“…These methods generated an artificial COS gas concentration gradient between the headspace of the enclosure and the soil, and thus not only led to an overestimation of natural emission strength by enhancing the diffusion of COS from the soil into the headspace, but also masked the potential of soils as a sink. As a result, these estimates all indicated that soils are a source of COS [ Aneja et al , 1979; Steudler and Peterson , 1985; Carroll et al , 1986; Goldan et al , 1987; Fall et al , 1988; Staubes et al , 1989], contributing around 21–25% of the total global source [ Khalil and Rasmussen , 1984; Chin and Davis , 1993; Johnson et al , 1993]. In contrast, Castro and Galloway [1991] suggested a much enhanced role for soils in the removal of COS from the atmosphere, and De Mello and Hines [1994], using ambient air as sweep gases, consistently found that soils took up COS at rates of 3 to 30 ng S m −2 min −1 (1.6–15.6 pmol m −2 s −1 ).…”