Information on soil CO2 emissions can be used in conjunction with data for C inputs from plants to estimate the soil C balance. Many studies assume a single-time-point measurement of soil CO2 flux taken in a day (Fh) is equal to its daily average value (Fd), which could result in over- or underestimation. A model using Fh, the temperature at the time the Fh is measured (Th), daily average temperature (Td) and Q10 factor to predict Fd was tested with extensive measurements of soil CO2 flux, temperature and moisture over 60 d from various treatments (no-till wheat, summer fallow and stubble, etc.) on a Swinton silt loam near Swift Current, Saskatchewan, Canada. A distinct hysteresis between flux and soil moisture was observed following rain events. Therefore, data for five rainy days were excluded from the analysis as the model does not consider the effect of hysteresis. Calculated Q10 factors were 1.86 and 1.54 for soil (Ts) and air (Ta) temperatures, respectively. The model using either Ts or Ta improved prediction of Fd in both calibration (49 d) and validation (6 d) datasets compared with Fh. Values of Fh measured in 6 yr were higher than modelled values of Fd in 96% of the 1602 treatment-days, hence if Fh is assumed to be equal to Fd, the averaged overestimation would be 17%.Key words: Soil CO2 flux, Q10 factor, temperature
Read full abstract