Proximity AE doublets are introduced as a new category of AE doublets in this paper.Proximity AE doublets have a similar waveform, and they occur in a succession with a short interval.Proximity AE doublets have the second event that arrives before attenuation of the first event.Their interval is shorter than 1 s.On the other hand, interval between events of the conventional AE doublets/multiplets ranges from some seconds to some days.Because of this short interval of the proximity AE doublets, relative location between AE sources can be estimated more accurately in the proximity AE doublets analysis than in the conventional AE doublets/multiplets analysis.
This paper also describes a signal processing technique to determine intervals of P-wave and S-wave between two events of proximity AE doublets.It is difficult to estimate similarity between the two waveforms and their interval because the two events overlap each other.Cepstrum analysis is applied to estimate both the interval and the similarity.Proximity AE doublets have one or two peaks in their cepstrum, and quefrencies of the peaks are intervals of P-wave and S-wave between the two events.Three-component AE signal is rotated in the directions of P-wave and S-wave, and data before the first S-wave arrival is replaced with zeros so that the intervals can be determined from the peaks in the cepstrum analysis.Cepstrums of these processed three-component AE signal can show intervals of P-wave and S-wave between the two events of proximity AE doublets.

View full abstract

In the hydraulic stimulation, a number of micro-seismic events are commonly observed. It is believed that injected fluid flows through subsurface pre-existing fractures, and the fluid flow raises pore-pressure in those fractures, and finally the raised pore-pressure triggers micro-seismic events. In the previous work of Osada et al. (2005), we have considered carefully such a scenario and come up with a method to estimate pore-pressure distribution in a rock formation by analyzing the observed data of micro-seismic events. On the other hand, pore pressure distribution along flow-pathways should change according to the location of flow-pathways and the distribution of hydraulic conductivity along them. Therefore, it could be possible to estimate flow-pathway structure as it gives a good explanation of the pore pressure distribution. Based on this idea, in the present work we propose a new method to estimate the flow-pathway structure. We assume an appropriate model of flow-pathway structure and optimize it as the pore pressure distribution estimated numerically by the model agrees well with the pore pressure distribution estimated from micro-seismic events by the method proposed in the previous work. Then we accept the most optimized model as the answer to represent the actual flow-pathway structure.We demonstrated how we can optimize the model of flow-pathway structure according to synthetic pore pressure distributions in 1D and 2D cases. Finally we applied this method to Soultz field in France and estimated the subsurface flow-pathway structure. The results were in good agreement with the flow logs carried out in the injection well.

View full abstract