Fracture Dynamics and Its Induced Seismicity
Fracture Dynamics and Its Induced Seismicity
Embedded Files
Fractures significantly impact the hydraulic properties of rocks, often serving as preferential pathways for fluids. Therefore, fracture characterization is essential for determining the scale of fluid transport in geological settings (e.g., estimating the volume of gases that escape from magma to the surface). Because fractures play a dominant role in controlling fluid flow within impermeable bedrock aquifers, volcanic systems, and petroleum reservoirs, understanding their dynamics is critical.
To investigate fracture behavior and the resulting seismicity, Dr. Askari’s team has developed several experimental apparatuses in the Physical Modeling Laboratory (PML) at Michigan Tech. In particular, the team studies a unique seismic mode known as the Krauklis wave, which is generated within fluid-filled fractures. Since this wave originates from fracture resonance, it can be used to estimate fracture size, stiffness, and the nature of the contained fluids, providing a powerful tool for linking laboratory observations to field-scale hydrogeological and volcanic processes.
We have developed a novel experimental framework that leverages photoelasticity and high-speed imaging to investigate fracture dynamics. Using a custom-built high-speed polariscope, we visualize and quantify the propagation of stress waves and fracture-induced deformations in real time, capturing transient stress distributions that conventional static imaging cannot resolve. This approach combines optical phase retardation measurements with precisely controlled boundary conditions to provide a physically grounded visualization of stress transmission in elastic media. The integration of high-speed imaging and advanced photoelastic analysis enables us to observe how internal pressure variations and mechanical interactions evolve within confined geometries, offering valuable insights into dynamic fracture processes and elastic wave behavior relevant to geophysical systems.
In Ray et al. (2025), we developed a controlled-source physical model to investigate the origin of long-period (LP) seismic events generated within fluid-filled cracks. The model, consisting of a 30 cm × 15 cm × 0.2 cm crack embedded in a 3 m × 3 m × 0.24 m concrete slab, enabled us to examine the influence of crack stiffness, fluid density and viscosity, radiation patterns, and trigger location on LP signals. Our results confirm the theoretical Krauklis wave model, showing that higher fluid density and viscosity reduce resonance frequency and quality factor (Q). At the same time, trigger location determines the excitation of transverse modes. Pressure and surface sensors recorded identical spectral features, demonstrating that surface instruments can capture LP-like crack waves. This study bridges the gap between theoretical models and observed LP seismicity, offering new insight into the role of fluid-filled cracks in volcanic and hydrothermal systems.
(a) Waveforms collected using a pressure transducer at the fifth position. (b)–(f) Waveforms obtained from the surface sensors, where the four panels on top of each other define the trigger locations as edge 1, edge 2, middle, and one-fourth, respectively. (g) The layout used in this experiment. From Ray et al. (2025).
In Ray et al. (2024), we demonstrated that fluid flow leaves a measurable signature on Krauklis waves, slow, resonant seismic waves that occur along fluid-filled fractures and are often linked to volcanic long-period (LP) events. Using a custom aluminum tri-layer apparatus with controlled water, oil, and polymer solutions, we found that increasing the flow rate enhances the phase velocity, resonance frequency, and quality factor of the Krauklis wave, whereas flow direction influences attenuation and waveform distortion. These results confirm that Krauklis waves can reveal subsurface fluid motion, offering a novel path to characterize hydrothermal and magmatic flow processes through seismic data.
Normalized amplitude spectra related to the flow rate conditions (a) no flow, (b) 12 ml/min (41.2% volume fraction/min), (c) 14 ml/min (48% volume fraction/ min), and (d) 16 ml/min (54.9% volume fraction/min) respectively, where the fluid used inside of the crack is water. (e) Comparison between the frequency spectrum of different flow rate conditions with respect to the no-flow condition, where the “black,” “red,” “green”, and “blue” lines in the comparison figure represent no‐flow, 12, 14, and 16 ml/min flow rates, respectively. From Ray et al. (2024).
