Field tests on water storage capacity of loess-gravel capillary barrier covers

The climate in northwest China is arid and the loess which is technically feasible and economical used for soil cover in landfills is widely distributed. At Jiangchungou Landfill, Xi'an, the first large size loess-grass capillary barrier cover 20 m×30 m is built, and the extreme rainfall experiments are carried out. The results of water distribution tests show that: with 214.8 mm rainfall, the slope runoff is 1.7 mm, accounting for 0.8% of the total rainfall, and the storage of soil (containing evaporation) is 199.57 mm, accounting for 92.9% of the total rainfall, and the leakage is 11.53 mm, accounting for 6.3% of the rainfall. The analysis of matrix suction and water migration shows that: with continuous rainfall, the pore pressure (or volume water content) of the surface soil (above depth of 15 cm) and the bottom soil (below depth of 85 cm) in the capillary-barrier cover are all high. The high pore pressure (or volumetric water content) of the underlying soil is due to the capillary-barrier effects at the gravel-loess interface, which is the distinct feature of rainfall infiltration water movement different from that of single soil layer. The evaluation of water storage capacity shows that: the effective water storage capacity of the loess-grass cover is 251.95 mm, measured by the rainfall experiments. The theoretical value of the effective water storage S_fac is 218.75 mm, evaluated by the indoor hygroscopic soil-water characteristic curve. The measured value is 15.18% larger than the the theoretical one, and the results are safe. The theoretical value of the effective water storage S_fac is 278.32 mm, evaluated by the field hygroscopic soil-water characteristic curve. The measured value is 9.47%, smaller than the theoretical one, and the results are dangerous. It is suggested that the indoor hygroscopic soil-water characteristic curve should be adopted in anti-seepage design.