Indirect Estimation of Soil Water Retention Curve Using Soil Imaging Penetrometer (SIP)

Various methods have been developed and tested to estimate soil water retention curves (WRC) indirectly. These methods do not always provide reliable results. Their accuracy typically depends on the datasets these methods are derived from or the assumptions they use to characterize the soil pore structure. We propose obtaining the soil pore structure information necessary to estimate soil WRC directly from soil images. The proposed approach minimizes the reliance of soil WRC estimates on other soil data sources and utilizes more information obtained in a quick and cost-effective manner for the soil in question for improved predictions. In order to test and validate the usefulness of soil images and analysis techniques, we propose to use soil image based pore structure information in three different modeling approaches: pedotransfer functions, physical models and fractal models in combination with percolation theory. These approaches represent the major categories of research done in the literature to estimate soil WRC indirectly. In Phase I we anticipate determining the best image analysis and indirect measurement methods to be used for our Phase II development. The ultimate goal is to develop a library of images and image processing routines for commercial applications for a wide range of soils.