Stress triggering and stress memory observed from acoustic emission records in a salt mine

Abstract

SUMMARY Some of the most interesting questions in geosciences are whether results from laboratory experiments can be applied to processes in the earth crust and whether in situ studies with high spatio-temporal resolution can bridge the gap between laboratory work and seismology. In this study, acoustic emission (AE) activity caused by stress changes due to the backfilling of a cavity in an abandoned salt mine is studied to answer questions regarding (1) the dependence of AE event rates, event distribution and b-value on the stress state, (2) the stress memory effect of rock (Kaiser effect), (3) the possibility to detect significant changes in the system like the initiation of macrocracks and (4) the possibility to estimate future activity from previous AE records. The large number of events studied (>3 × 10 5 ) allows a spatial resolution of the order of 1 m and a temporal one on the order of 1 hr. Stress changes are created due to the thermal expansion and contraction of the rock mass in response to the temperature changes caused by the backfilling. A roughly 20 × 50 × 50 m section of the mining complex just above the backfilled cavity is well covered by a network of 24 piezo-electric receivers and poses an optimal volume for the study. Results of a 2-D finite element thermoelastic stress model are in agreement with the spatio-temporal AE event distribution. In addition to the initial upward migration of the AE event front, which correlates with the calculated stress field, the rock salt exhibits a pronounced Kaiser effect for the first few thermal loading cycles throughout the whole study region. The deviation from the Kaiser effect during later loading cycles seems to be caused by the initiation of a planar macroscopic crack, which is subsequently reactivated. AE activity tends to concentrate along this macrocrack. Calculated b-values decrease before and increase after the supposed initiation of the macrocrack supporting this explanation. In intact rock volumes not subjected to macrocracking a linear relation between the maximum event rate and the calculated absolute Coulomb stress increase is observed. This indicates that future maximum AE event rates can be estimated from expected loading. AE activity during stress loading cycles is most prominent in regions with Coulomb stress maxima indicating possible shear cracking that has the potential to create macrocracks. Strong bursts of AE activity observed during thermal unloading phases are concentrated in regions for which the minimum principal stress becomes tensile. These regions exhibit significantly higher b-values than those active during thermal loading. We interpret these weak events with tensile microcracking.