The US Geological Survey (USGS), in response to potential geohazards, energy resource potential, and climate issues associated with marine gas hydrates, has developed a laboratory research system that permits hydrate genesis and dissociation under deep-sea conditions, employing user-selected sediment types and pore fluids.
The apparatus, GHASTI (gas hydrate and sediment test laboratory instrument), provides a means to link field studies and theory and serves as a tool to improve gas hydrate recognition and assessment, using remote sensing techniques.
GHASTLI's use was proven in an exploration well project led by the Geological Survey of Canada and the Japanese National Oil Corp., collaborating with Japan Petroleum Exploration Co. and the USGS. The site was in the Mackenzie Delta region of the Northwest Territories (Mallik 2L-38 drillsite).
From tests on natural methane hydrate-bearing sand recovered at about 1,000 m subsurface, the in situ quantity of hydrate was estimated from acoustic properties, and a substantial increase in shear strength due to the presence of the hydrate was measured.1 2
GHASTI can mimic a wide range of geologic settings and processes. Initial goals involve improved recognition and mapping of gas hydrate-bearing sediments, understanding factors that control the occurrence and concentration of gas hydrates, knowledge of hydrate's significance to slope failure and foundation problems, and analysis of gas hydrate's potential use as an energy resource.
Gas hydrates
Gas hydrate is a crystalline form of water, like ice, but has an expanded lattice that holds gas molecules. Gas can be highly concentrated in this lattice; for example, 1 scf of methane hydrate, can release more than 160 scf of gas.
Methane hydrate is by far the most commonly occurring form of hydrate. The pressure and temperature combinations that favor its formation are widespread, as are its ingredients. Consequently, it is present in vast quantities in ocean sediments, especially at depths deeper than 500 m, and is relatively commonplace beneath permafrost in the Arctic.
Despite its apparent abundance, our ignorance of the processes and effects associated with methane hydrate is profound. Why are these hydrates concentrated in certain places? How would one prospect for them or determine their amount? Will they "melt," release gas, and cause foundation problems for offshore platforms, or could their presence improve foundation support? Would a rapid, massive release of methane (a potent greenhouse gas) have an impact on climate?
Since they were first recognized in marine sediments more than a quarter of a century ago,3 about 20 holes in the Deep Sea Drilling Project and Offshore Drilling Program produced one or more samples containing gas hydrate (Reference 4 contains a database on the holes). Also, gas hydrate-bearing strata have been widely recognized in seismic reflection profiles and subsequently mapped in a few locations.5 6
There has also been substantial progress in understanding the properties of pure gas hydrates. Theoretical and experimental work has been conducted by many investigators since the 1940s.7 Little work, however, has been done to establish relationships between remotely sensed data, such as seismic profiles, and properties of hydrate-bearing sediment, critical knowledge for exploration or engineering.
Similarly, knowledge of relationships between gas hydrates, hydrate processes, sediment properties and geological processes is essential for predicting hydrate occurrence and assessment of geohazards.
GHASTI system
Fig. 1 shows the basic concept of GHASTLI. Experiments or tests take place within a pressure chamber that holds a cylindrical sediment specimen (Fig. 2). Inside this chamber, pressure systems emulate the effect of water depth (to 2,500 m) and sediment overburden (to hundreds of meters in thickness). Flow of seawater and/or gas through a sediment specimen is precisely controlled and monitored.
GHASTI can operate with natural gas hydrate or specimens with prescribed characteristics (Fig. 3b) and can simultaneously measure a suite of key variables. This article will focus on fluid volume, acoustic velocity-compressional (P-wave) and shear wave (S-wave), electrical resistance, temperature, pore pressure, shear strength, and permeability.
Within the time constraints imposed by laboratory testing, the hydrate-bearing sediment created in this apparatus is virtually identical to its natural counterpart in the manner that it formed.
Simulating natural conditions
GHASTI enables creation and dissociation of compounds in a manner such that "natural" hydrates can be studied under "natural" conditions. This involves simulating pressure, temperature, and sediments, including the solids and liquids contained within the specimen.
Simulating pressure
There are four pressure systems in GHASTI (Fig. 4) that can be accurately controlled (plus or minus 5 kPa or <1 psi).
System No. 1 controls pore-water pressure within the sediment and thus represents hydrostatic pressure, which is the ocean depth plus the additional distance below the seafloor down to the sediment level in question (Fig. 1). Hydrostatic pressure (P) and temperature (T) are the two primary factors that control formation and dissociation of gas hydrates (Fig. 3a).
GHASTI can create pore-water pressure equivalent to that of a large portion of the sub-seabed environment in which methane hydrate occurs (Fig. 3a).
Sediments buried beneath the seafloor have been compacted as a result of overburden pressure. This is pressure applied by the weight of the overlying sediment through direct grain-to-grain contact and is therefore generally independent of pore-water pressure.
