Use of existing pore models and X-ray computed tomography to predict saturated soil hydraulic conductivity

This study investigates the use of 3D soil pore characteristics (volume, surface area, and tortuosity) for prediction of saturated hydraulic conductivity (Ks). The pore characteristics were determined by cluster labeling of X-ray CT images and random-walk simulations of 3D pore space. The flow characteristics of the four soil cores were measured in the laboratory pre- and post-CT imaging. The predicted values of Ks were arrived at through the use of Darcyʹs equation and a modified Poiseuille equation. For comparison the 2D pore characteristics were determined from individual slice of X-ray CT imagery, and fed into three established methods of predicting Ks. The predicted Ks by the proposed method exhibited a positive correlation to both pre- and post-imaging measured Ks, and attained a greater correlation than the 2D Ks prediction.