time scales in order to determine whether the relationships are dependent on the time step at which they are evaluated.

Relationships with temperature were in the range of those reported for other temperate and boreal forest sites; Q₁₀ was 4.5 relative to temperature at 20 cm depth with volumetric soil moisture greater than 15% using half-hourly data, and decreased to less than 4.1 when evaluated on a daily basis. Q₁₀ decreased to 2.7 when evaluated with respect to daily temperature at 5 cm depth; this value is larger than that of CTEM (~2.0) for a similar temperature range and optimal moisture conditions. Interannual variability was strongly controlled by changes in soil moisture. Drying of the soil resulted in reductions in the observed R_s and a significant moisture control was evident as soil moisture decreased below 10% by volume.