A Hybrid Solution Worth its Salt

Earth Explorer March 2015

In many sedimentary basins around the world, ‘seeing through’ thick or steeply-dipping layers of salt is a major challenge for oil and gas exploration, especially in deep offshore environments. Seismic methods alone are usually insufficient to visualize the base of salt and the underlying sediments.

One thing is clear: Exploration success depends on knowing the location of the base of salt. Geophysical surveys help exploration teams evaluate the field, define prospects and determine drill targets. For example, if the geophysics indicates that the field extends farther updip due to the salt geometry, then more drilling may be needed or drill targets may be repositioned, whereas if the field is found to be more limited in the updip direction, a lot of unnecessary and costly deepwater wells can be avoided.

Delineating salt and sub-salt boundaries is a recalcitrant challenge for seismic interpreters. Additional data are often sought to supplement the seismic information, particularly in offshore and deepsea environments.

For years, researchers have worked at incorporating gravity with seismic data. “Since the beginning of modern subsalt petroleum exploration, geophysicists have tried using gravity data to help locate the base of salt in cases where seismic interpretations are uncertain,” says Gerry Connard, Senior Scientist, Potential Field Modeling, at Geosoft. “Typically the initial seismic interpretation is used to build a gravity model, which is adjusted to fit the gravity data and the parts of the seismic interpretation that have high confidence,” he says.

One integrated approach using wave-equation depth imaging and full tensor gradiometry inversion had some success in resolving the base of salt in an area where Kirchhoff depth imaging alone had failed. That attempt was carried out a decade ago at the K-2 Field in Green Canyon Block 562 in the Gulf of Mexico, some 180 miles south of New Orleans. However, as Connard notes, “the published results suggest that the gravity models required intensive manual intervention to accurately represent the seismic interpretations.”

A better modelling approach has eluded researchers. “Numerous exploration examples show that seismic data alone has difficulty mapping the base of salt,” says Rob Ellis, Principal Scientist, Earth Modelling, with Geosoft. “There is no established product for easily integrating gravity data with seismic data to map the base of salt.”

VALEM, (Voxel Assisted Layered Earth Modelling) technology was developed by Geosoft, in collaboration with potential field experts in the energy sector, to address this challenge.

“There have been a number of attempts to incorporate gravity with seismic,” Ellis says, but VALEM is “the first attempt to deliver this incorporation in a commercial, integrated, easy-to-use application specifically focused on the base of salt problem.”

A high-performance computing service for inversion modelling in the cloud, VALEM represents a significant advance in base of salt and subsalt imaging for energy exploration. Tightly integrated with Geosoft’s GM-SYS 3D modelling program, it results in an interactive solution for sub-salt interpretation that resolves the base of salt from gravity data bound by the constraints of seismic modelling.

Research on VALEM started in 2010, and it holds the best promise yet. A number of challenges remain, however; the chief one being how to integrate a seismic interpretation with a gravity one to provide an improved geologic interpretation. Or put another way, can gravity data supplement seismic data in determining the base of salt? “It’s possible that gravity data can help in some situations,” Ellis says, adding that VALEM’s unique approach to inversion modelling enables geophysicists to explore that possibility.

VALEM proposes a ‘hybrid inversion’ solution to the base-of-salt problem. Such an approach takes the best of layered-earth modeling and voxel-based inversion to allow for the full range of geometries associated with salt/subsalt structures. As Connard explains, VALEM advances the modelling technology in several ways. “VALEM models allow the use of three different types of data structures – layers, voxels, and 3D triangulated surfaces – within a single model. Modelers can use the most appropriate representation for each model element and quickly build complicated models.”

VALEM also uses hybrid calculations by combining both frequency-domain and space-domain calculations where appropriate. Says Connard: “Because space-domain calculations are computationally expensive, VALEM performs the space-domain calculations using server clusters in the cloud, providing high-performance computing for the desktop.”

Connard and Ellis, together with Geosoft researchers Tom Popowski and Gaud Pouliquen, recently did some theoretical testing of VALEM. Their findings were presented at a 2014 workshop organized by the Society of Exploration Geophysicists (SEG) and the Chinese Geophysical Society in Chengdu, China; and more recently at Finding Petroleum’s Non-Seismic Geophysics event held in February.

For the presentation, entitled Hybrid strategies for modelling gravity gradient data, the research team used SEAM I Earth model data and hybrid modelling. SEAM (the SEG Advanced Modeling Program) is a partnership between industry and SEG which aims to advance geophysical science and technology through the construction of subsurface models and the generation of synthetic data sets. The SEAM I model used by the Geosoft researchers describes complex salt structures comparable to those found in the Gulf of Mexico. The researchers performed a seismically constrained inversion of Falcon airborne gravity gradient data to map the base of salt in the SEAM model. The results demonstrate the capabilities and potential of VALEM.

“The exercise points to the progress achieved in VALEM while demonstrating some of the challenges that remain,” Connard says. The main challenge in combining voxel and layered models in a hybrid inversion is that voxel inversions generally produce models which have ‘smoothed’ transitions between domains whereas layered earth models have discrete transitions. “To overcome this inconsistency, we developed a voxel method which effectively produces either a target domain (salt) or a host domain (sediments) with sharp transition,” he says. “This is the key to hybrid modelling.”

For optimal results, the use of VALEM in the base of salt scenario requires an interpreted, depth-converted, 3D seismic survey plus a gravity or gravity gradient survey covering the area of interest. Regional models of the surrounding area will help improve the results. The better the quality of the data, the better the outcomes.

Since December 2014, three exploration and production companies and one consultant have been involved in testing VALEM with real project data from salt basins around the world.

These tests are proving out the application of the hybrid inversion approach in a realistic environment, says S. Bruce Kohrn, Energy Segment Lead at Geosoft. “VALEM is the result of five years of research and development, and the application is now ready for release,” he says, adding that he anticipates full commercial roll-out by mid-2015.

Longer term, VALEM applications likely won’t be restricted to base of salt determinations. Other potential applications include basin analysis and sub-basalt imaging to support subsurface investigations. “In fact,” says Ellis, “there is potential to apply VALEM inversion modelling in any situation where a discrete contrast is required and there is a mixture of surfaces/layers and compact bodies.”

View story online including illustrations and examples:

A Hybrid Solution Worth its Salt

Copyright 2017 Graham Chandler | All Rights Reserved | Contact | Site by Element Media Ltd