GHASTI applies this overburden through its confining pressure system (System No. 2, Fig. 4), which "squeezes" the sediment specimen. Up to several hundred meters of burial may be simulated in this manner.
System No. 3 controls the gas entry pressure, which is the pressure needed to push gas into the sediment and displace the pore water (seawater) so that the ingredients to make a gas hydrate can be brought together. Back pressure (System No. 4), in combination with seawater or gas pressure, can create a pressure gradient, and therefore induce fluid flow between ends of the specimen.
Simulating temperature
Temperature is the other main environmental factor that controls the formation and dissociation of hydrate. Fig. 4 shows the GHASTI heat-exchange system. Most natural methane hydrates exist in the temperature range of 2-22° C.
Methane hydrate is formed in GHASTI by lowering temperature and/or raising pressure such that the specimen pressure-temperature (P-T) conditions are within the methane hydrate P-T stability field (Fig. 3a). At an appropriate pressure, a cooling front at the top of the specimen is initiated (Fig. 2) and the entire specimen is progressively moved into the stability zone.
Alternatively, pressure can be appropriately increased, and hydrate nucleation may begin anywhere in the specimen. Table 1 lists the environmental controls and provides information on how those controls relate to geologic setting.
Host sediment options
Sediment specimens are fabricated with natural sands, silts, and clays or artificial grains. Also, core samples can be used, such as those recovered from the Ocean Drilling Program. Sediment properties, such as grain size, and geo technical index properties, such as porosity, can be preset, depending on the experiment's objectives.
Moreover, a specimen's permeability, pore size distribution, and other properties can be determined before experiments begin. Seawater or other liquid can be used to saturate the sediment. A specimen can also be constructed to have certain properties or simulate specific characteristics, such as fissures.
Measurements
To characterize gas hydrate in sediment, one must know hydrate volume, how much methane is actually contained in the hydrate, what form it is in (grains, layers, etc.), and where and how it is distributed. Fig. 5 shows the sensor-data acquisition system. The key variables include fluid volumes, temperature, electrical resistance, acoustic properties, triaxial shear, strength, and permeability.
Fluid volumes
The amount of gas in the hydrate is known by monitoring seawater and gas volumes during hydrate formation. Because the volume of gas pushed into the specimen and the volume of both gas and pore water that passed out into the collector are known (to approximately equal to 1 ml), one can calculate the methane available for hydrate formation.
The change in confining fluid volume gives an indication of hydrate volume. This volume roughly corresponds to a change in total specimen volume, which results from growth or dissociation of hydrate.
Temperature
The temperature data provide an approximation of the distribution and amount of hydrate because formation of methane hydrate involves an exothermic reaction. Heat is given off as a hydrate grows and this heat registers as a thermal anomaly within the array of temperature sensors (Fig. 6 inset).
Plots of temperature vs. time (Fig. 6) show three sequential anomalies as hydrate forms from the top down. The temperature trace of Thermocouple No. 4 shows no anomaly, signifying that hydrate did not grow throughout the entire specimen or was relatively sparse in its lowermost portion.
Electrical resistance
A specimen's electrical properties also may provide an indication of hydrate volume, show its distribution, and possibly infer its mode of occurrence. The gross difference in resistivity between seawater and hydrate (greater than or equal to two orders of magnitude) provides an image of the sediment areas occupied by hydrate.
Frequent measurements allow a time-lapse view of changing resistances, which can be plotted in the form of contour maps (Fig. 7).
Acoustic properties
Knowledge of changes in acoustic properties caused by gas hydrate will foster development of seismic reflection profiling as a dependable tool to map and determine the concentration of gas hydrate in sediment.
GHASTLI's design allows correlation of laboratory and shipboard data, and because it can accommodate downhole samples, it lends itself to correlation with logging data as well. Furthermore, because GHASTI can sustain a thermal gradient, the phase boundary can exist within a specimen, which makes possible studies of acoustic phenomena at the phase boundary.
Fig. 8 shows the acoustic setup, and also presents an example of acoustic data output from a Mackenzie Delta natural hydrate test.
Because velocities for each sample are recorded with and without methane hydrate, the effect on velocity is easily determined. In this example, the P-wave velocity was 2,370 m/sec and the hydrate volume was about 60% of the pore space.
Triaxial-shear strength
The triaxial strength subsystem (Fig. 9) measures stress-strain relationships and shear strength, which are an integral part of foundation analysis and slope stability calculations. These measurements may also yield valuable input for acoustic models.
Due to the wide range of rates accommodated by the syringe pumps and load cells, GHASTI can study a wide range of sediment types from very soft, wet mud to hydrate-bearing sands (Figs. 3 and 9).
