The goal of this project is to determine structural and stratigraphic controls on hydrate occurrence and distribution in Green Canyon (GC) 955 and Walker Ridge (WR) 313 blocks with special emphasis on hydrate-bearing sand reservoirs. Structural and stratigraphic controls on hydrate distribution are examined by jointly analyzing surface-towed, multichannel seismic (MCS) and Ocean Bottom Seismometer (OBS) data and well logs through a combination of pre-stack depth migration (PSDM), travel time and full-waveform inversion (FWI), and rock physics modeling methods.

Oklahoma State University, Stillwater, OK 74078-1026

Vast volumes of hydrate and associated free gas may exist beneath the seafloor along continental margins, making hydrates a potential energy resource, a deep-water geohazard, and a potential component of the carbon cycle. Thus, hydrate research has a multi-dimensional significance.

Geophysical and subsurface investigations have indicated that hydrates are heterogeneously distributed and demonstrate multiple arrangement styles such as freely floating within pore spaces, embedded in the rock matrix, cementing outside mineral grains, creating and filling fractures, and forming massive seafloor outcrops. Hydrates affect elastic velocities of their host sediments. The magnitude of velocity change, however, depends both on the in situ volume and arrangement styles of hydrate.

The most ambitious and successful hydrate research in the Gulf of Mexico (GOM) was carried out in 2009, in the second leg of the Joint Industrial Project (JIP) where several boreholes in WR313 and GC955 were completed using a comprehensive set of logging-while-drilling (LWD) tools. JIP Leg 2 inferred hydrate in at least two arrangement styles: pore-filling in sand-dominated reservoirs and fracture-filling in clay-dominated reservoirs. Presence of hydrate in highly concentrated form within sand-dominated reservoirs was confirmed indicating a high possibility of hydrate recovery with near-future technology.

Researchers currently do not have detailed knowledge of structural and stratigraphic controls on hydrate distribution in the study area, and it is critical to address this knowledge gap to fully appreciate both the feasibility of hydrate extraction from a resource perspective and the environmental impact of perturbing these natural systems. This effort will address this knowledge gap by resolving questions related to heterogeneity of hydrate distribution within sand reservoirs and association of fracture-dominated and sand-dominated reservoirs and hydrate and free-gas interaction, which in turn will enable elucidation of hydrates as an energy source and their significance in natural environments.

Project personnel will use 3-D MCS, 2-D MCS, and OBS data that are co-located with GC955 and WR313 wells to construct high-fidelity depth images using PSDM and high-resolution velocity and attenuation models using tomographic methods such as travel time inversion and FWI. The depth image, the velocity and the attenuation models will be used for detailed interpretation of the study areas to identify the structural and stratigraphic features and facies relevant to hydrate occurrence and distribution. Rock physics models will be used to relate hydrate saturation and growth style to the LWD sonic velocities. The sonic log and the rock physics template will be up-scaled to seismic wavelengths. Hydrate saturation will be extended outwards from the well using the FWI velocity model to create a high-resolution map of hydrate saturation in the study area. The saturation maps will be analyzed in conjunction with previously created structural, stratigraphic, and facies models to understand how, where, and why hydrates form and concentrate in different locations within the study area.

The project will greatly advance the tools and techniques used for delineating specific hydrate prospects. Results from the proposed effort can be incorporated into other DOE-supported projects on modeling the potential productivity and commercial viability of hydrate from sand-dominated reservoirs.

Application of FWI, PSDM, and rock physics-based seismic quantification will greatly increase knowledge of hydrate and free gas systems in the northern GOM in general, and at sites WR313 and GC955 in particular. A combination of reflectivity attributable from the PSDM image and the FWI velocity model will unravel the heterogeneity of hydrate distribution within the sand reservoir. The structural (fault, fractures, etc.) and stratigraphic (pinch-outs, channel-levee architecture, etc.) interpretation of the PSDM image combined with the attenuation from FWI will help establish the relationship between fracture-dominated and sand-dominated environments, including the extent of their inter-fingering. Rock physics-based seismic quantification will help determine the stratigraphic controls on hydrate and free gas distribution.

This effort will provide insight into the geological controls on hydrates and will shed light on fundamental in-situ properties (porosity, pore-interconnectivity, and mineralogy) of hydrate-bearing sediments. These results will be of great interest to industries that view hydrate as either an energy source or a geo-hazard.

The project team created a high-resolution structural image using 3-D seismic data procured from M/S CGG-Veritas Inc.  The navigation quality control and binning of the 3-D data were completed in February 2013 and data processing (i.e., filtering, muting, stacking velocity analysis, and brute stack generation) was completed in July 2013.

