Electronic Thesis and Dissertation Repository

Thesis Format

Monograph

Degree

Master of Engineering Science

Program

Civil and Environmental Engineering

Supervisor

Bing Li

Abstract

This thesis quantitatively investigates the waveform similarity of micro-earthquakes detected during the Tony Creek Dual Microseismic Experiment (ToC2ME) located in Fox Creek, Alberta. Subsurface anthropogenic activities, especially processes involving injection, can lead to accumulation of porepressure and elastic strain in the geologic formations and rocks, resulting in induced earthquakes. Four hydraulic-fracturing treatment wells were drilled and completed in the study area. The microseismic acquisition system monitored continuous seismic waveforms from 25 October to 15 December 2016 revealing 21619 hydraulic fracture-related events with magnitude ranging from Mw -1 to 3.2. The present study compares and discusses similarity of waveforms captured by 69 stations using two waveform similarity clustering algorithms: 1) Clusty and 2) PageRank, to form clusters of earthquakes with similar waveform content. The performance of this clustering is validated by visual inspection of these waveforms and their spatial and temporal distribution. These clusters are analyzed in terms of their b values, nearest neighbour statistics, and corresponding spanning trees. A porepressure model is also developed to understand the spatial and temporal distribution of porepressure throughout the injection process. This model was then used to identify the mechanisms of different clusters as it relates to the hydraulic fracturing program.

These analysis suggest earthquakes at the site can be attributed to two main sources. Clusters in lower zone #1, zone #7, and zone #5 are likely induced by the activation of pre-existing faults, while clusters in zone #2 and #4 tend to relate to fractures directly induced by the hydraulic fracturing process. Waveforms relating to hydraulic fractures tend to exhibit shorter P-wave packets compared to the activation of natural fractures. These results could be applied at hydraulic fracturing sites such as unconventional oil&gas or geothermal energy extraction to gain insight into the size and shape of the stimulated reservoir.

Summary for Lay Audience

The large number of earthquakes that occurred at the Tony Creek Dual Microseismic Experiment (ToC2ME) site are considered to have a causal relationship with the hydraulic fracturing that was conducted at the site. The underground fluid injection causes a transient pressure change that makes the local faults and fractures in rocks unstable and leads to earthquakes. There are four injection wells completed at the ToC2ME site between 25 October and 15 December 2016, accompanied by 21619 events (with magnitudes between - 1and 3.2). In this thesis, two clustering methods 1) Clusty and 2) PageRank are applied on the dataset, to evaluate the waveform similarity of earthquakes detected by the 69-station monitoring system. The clustering results are assessed by waveforms and the location of the clusters. The clusters are then analyzed using seismic parameters such as b value, nearest neighbour statistics, and corresponding spanning trees. Meanwhile, a porepressure model is also developed to understand the interaction between the occurrence of earthquakes and injection process.

Results show earthquakes at ToC2ME site can by classified into two types. Events of clusters away from the injection wells are related to slip on pre-existing faults, while clusters close to the injection wells are triggered by new fractures created by hydraulic fracturing.

For waveforms of events on pre-existing faults, the length of P- wave is shorter than events triggered by hydraulic fracturing. This result gives insight into the mechanisms of human- induced earthquakes at hydraulic fracturing sites such as unconventional oil&gas or geothermal energy extraction.

Download

Share

COinS