It is well known that soil CO2 flux can exhibit pronounced day-to-day variations; however, measurements of soil CO2 flux with soil chambers typically are done only at discrete points in time. This study evaluated the impact of sampling frequency on the precision of cumulative CO2 flux estimates calculated from field measurements. Automated chambers were deployed at two sites in a no-till corn/ soybean field and operated in open system mode to measure soil CO2 fluxes every hour from 4 March 2000 through 6 June 2000. Sampling frequency effects on cumulative CO2-C flux estimation were assessed with a jackknife technique whereby the populations of measured hourly fluxes were numerically sampled at regular time intervals ranging from 1 d to 20 d, and the resulting sets of jackknife fluxes were used to calculate estimates of cumulative CO2-C flux. We observed that as sampling interval increased from 1 d to 12 d, the variance associated with cumulative flux estimates increased. However, at sampling intervals of 12 to 20 d, variances were relatively constant. Sampling once every 3 d, estimates of cumulative C loss were within +-20% of the expected value at both sites. As the time interval between sampling was increased, the potential deviation in estimated cumulative CO2 flux increased such that sampling once every 20 d yielded potential estimates within 60% and 40% of the actual cumulative CO2 flux. A stratified sampling scheme around rainfall events was also evaluated and was found to provide more precise estimates at lower sampling intensities. These results should aid investigators to develop sampling designs to minimize the effects of temporal variability on cumulative CO2-C estimation.
