Some effects of trapped air in wet soils
The effects of air trapped in soils at zero or small negative values of matric potential were studied in relation to saturated and unsaturated hydraulic conductivity, the water release characteristic, and root growth. A reliable method was developed for measuring the trapped air porosity of 'saturated' soils. Using this method, the surface layer of three soils of contrasting texture, in a relatively undisturbed state, was found to have similar amounts of trapped air (5 to 7% of total soil volulme). After digging, the amount of air trapped by ponding water on the surface increased to a value of about 10%. Values of trapped air porosity below a stationary water-table were very low (1%) and this can be attributed to opportunity which the trapped air has to go into solution and diffuse to the water-table. The effects of trapped air on the water release characteristic were investigated in both field and laboratory studies using the neutron-probe and both the neutron-probe and the gamma-probe to measure the water content in the field soil and in a tank packed with sand in the laboratory, respectively. The presence of trapped air had an effect on the field water release characteristic. This effect was more pronounced in dug soil than in undisturbed soil. In the laboratory, the amount of trapped air differed according to the method of wetting, less trapped air occurred when wetting was from the bottom upwards compared to wetting by ponding water from the top. Amounts of trapped air decreased with increasing depth of sand. The water release characteristic was found to depend not only on the method of wetting but also on the previous history of wetting. In a comparison between a tension-table water release and the water release characteristic measured in a sand tank, they were found to be closely similar except at zero and small negative values of matric potential, where effects due to varying amounts of trapped air with depth and wetting history caused differences. Unsaturated hydraulic conductivity measured in the field was affected by the presence of trapped air but only at potentials close to zero. In the laboratory, the values of saturated hydraulic conductivity measured in the sand tank and the values calculated from the water release characteristic using Marshall's theory agreed to within the limits of experimental accuracy. Unsaturated hydraulic conductivity values were found to be not only dependent on the method of wetting but also on the previous history of wetting. The results suggested that once dry sand has been wetted and allowed to drain, that rewetting leads to trapping of air in large pores which reduces the flow during the next drainage period. There was an agreement between unsaturated hydraulic conductivity variation with matric potential obtained from the instantaneous profile method in the sand tank and that using Marshall's theory, however, it is not to be expected that such an agreement will be found for most soils which are not comparable to the relatively homogenous sand used here. In a laboratory experiment with winter barley, the relationship between the presence of trapped air, soil aeration and root growth was studied at 9 and 15°C. The techniques of measuring redox potential with bare platinum electrodes, measuring exygen flux with the same platinum electrodes, and measuring the concentrations of O2, CO2 and N2O in samples extracted through a hypodermic needle gave results which were consistent with one another and with observed root growth rates. Root extension rate was much slower in the saturated soil than in the freely-drained soil. At both temperatures, root growth during 'daytime' (the time of illumination) was about 3 times greater than at night. At 15°C, there was a greater rate of root growth in soil cores which had been flooded rather than vacuum saturated which was related to a slightly greater measured rate of oxygen flux from air trapped by the flooding procedure.