The Kaiser stress is then: = ( ) (12) This operational definition corresponds to the first statistically significant increase of AE activity during loading, consistent with the standard Kaiser-effect concept used for stress memory in rocks. 3. RESULTS AND ANALYSIS Specimens were evaluated using three predefined quality criteria. A specimen was classified as valid only if: • a Kaiser point could be identified (σK not empty), meaning a distinct AE onset attributable to stress-memory could be determined; • the overall AE activity was sufficient (Total hits ≥ 5000) to ensure robust detection; • the Kaiser point did not occur too late relative to peak stress (σK/σpeak ≤ 0.65), so that the detected onset reflects a reproducible stress-memory threshold rather than late-stage damage close to failure. Table 2. Validity of samples cut from different boreholes Borehole / dataset Lithology Total n Valid Not Valid Validity rate BH1-roof Dolomite 9 5 4 55.6% BH1-roof Anhydrite 17 13 4 76.5% BH3-roof Dolomite 9 8 1 88.9% BH4-roof Dolomite 25 14 11 56.0% BH1-sidewalls Anhydrite 6 4 2 66.7% BH2-sidewalls Anhydrite 6 4 2 66.7% BH3-sidewalls Sandstone 9 8 1 88.9% BH4-sidewalls Sandstone 4 4 0 100% The plot in Figure 4 shows cumulative AE hits versus axial stress σ for dolomite samples extracted from roof strata (BH-3). Most curves, marked green, follow a consistent pattern: a short low-activity phase, then a clear onset of AE (typically ~8–15 MPa), a rapid increase in CumHits up to ~25–30 MPa, and finally a plateau (stabilisation) at different levels (~40–100k hits), reflecting natural material variability and measurement effects (e.g., sensor coupling/threshold). Still, one sample was a clear outlier: it remains almost inactive until ~35 MPa and then shows a strong increase between ~35–50 MPa before reaching a plateau. This right-shifted AE onset is inconsistent with the rest and would artificially bias the estimated Kaiser stress (σK) upward, therefore it was excluded as an outlier.
RkJQdWJsaXNoZXIy MTM0Mzk2