Imbibition into Soil and Rock: What's the Same, What's Different?

Water imbibing into a dry soil is affected by buoyancy, capillary, and viscous forces, as well as by properties of the soil at the scale of both the pore and the pedon. Imbibition can therefore be used to assess some basic soil properties, such as sorptivity and saturated hydraulic conductivity; this is the point of (for example) the Bruce & Klute horizontal infiltration experiments. A change in imbibition rate with time should indicate a change in the soil encountered, such as the wetting front going from a coarse to a fine-textured soil. We conducted a suite of imbibition experiments using rock cores that were uniform, and yet we found changes in the slope of log(mass imbibed) versus log(time). Repeated experiments confirmed that it was not an error, and statistical analysis showed that the change in slope was real. We hypothesized that the change in slope was related to low pore connectivity in the rock. If pore connectivity is indeed the controlling variable, then the distance from the core's inlet face to the cross-over point should vary with core height/diameter ratio. This variation was observed in both experiments and simulations, supporting our conclusion that a key variable for fluid flow in rock and soil is the connectivity of the pores.