The industry is notoriously cyclical. As the price of oil fluctuates, so does the level of E&P activity that is the industry’s calling and passion. One fact remains constant throughout good times and bad: The oil and gas industry is driven by technical innovation. When prices are high, the industry innovates new ways to find resources previously out of reach. When prices are low, the industry develops new ways to reduce the cost of producing the resources it has in hand.

In recent years, the prodigious development of technology for microseismic monitoring of hydraulic fracturing is one example of a technical innovation driven by high commodity prices and the resulting push to produce hydrocarbons from previously inaccessible shales. Early realizations of microseismic monitoring technology provided estimates of the source location of the sounds recorded during the fracturing process. These microseismic events were interpreted to be the result of rock breaking during hydraulic stimulation. Mapping the event locations provided an estimate of the size and shape of the treated rock volume and a visual record of how fractures evolved. As this technology matured, it became clear that an adequate deployment of geophones also could estimate the nature of the rock break (i.e., the event’s focal mechanism). The focal mechanism describes the size and orientation of the break and the direction of slip.

The event’s moment tensor is the mathematical representation of the source mechanism. The established process for estimating the moment tensor of an event presents several challenges. First, an adequate sampling of the event wavefront is required to map the phase and amplitude distribution of the signal over a significant fraction of the focal sphere surrounding the event. Second, the event first-arrival P-wave amplitude must be hand-picked to determine the phase and amplitude spatial distribution, which is a time-consuming and error-prone process. Also, the process of solving for all of the components of the
moment tensor can be unstable and nonunique.

Most estimates of focal mechanisms are overly simplified because they evaluate only the focal mechanism of the largest events, limiting the solution to double-couple (shear only) mechanisms and limiting the number of distinct mechanism types to a small number (usually three or less). Even using this simplified method, the effort required for the process of estimating focal mechanisms makes real-time delivery of the analysis challenging. Yet the need for such real-time analysis is becoming increasingly important in today’s environment.

Recently, MicroSeismic Inc. developed a method for analyzing focal mechanisms in real time using an automatic moment tensor inversion to quickly calculate event moment tensors. This automated process uses a cross-correlation technique to determine the station-by-station relative amplitude and phase of each event in comparison to an estimate of the source function derived from a linear inversion of the data. This replaces the laborious method of manually hand-picking P-wave arrival amplitudes and allows for a unique focal mechanism estimate for every recorded event. This provides a more accurate representation of the discrete fracture network geometry and a more robust event catalog. An uncertainty in the estimates also is available, allowing for spurious solutions to be rejected. The result is a more complete and accurate event catalog delivered in real time, enabling important completion decisions to be made while they are still relevant.

Such real-time results enable optimal completion of each stage, more complete refracturing jobs along the entire wellbore, detection of geohazards before they cause wasteful pumping or even failure of the completion and targeted treatment of fractures of interest. The ability of automatic moment tensor inversion to provide microseismic data in real time can improve these real-time decision-making capabilities.