Examining the Effectiveness of Digital Rock Physics Without Segmentation for Thermal Conductivity Estimation of Sandstones

To elucidate the temperature distribution at depth, thermal properties of the mantle, crusts, and sediments are needed in addition to temperature logs. Although a lot of rock core samples have been retrieved by scientific drilling projects including International Ocean Discovery Program (IODP), there are no other ways to know thermal properties of these cores except for measurements currently Since it is almost impossible to measure many samples easily, a method to determine thermal properties without measurements is now required. "Digital Rock Physics (DRP)" may be a solution that utilizes data called "CT number" of X-ray computerized tomography (X-ray CT) images and estimates physical properties of rocks through numerical processes. In this research, we estimated the thermal conductivity of sandstones without segmentation (known as the "segmentation-less method"). The results were examined by comparing with measurement data by the Hot-Disk method (also called the transient plane source (TPS) technique). At first, CT number was converted to density of each voxel by a best-fit continuous function, and then density was converted to porosity based on the monomineralic assumption. Thermal conductivity and specific heat were calculated at each voxel by referring to porosity with four types of mixing laws: the harmonic mean, the arithmetic mean, the geometric mean, and the square root mean, creating four digital rocks. After solving the three-dimensional thermal conduction equation in four digital rocks using the finite difference method, thermal conductivity of the whole inspection area was calculated. As a result, we found that the segmentation-less DRP is effective for estimating thermal conductivity and the geometric mean gives the most valid model for applying DRP to sandstones. Although DRP cannot reproduce smaller structures than the size of a voxel, the segmentation-less method