Gas Transport Parameters for Peaty Soil: Effect of Peat Shrinkage Induced by Successive Drainage

Accurate prediction of gas transport parameters (soil-gas diffusion coefficient, Dp, and air permeability, ka) is important when investigating the fate and transport of gaseous phase contaminants and in quantifying the emission of methane from peat (wetland) soil. However, only limited measurements and knowledge of Dp and ka, especially for peat soils, are available. In this study, Dp and ka were measured on undisturbed 100-cm3 peat soil samples (triplicate) taken from the Bibai wetland in Hokkaido, Japan at 30-, 60-, and 90-cm depths. The undisturbed cores collected at 60-cm depth were sampled in both horizontal and vertical directions. Each soil sample was drained to different soil-water matric potentials of pF (= log (-ψ, cm H2O)) 1.0, 1.5, 1.8, 2.0, 3.0, 4.1, and air-dry condition before measurements of Dp and ka. The peat soil samples showed remarkable shrinkage during successive drainage to the different pF values. Sudden decrease in soil sample volume was seen at the transition from pF 1.5 to 1.8. At pF > 1.8, the soil sample volume continuously decreased to around 30 percent of the original volume at air-dry condition. Shrinkage of the soil sample affected the soil pore structure, and, consequently, markedly influenced the magnitudes of Dp and ka. As a result, Dp and ka did not exhibit an expected monotonically increase with soil-air content, ɛ (m3 m-3). At pF 2 to pF 4.1, there was no significant increase in Dp and ka. These variations of Dp and ka with ɛ could be explained from the changes in the pore structure of a peat soil which is classified as macropores (pores in between structures) and matrix pore (structural pores). We evaluated the pore connectivity shape factor X (defined as X = log(Dp/Do)/log(ɛ)) and equivalent pore diameter dg defined as dg = (8ka/(Dp/Do))1/2) as indices of changes in pore structure. The variation of the X shape factor with the increase in ɛ involves a three\- fold process. First, at pF<1.5, the large almost highly connected straight macropores were drained resulting to X values less than 2. More macropores become connected increasing the tortuous pathways for gas diffusion thereby increasing the magnitude of X. At pF 1.5, the macropores are assumed to be completely drained, in agreement with the observed high equivalent pore diameter dg. Also at pF 1.5, the soil sample will likely result in a sharp decrease in sample volume due to the closure of macropore apertures because of the absence of soil- water in the macropores. Second, after the sudden sample shrinkage at pF 1.5 to 1.8, the X value followed an almost linear increase with ɛ within the range of 0.2 to 0.4 m3 m-3. This occurs at the condition where the matrix pores are starting to drain. Both the decrease in soil-water and the further decrease in soil volume resulted in the gentle increase in Dp/Do and ka at pF 1.5 to pF 4.1. Third, at pF > 4.1, the additional matrix pore spaces become highly tortuous because of the observed remarkable shrinkage, at this dry condition. Thus, pore structure changes due to soil shrinkage must be taken into consideration in order to characterize accurately the gas transport parameters.