Specifying the Quantitative Detection of Non-trivial Nucleation under Acoustic Emission Observation

Abstract:

The concept of contaminant sphere radius (Reches and Lockner, 1994) and AE nucleation point (Figure 1(b)) has helped Reches (1999) in specifying the analogues among a quasi-static rock rupture test (Figure 1(a), Lockner et al. (1991)), the earthquake modeling prediction (Reches and Lockner, 1994), and real earthquake observations (Ellsworth and Beroza, 1995). This result had highlighted the “rupture” concept as an earthquake mechanism, which had been overlooked for decades (Reches, 1999). However, determination of the AE nucleation point is empirical and to a certain extent, subjective. We devise a randomization process to give the AE nucleation point an objective and quantitative definition. This randomization process shuffles the sequence, but will not change the location of the events, as such the contaminant sphere radius curve for the randomized AE data represents the natural fluctuation of the AE event concentration. We validate the devised method on the AE events released from a well-known flaw-controlled and symmetric fracture process (Figure 1(c) and (d)). The contaminant sphere radius curve for the randomized data is significantly different than that of the original data (Figure 1(e)). The randomized data results in a noise-like curve of the minima contaminant sphere radius (Figure 1(f)), and the distribution of those minima obey normal distribution (Figure 1(g)). As such, we can define the non-trivial AE nucleation point based on the statistical confidence levels and imposed these levels into the original data (Figure 1(e)). One surprising discovery through the observation at the non-trivial AE nucleation point is that, the macro-scale well-known flaw-controlled symmetric fracture process can, in meso-scale, display temporal anti-symmetry (Figure 1(h)). This anti-symmetry is a reflection for the existence of asynchrony from a macro-scale symmetric fracture development process. We further have proven this observation of asynchrony is neither a result from the bias of the AE system sensitivity nor from a special selection on the window length for calculating the contaminant sphere radius. As such we conclude this meso-scale anti-symmetry and asynchrony in fracture development is a natural phenomenon in the macro-scale symmetric fracture process.