Depth imaging, which includes velocity model building and pre-stack depth migration, has been completed. At this stage, the potential BSRs in both WR313 and GC955 blocks have been interpreted. The team attempted FWI with the CGG-Veritas dataset in October 2013 but did not achieve the desired resolution (<100m) due to the scarcity of the 3-D data.

The project team obtained high-resolution 2-D and OBS data from the U.S. Geologic Survey (USGS) in January 2014. As with the CGG-Veritas dataset, the team performed velocity model building using travel time inversion and depth imaging of the high-resolution 2-D data. This effort was completed in March 2014.  The project team then attempted FWI of GC955 OBS data and were successful in obtaining the desired resolution in the velocity model. This effort was completed in October 2014. The project team further generated the attenuation model using FWI. This effort was completed in November 2014. The project team then jointly interpreted the depth-migrated image, the FWI velocity model, and the FWI attenuation model to identify the key features of the stratigraphy relevant to hydrate accumulation. This effort was completed in March 2015. This joint interpretation showed that the difference in hydrate concentration between Wells GCC955-H and GCC955-Q could be due to the relative placement of the two wells within a submarine channel complex. According to their interpretation, it is likely that Well GCC955-H was placed within the axis of this channel complex; explaining the large volume of hydrate inferred in the GCC955-H logs. Likewise, it is likely that Well GCC955-Q was placed within the levee of this channel complex, which lacks the reservoir quality sands; explaining why GCC955-Q logs did not indicate presence of any appreciable hydrate volume.

The project team has been able to develop a good understanding of the stratigraphy along the profile and the structure of the BSR upon completing the 3-D imaging using the CGG-Veritas data. Although depth imaging has been done in 3-D, the tomography must be done in 2-D due to software limitations. A collaboration with USGS was formed to provide project personnel with state-of-art high resolution 2-D MCS and collocated OBS data from WR313 and GC955. The initial tomographic models were completed in July 2014. FWI of GC955 OBS data to produce high-resolution velocity and attenuation models was competed in November 2014. To gain further insight into the hydrate-free gas system in GC955, the project team applied rock physics model to logs from well GC955-H and found out that the hydrate bearing sands are not massive. They likely have a layered stratigraphy with thin clay layers separating the sand sheets. This effort was completed in June 2015. 

The project team attempted to apply the same modeling methodology on data from WR313, but did not get good results in that the velocity perturbations did not explain what we know about the system. Initially the reasons were unclear. After an in-depth investigation of the data quality, the research team found out that the initial processing of the WR313 OBS data had removed the low frequencies that are needed in full-waveform inversion. The project team redid the initial processing to include lower frequencies and tried velocity modeling again. The results were still not encouraging. The research team realized that the OBS data had very low signal-to-noise ratio. Therefore, they attempted a new processing methodology known as the wavefield decomposition where the hydrophone and the vertical geophone data are jointly used to separate the downgoing and the upcoming wavefields. The upcoming wavefield created in this manner is supposed to be free of source reverberations. The full-waveform inversion could only then be attempted on the upgoing wavefield. Wavefield decomposition was adequately accomplished in February 2016.

(May 2016)
The project team is attempting to recreate the results for GC955 to make sure that the new processing gives better results. In the GCC955 the reflection arrival times from key geological horizons for travel time inversion and depth migration have been picked. The travel times have been inverted and a background velocity model has been prepared. The research team is now attempting FWI of the OBS data using the traveltime velocity model as the starting model. The project team expects that the new velocity models will be superior to the velocity models obtained from previous processing. After affirming this, the process will be replicated for WR313.

$304,104

$76,193

NETL – Joseph Renk (Joseph.Renk@netl.doe.gov or 412-386-6406)
Oklahoma State University – Priyank Jaiswal (priyank.jaiswal@okstate.edu or 405-744-6041)

Quarterly Research Performance Progress Report [PDF-1.90MB] January - March, 2016

Quarterly Research Performance Progress Report [PDF-1.08MB] October - December, 2015

Quarterly Research Performance Progress Report [PDF-1.66MB] July - September, 2015

Quarterly Research Performance Progress Report [PDF-1.38MB] April - June, 2015

Quarterly Research Performance Progress Report [PDF-1.46MB]  January - March, 2015

Quarterly Research Performance Progress Report [PDF-1.10MB]  October - December, 2014

Quarterly Research Performance Progress Report [PDF-788KB]  July - September, 2014

Quarterly Research Performance Progress Report [PDF-1.38MB]  October - December, 2013

Quarterly Research Performance Progress Report [PDF-3.07MB]  July - September, 2013

Quarterly Research Performance Progress Report [PDF-5.49MB]  April - June, 2013

Quarterly Research Performance Progress Report [PDF-553KB]  January - March, 2013

Quarterly Research Performance Progress Report [PDF-248KB]  October - December, 2012