Forest floor CO2 flux estimated from soil CO2 and radon concentrations.

Abstract Having a quantitative understanding of the carbon cycle in forests is of great importance for predicting global warming issues. Carbon dioxide production in soil is the largest CO 2 source in forests, and exhibits large temporal and spatial variations. Continuous observation of soil CO 2 flux at many sites over a forest is therefore necessary to obtain representative soil CO 2 fluxes for the forest. In this study, a gradient method to measure soil CO 2 flux indirectly from soil radon and CO 2 measurements was theoretically modified to conveniently measure the soil CO 2 flux from soil radon and CO 2 concentrations measured at one soil depth. To experimentally test the modified method, a field observation was conducted continuously in a forest over a 31-day period. Since changes in the soil water content near the soil surface were small throughout the observation, a constant effective diffusivity for CO 2 was assumed for the soil CO 2 flux estimation. The soil CO 2 flux was then calculated as the product of the effective diffusivity and the gradient of the soil CO 2 concentration, each calculated from soil radon and CO 2 concentrations. The estimated flux ranged from 1.9 to 5.8 μmol m −2  s −1 , and, correlating well with the reference value, measured with a conventional ventilated-chamber method. We therefore conclude that the modified gradient method based on the measurement of soil CO 2 and radon concentration at one depth is reliable, at least under conditions where the change in the soil water content is small.