ABSTRACT
This paper presents a methodology for evaluating the complete stress field and its variation versus depth. The validity of the protocol is visualized in the case study at the Äspö Hard Rock Laboratory (HRL), south-eastern Sweden, a site predominantly sampled using the overcoring method. Overcoring data involve explicit (measurement-related) as well as implicit uncertainties. The former include for example uncertainties regarding determination of the location of the test sections in physical space and of the value of elastic parameters, as well as uncertainties in strain-/displacement measurements, etc. The explicit types of uncertainties are fairly straightforward to analyze and correct for during the stress calculation procedure. The implicit uncertainties, on the other hand, such as the assumption of homogeneity and linear-elasticity, are much more difficult to appreciate and correct for, if possible at all. Yet, as for explicit errors, they may render an individual test or a series of tests completely meaningless, and it is therefore crucial that both categories of uncertainties are identified, understood, and properly considered within the process of stress field determinations.
The proposed methodology follows the directions outlined by ISRM for rock stress estimation using overcoring methods (Sjöberg et al., 2003; Sjöberg & Klasson, 2003). In addition, we pay particular attention on avoiding, identifying, and correcting for various potential sources of error, the sampling strategy, and considerations of the continuity hypothesis.
