Reduction of bias in static closed chamber measurement of δ13C in soil CO2 efflux

The 13C/12C ratio of soil CO2 efflux (δe) is an important parameter in studies of ecosystem C dynamics, where the accuracy of estimated C flux rates depends on the measurement uncertainty of δe. The static closed chamber method is frequently used in the determination of δe, where the soil CO2 efflux is accumulated in the headspace of a chamber placed on top of the soil surface. However, it has recently been shown that the estimate of δe obtained by using this method could be significantly biased, which potentially diminish the usefulness of δe for field applications. Here, analytical and numerical models were used to express the bias in δe as mathematical functions of three system parameters: chamber height (H), chamber radius (Rc), and soil air‐filled porosity (θ). These expressions allow optimization of chamber size to yield a bias, which is at a level suitable for each particular application of the method. The numerical model was further used to quantify the effects on the δe bias from (i) various designs for sealing of the chamber to ground, and (ii) inclusion of the commonly used purging step for reduction of the initial headspace CO2 concentration. The present modeling work provided insights into the effects on the δe bias from retardation and partial chamber bypass of the soil CO2 efflux. The results presented here supported the continued use of the static closed chamber method for the determination of δe, with improved control of the bias component of its measurement uncertainty. Copyright © 2009 John Wiley & Sons, Ltd.