Identifying Facies Using Microseismic Analysis

Researchers from the University of Calgary found just what they weren’t looking for when they monitored microseismic signals from an unconventional resource unit in the Hoadley field in central Alberta.

Their work produced a new advance in reservoir characterization using microseismic data.

Hoadley, discovered in 1977, is a giant gas-condensate field. Its main reservoir lies in the Glauconitic formation of the Lower Cretaceous Upper Mannville group, a shallow-marine and nonmarine clastic wedge.

“We’ve been collecting our own data on Hoadley and publishing (research papers) on that. In this particular instance it was a tight sand reservoir, maybe more heterogeneous than a shale reservoir,” said David Eaton.

Eaton is a professor of geophysics and the Natural Sciences and Engineering Research Council of Canada/Chevron industrial research chair in microseismic system dynamics in the geoscience department at the University of Calgary, and an AAPG associate member.

He was a co-author of the 2016 paper describing the new approach, “Reservoir characterization using microseismic facies analysis integrated with surface seismic attributes,” published in Interpretation, the peer-reviewed journal issued jointly by AAPG and the Society of Exploration Geophysicists.

Other authors were Per Pedersen, a petroleum geologist in the department, and students Aamir Rafiq and Adrienne McDougall.

Seismic Serendipity

It was Rafiq’s interest that ultimately led to the new approach.

“What he was really interested in doing was partitioning the reservoir,” Eaton explained. “That sort of evolved into what we’re calling ‘microseismic facies analysis.’”

The university set up its own equipment to monitor multistage hydraulic fracturing treatments at the tight-sands unit in September 2012. Acquisition of continuous passive seismic data continued for more than 10 months during flowback and production.

Please log in to read the full article

Researchers from the University of Calgary found just what they weren’t looking for when they monitored microseismic signals from an unconventional resource unit in the Hoadley field in central Alberta.

Their work produced a new advance in reservoir characterization using microseismic data.

Hoadley, discovered in 1977, is a giant gas-condensate field. Its main reservoir lies in the Glauconitic formation of the Lower Cretaceous Upper Mannville group, a shallow-marine and nonmarine clastic wedge.

“We’ve been collecting our own data on Hoadley and publishing (research papers) on that. In this particular instance it was a tight sand reservoir, maybe more heterogeneous than a shale reservoir,” said David Eaton.

Eaton is a professor of geophysics and the Natural Sciences and Engineering Research Council of Canada/Chevron industrial research chair in microseismic system dynamics in the geoscience department at the University of Calgary, and an AAPG associate member.

He was a co-author of the 2016 paper describing the new approach, “Reservoir characterization using microseismic facies analysis integrated with surface seismic attributes,” published in Interpretation, the peer-reviewed journal issued jointly by AAPG and the Society of Exploration Geophysicists.

Other authors were Per Pedersen, a petroleum geologist in the department, and students Aamir Rafiq and Adrienne McDougall.

Seismic Serendipity

It was Rafiq’s interest that ultimately led to the new approach.

“What he was really interested in doing was partitioning the reservoir,” Eaton explained. “That sort of evolved into what we’re calling ‘microseismic facies analysis.’”

The university set up its own equipment to monitor multistage hydraulic fracturing treatments at the tight-sands unit in September 2012. Acquisition of continuous passive seismic data continued for more than 10 months during flowback and production.

ConocoPhillips was a sponsor for the project, hoping to learn more about the results of fracturing and production in the unconventional reservoir, Eaton said.

But the researchers got more than the usual amount of seismic input.

“When we were monitoring that, we noticed that we were picking up a big uptick in data. It turned out we were picking up signals from a 3-D seismic survey right on top of us,” Eaton said.

And when someone does a 3-D seismic shoot where you have a microseismic monitoring project, you’re going to notice.

“We were getting many thousands of detections,” Eaton noted.

It turned out to be serendipity. When the researchers developed their theories about microseismic facies analysis, they had the 3-D survey results as another view and a sort of reality check.

“The 3-D seismic was very important to help us validate the concepts,” Eaton said.

The researchers came out with a new method of partitioning an unconventional reservoir into distinct facies units, based on their microseismic response.

And enhanced, microseismic-based input to help direct field development, the original target of the research project?

Not so much.

“We saw a lot of interesting phenomena. But we didn’t actually see what we were looking for,” Eaton said.

The Experiment

The Hoadley Flowback Microseismic Experiment began with four main scientific objectives:

  • Undertaking real-time monitoring of an open-hole multistage fracture treatment in the Glauconitic member;
  • Performing long-term monitoring of post-frac microseismicity during flow-back and production;
  • Developing a geomechanical model for that activity; and
  • Integrating the analysis of microseismic data with ancillary studies, including rate-transient analysis and production analysis.

Monitoring equipment for the project consisted of a 12-sensor retrievable array of geophones installed at the end of a multiconductor wireline, in a vertical observation well between two horizontal treatment wells.

In each well, completion included 12 fracturing injection stages. A total of 1,660 microseismic events were recorded and located during the 24-stage treatment, including 259 postpumping events.

Surprisingly, only about half of the microseismic events were oriented in the direction of maximum horizontal stress orientation, or SHmax. The other events were oblique to SHmax and were inferred to represent reactivation of pre-existing fractures.

“Those monitored events moving back toward the well we interpreted as fracture closures,” Eaton noted.

The researchers classified the seismic events into clusters and then computed several microseismic attributes, including azimuth, net seismic moment, length and duration. Based on an interactive clustering approach, 94 percent of the microseismic events were contained within 17 event clusters.

Cluster and attribute analysis then became the basis of facies identification.

Among the set of calculated attributes, incoherence, most-positive curvature and most-negative curvature provided the most clearly interpretable association with the microseismic signals, the researchers found.

“We’re still trying to explore the microseismic attributes that could be used. We’re working on a number of datasets, not just from Canada, but also from the U.S.,” Eaton said.

Even though the two horizontal wells were treated the same in stimulation, there were significant differences in microseismic response between the wells. One showed more postpumping activity and an inferred higher density of event clusters misaligned with the stress field.

Because of the differences, the researchers interpreted the reservoir as compartmentalized, with the two wells intersecting distinct facies with varying rock fabric.

“Taken together, attribute analysis, magnitude statistics and well log analysis are integrated here to characterize reservoir heterogeneity, rock fabric and compartments in the reservoir.

“These reservoir characteristics may, in turn, reflect variations in depositional environment, structural deformation and lithofacies,” the researchers concluded in their paper.

In unconventional plays, microseismic monitoring commonly has been associated with high-definition imaging of microseismic events and hydraulic fracturing results. But that has grown into numerous other uses, including reservoir characterization.

Getting More Out of the Data

Facies identification using seismic and even microseismic data isn’t something new in the world, Eaton acknowledged. An earlier Barnett shale project plotted microseismic event locations over a 3-D geological model to establish the relationship between hydraulic fracturing and facies.

“What our work is highlighting is a different strategy, where they get more out of the microseismic data. It’s drawing on work done, really, decades ago, using regular seismic,” Eaton observed.

The University of Calgary research developed a new approach using different tools, like its approach to cluster identification and attribute analysis, and in some cases tools used differently.

“I think it’s new and kind of unique to our group at this time,” Eaton said, “until the world discovers it and everyone starts using it.”