In Cao et al. 1, we studied the fundamental parameters affecting dispersion, dissipation, and resonance properties of the Krauklis waves, such as fluid viscosity, fracture geometry, fracture compliance, and stiffness ratio, to enhance the interpretation of the Krauklis wave for fracture characterization. We developed an experimental apparatus to study the Krauklis wave within a trilayer model consisting of a pair of aluminum plates and a mediating viscous fluid layer. We utilized a piezoelectric source and miniature pressure transducers in our measurements. Through comprehensive experiments, we noted that the phase velocity, resonant frequencies, and quality factors of the Krauklis wave (1) increase with the expansion of the fracture aperture and (2) decrease with the increase of the fluid viscosity. Additionally, (3) phase velocity, resonant frequencies, and attenuation decrease with the increase of mechanical compliance. In addition, we found out that fracture geometry can significantly affect the Krauklis wave as in two modeled wedge-shaped and rough fractures, the phase velocity of the Krauklis wave decreased.
The Sompi analysis of a Krauklis wave record the fracture apertures of (a) 1.5 mm, (b) 2.5 mm, (c) 4.5 mm, and (d) 8.5 mm (from Cao et al. 1).
Visualization of the Krauklis wave propagation with a trilayer model using the dynamic photoelasticity experimental apparatus (from Cao et al. 2).
In Cao et al. 2, we developed an optical apparatus based on the dynamic photoelasticity technique to visualize the propagation of the Krauklis wave within fluid-field fractures. Using this apparatus, we visualized and analyzed the propagation of the Krauklis wave within an analog fluid-filled fracture. We demonstrated that the dynamic photoelasticity technique can quantitatively describe seismic wave propagation with a quality similar to experiments using conventional transducers (receivers) while additionally visualizing the seismic stress field. We also determine the capability of the method to analyze seismic data in the case of complex geometry by modeling a saw-tooth fracture. We showed that the fracture’s geometry could strongly affect the characteristics of the Krauklis wave as we noted a higher Krauklis wave velocity for the saw-tooth case, as well as greater perturbation of the stress field.
In Cao et al. 3, we focused on the dynamics of fluid-filled fractures using the dynamic photoelasticity technique. We studied two phenomena stemming from the solid-fluid interaction that can significantly fracture dynamics. The first phenomenon is cavitation (also called bubble nucleation in literature), which is initiated when the pressure fluid within a fracture suddenly drops below its vapor pressure, forming the nucleation of vapor bubbles. Cavitation has a significant effect on fracture dynamics. We showed that fracture geometry can contribute to cavitation when the fluid viscosity is low. Another phenomenon pertinent to fracture dynamics is fracture opening due to an increase in fluid pressure. We demonstrated that bubbles accumulated at the tip of a fracture can enhance fracture opening.
In an internationally collaborative project, we studied the deformation of fractures and vugs in carbonate rocks due to changes in the effective stress by conducting a series of loading-unloading experiments (up to 20 MPa effective stress) using a high-resolution 3D X-ray computed tomography technique. We showed that after loading, the porosity of carbonate rocks decreased exponentially, followed by an increase during unloading; however, it did not recover to its initial value.
Currently, Dr. Askari's team is working on two NSF-funded projects pertinent to fracture dynamics and its induced seismicity. In one research3, we have developed a large-scale apparatus in a laboratory-controlled condition to simulate long period (LP) events that are observed in magmatic settings. Our results will improve the interpretation of LP events to enhance estimating the scale of magma transports in volcanoes. In another research, we are developing a novel remote sensing technique based on the enhanced Moiré technique to record seismological signals at locations close to volcanic craters. Those seismological signals can provide crucial information about gas and magma through the shallow crust.
(a-g) Image sequence of the bubbles expanding and collapsing for a saw-tooth crack model visualized by the photoelasticity apparatus. (h) The tracked pixel brightness intensity at the saw-tooth location indicated by a white arrow in (a–g). From Cao et al. 3.
3-D visualizations of (a) digital rock and corresponding (b) pore network model. In (a), the blue and blue areas correspond to the pore space and the matrix respectively. In (b), blue spheres are the pores and white tubes are the throats (from Yang et al.).
Page updated
Google Sites
Report abuse