Fig. 9 shows the stress-strain curves of hydrate-bearing and nonhydrate-bearing specimens from the Mackenzie Delta study. There is at least a four-fold increase in the strength due to the presence of hydrate in the sands.
Permeability
Measuring the ease with which fluid can move through hydrate-bearing sediment may reveal hydrate habit (cement? loose grains?) and provide information on the progressive capacity of a hydrate to act as a seal or trap. Knowledge of permeability changes caused by hydrate dissociation and the flow of liberated gas may be important for gas production and predicting changes in sediment strength parameters.
Mackenzie Delta samples were subjected to permeability tests and demonstrated a consistent hydraulic conductivity in hydrate-free specimens. In contrast, hydrate-bearing sands showed a considerable range in permeability, from which we inferred that fissures or cracks might be present in the system.
Geologic processes, human activities
Table 2 shows how GHASTI would simulate some common, hydrate-related geological or human-induced processes that could occur on continental slopes and rises, where most gas hydrate exists.
Natural processes include sea level change, uplift and subsidence, erosion and deposition, mass movement (rapid removal of overburden), mass deposition (rapid loading), change in geothermal gradient, and injection of warm, gas-rich fluids into faults or fracture zones.
GHASTI can also simulate foundation loads, downhole pressure reduction, rising well fluids, and other characteristics of offshore oil and gas operations. The data from the experimental work could apply directly to these and other engineering considerations.
Acknowledgments
USGS personnel who have contributed to GHASTLI's progress include Dave Mason (head technician), Ray Davis, Chris Green, and Monica Relle. Initial funding for the purchase and development of the GHASTI system was provided by the Federal Energy Technology Center of the US Department of Energy; we specifically acknowledge the support of Rodney Malone of the FETC.
Ingo Pecher (University of Texas at Austin), and Ben Clennell (University of Leeds, UK) have made major contributions to the development of GHASTLI.
References
- Pecher, I.A., Booth, J.S., and Winters, W.J., "Gas hydrate distribution in sands results from seismic laboratory studies," American Geophysical Union, Eos supplement, Vol. 80, No. 17, 1999, p. S338.
- Winters, W.J., Wright, J.F., Booth, J.S., Dallimore, S.R., Pecher, I.A., Mason, D.H., Relle, M.K., and Dillon, W.P., "Laboratory tests of gas hydrate samples from the Mackenzie Delta," American Geophysical Union, Eos supplement, Vol. 80, No. 17, 1999, p. S338.
- Stoll, R.D., Ewing, J., and Bryan, G.M., "Anomalous wave velocities in sediments containing gas hydrates," Journal of Geophysical Research, Vol. 76, 1971, p. 2090.
- Booth, J.S., Rowe, M.M., and Fischer, K.M., Offshore gas hydrate sample database, USGS Open-File Report 96-272, 1996.
- Tucholke, B.E., Bryan, G.M., and J.I. Ewing, "Gas-hydrate horizons detected in seismic profile data from the western North Atlantic," American Association of Petroleum Geologists Bulletin, Vol. 61, 1977, pp. 698-707.
- Dillon, W.P., Lee, M.W., and Coleman, D.F., Identification of marine hydrates in situ and their distribution off the Atlantic coast of the United States, in Sloan, E.D., Jr., Happel, John, and Hnatow, M.A., (editors), Annals of the New York Academy of Sciences, International Conference on Natural Gas Hydrates, Vol. 715, 1994, p. 364-380.
- Sloan, E.D. Jr., Clathrate Hydrates of Natural Gases, Marcel Dekker, Inc., New York, 1998.
- Englezos, P., and Bishnoi, P.R., "Prediction of gas hydrate formation conditions in aqueous electrolyte solutions," American Institute of Chemical Engineers Journal, Vol. 34, 1988, pp. 1718-21.
- Dickens, G.R., and Quinby-Hunt, M.S., "Methane hydrate stability in seawater," Geophysical Research Letters, Vol. 21, 1994, pp. 2115-18.
The Authors
James S. Booth is a research geologist with the US Geological Survey and has acted as science advisor for GHASTI development and research planning. His GHASTI experiments focus on relationships between gas hydrate-related phenomena and the geomechanical behavior of sediments. He received his PhD in marine geology from the University of Southern California.
William J. Winters is a civil engineer with the USGS at the Woods Hole Center for Coastal & Marine Geology. He has participated in gas-hydrate field programs on the Blake Ridge during Ocean Drilling Program Leg 164 and on the Mackenzie Delta project. He has an ME in geotechnical engineering from Cornell University.
William P. Dillon is project chief of the USGS's gas hydrate project, encompassing seismic studies, geochemical research, and laboratory analyses of the physical characteristics of gas hydrate. He received a PhD from the University of Rhode Island graduate school of oceanography.
