Soil pollution with oil is a serious environmental problem. Although the oil movement in soil must be recognized, the thermal properties of oil-polluted soil, which are required for simulation, were less studied. In this report, we focused on thermal conductivity of the oil-polluted soil. The objectives of this study were to clarify the effects of liquid phase ratio, temperature and water-oil ratio on the thermal conductivity of oil-polluted soil. Washed Tottori dune sand and rape seed oil were used as experimental samples. The thermal conductivity of samples was measured by KD2 under several experimental conditions of the liquid phase ratio, temperature and water-oil ratio. The thermal conductivity of oil-polluted soil increased with the liquid phase ratio, even when the soil samples contained both of oil and water, and showed different trends according to the liquid phase ratio and water-oil ratio when the temperature increased. It was suggested that the thermal conductivity of soil polluted with organic oil might show the opposite trend to that of the non-polluted soil by increase in temperature. When the liquid phase ratio and temperature were fixed, the thermal conductivity of oil-polluted soil decreased linearly to the water-oil ratio. Finally, we proposed an empirical formula to predict the thermal conductivity of oil-polluted soil which contains water and oil by using the liquid phase ratio, water-oil ratio, and thermal conductivities of soil non-polluted and polluted with only oil.
The soil hydraulic parameters for analyzing soil water movement can be determined by fitting a soil water retention curve to a certain function, i.e., a soil hydraulic model. For this purpose, a program which performs nonlinear fitting of soil water retention curves to 4 models, Brooks and Corey model, van Genuchten model, Kosugfs lognormal pore-size distribution model, and Durnefs bimodal pore-size distribution model, by Levenberg-Marquardt method was developed. The program was written in GNU Octave, a high-level language primarily intended for numeric calculation. It was shown that the developed program actually works fine with many soil water retention curves in UNSODA database. It was also shown that by comparing coefficients of determination and fitting curves of each model, the most appropriate hydraulic model for a certain soil can be selected. Therefore this program not only calculates the necessary parameters, but also helps determining which model is to be used for a particular purpose.