The effect of gas hydrate saturation and distribution on the geophysical properties of marine sediments

Abstract

Gas hydrates are ice–like compounds found in marine sediments and permafrosts. A significant fraction of all known hydrocarbons in nature is in the form of hydrate. Gas hydrates are a potential energy resource, with possible roles in seafloor slope stability and climate change. As such, improved geophysical methods are needed to identify and quantify in situ natural hydrates to better study their potential impacts. Current estimates of the distribution and volume of gas hydrates vary widely, by orders of magnitude, largely because of uncertainties in geophysical inversion results. The presence of hydrate affects the geophysical properties of the host sediment, creating anomalies that can be detected by seismic or electrical methods measurements. However, the precise relationships between measured geophysical properties and hydrate content (and distribution) are not fully understood, leading to uncertainties in hydrate estimates. Previous studies have shown that both the hydrate saturation (content) and its distribution (morphology or habit) affect the geophysical properties of the host sediment, and separating these effects presents a challenge to geophysical data interpretation. As this knowledge is generally required to interpret field data, this thesis instead seeks to gain this understanding from controlled laboratory experimental studies.
I studied laboratory hydrate formation and dissociation in Berea sandstone and Leighton Buzzard sand to understand their effect on P- and S-wave velocities and attenuations, and on electrical resistivity. I used high resolution synchrotron radiation X-ray tomography (SRXCT) to visualize the pore-scale evolution of hydrate morphology with saturation. These observations could be important for seismic data interpretation in terms of hydrate content and sediment strength, which are needed for natural resource and geohazard assessments (also for joint seismic and electromagnetic survey data interpretation). Hence, I was able to observe how hydrate distribution within the pores (morphology or habit) changes with hydrate formation and dissociation, and how these changes affect the P- and S-wave velocities and attenuations. I calculated hydrate saturation continuously from changes in pressure and temperature and independently from electrical resistivity during hydrate formation and dissociation. I applied a new rock physics model to relate P- and S-wave velocities and attenuations with changes in hydrate saturation and morphology.
I found that not all the gas formed hydrate, even when the system was under hydrate stability conditions with excess water. The synchrotron CT results suggest that the dominant mechanism for co-existing gas is the formation of hydrate films on gas bubbles; these bubbles either rupture, releasing trapped gas, or remain trapped within an aggregate of hydrate grains. From a geophysical remote sensing perspective, such co-existing gas could cause errors in hydrate saturation estimates from electrical resistivity as both gas and hydrate are resistive compared to saline pore fluid. I saw that hydrate starts forming in the pore-floating morphology (where hydrate grains are surrounded by brine) and evolves into the pore-bridging morphology (where hydrate connects mineral grains). Eventually, hydrate from adjacent pores joins and forms a pore hydrate framework, interlocking with the sand grain framework and separated by thin water films. I was able to relate these changes in morphology to our elastic wave measurements using the HBES (Hydrate Bearing Effective Sediment) rock physics model. For low hydrate saturations, both P and S wave velocity follows the pore-floating model curve. As hydrate formation continues, the P-wave velocity follows the pore-bridging model curve, similar to other studies. In contrast, the S-wave velocity was lower than the pore-bridging model but higher than the pore-floating model curves. I think that the presence of water films between hydrate and the rock frame inhibited the ability of pore-bridging hydrate to increase the frame shear modulus. The higher S-wave velocity than the pore-floating model predictions is likely due to interlocking rock and pore-bridging hydrate frameworks. The magnitude of relative changes in attenuation is much higher than that of velocity due to changes in hydrate content and distribution. Elastic wave attenuation frequency spectra between 448 and 782 kHz show systematic and repeatable changes during hydrate formation and dissociation. In our experiments, the dominant mechanism of attenuation and velocity changes with an increase in hydrate saturation is (i) a decrease in methane gas bubble radius and (ii) an increase in secondary porosity with hydrate formation. The accurate measurement of both velocity and attenuation at multiple frequencies in the pulse-echo system allow us to constrain the dominant attenuation mechanisms using the HBES rock physics model.
Overall, I conclude that hydrate-sediment systems are complex with interlocking solid hydrate aggregate and host grain frameworks separated by water films, with isolated pockets of gas within the hydrate. Such an interlocking pore hydrate framework and co-existing gas, if widespread in nature, should be considered in hydrate quantification from elastic wave velocities. For more reliable estimates of in situ hydrate, multiple geophysical parameter measurements are required (e.g., P and S wave velocities and attenuation, electrical resistivity, and at multiple frequencies), and hydrate estimates from seismic velocities alone could lead to significant errors at low hydrate saturations (< 40%).

More information

Identifiers

Catalogue record

Export record