Lawrence Berkeley National Laboratory

A general approach is presented here which allows estimation of field-scale thermal properties of unsaturated rock using temperature data collected from in situ heater tests. The approach developed here is used to determine the thermal conductivities of the unsaturated host rock of the Drift Scale Test (DST) at Yucca Mountain, Nevada. The DST was designed to obtain thermal, hydrological, mechanical, and chemical (THMC) data in the unsaturated fractured rock of Yucca Mountain. Sophisticated numerical models have been developed to analyze these THMC data. However, though the objective of those models was to analyze "field-scale" (of the order of tens-of-meters) THMC data, thermal conductivities measured from "laboratory-scale" core samples have been used as input parameters. While, in the absence of a better alternative, using laboratory-scale thermal conductivity values in field-scale models can be justified, such applications introduce uncertainties in the outcome of the models. The temperature data collected from the DST provides a unique opportunity to resolve some of these uncertainties. These temperature data can be used to estimate the thermal conductivity of the DST host rock and, given the large volume of rock affected by heating at the DST, such an estimate will be a more reliable effective thermal conductivity value for field scale application. In this paper, thus, temperature data from the DST are used to develop an estimate of the field-scale thermal conductivity values of the unsaturated host rock of the DST. An analytical solution is developed for the temperature rise in the host rock of the DST; and using a nonlinear fitting routine, a best-fit estimate of field-scale thermal conductivity for the DST host rock is obtained. Temperature data from the DST show evidence of two distinct thermal regimes: a zone below boiling (wet) and a zone above boiling (dry). Estimates of thermal conductivity for both the wet and dry zones are obtained in this paper. Sensitivity of these estimates to the input heating power of the DST is also investigated in this paper. These estimated thermal conductivity values are compared with core measurements and those estimated from geostatistical simulations. Note that the approach presented here is applicable to other host rock and heater test settings, provided suitable modifications are made in the analytical solution to account for differences in test geometry.