Claudio Paschoa discusses visualization technology with representatives of TGS, Petroleum Geo-Services, and Schlumberger.
Test image of Petrel structural modeling. Photo from Schlumberger.
Visualization technology is used to interpret, through 2D and 3D imaging, complex subsurface geological structures, view fluid movement pathways in hydrocarbon bearing reservoirs, optimize placement of wellbores and visualize surface and subsea facility design implications. Visualization technology was used by the oil and gas industry in the 1980s and 1990s, through dedicated interpretation workstations. By the 2000s, enhanced visualization technologies were periodically introduced, including hardware and software. Geoscientists needed new and effective ways to explore the large quantities of seismic data showing geologic formations and also showing well data rock properties that are acquired through seismic surveys. The reservoir models, traditionally characterized by 2D maps and cross-sections, became fully populated 3D earth models with geological, geophysical, and petrophysical properties, all captured within a 3D visualization environment. Visualization and depth imaging of 3D seismic data are some of the key elements of a rapid technological and qualitative evolution in the remote sensing of the subsurface.
The main aim for interpreters is to better understand the structural content of their datasets, incorporating all the available information in one integrated display. 3D visualization data processing requires massive computing power, due to the volume sizes, particularly pre-stack, which typically involves orders of magnitude greater than standard data viewing. However it is now possible for data processors to quality control (QC), visualize and interpret seismic data in an immersive, multi-volume and collaborative visualization environment, in real time, enabling rapid, informed, strategic and cost effective decisions.
Petroleum Geo-Services (PGS), headquartered in Oslo, Norway, has a specialized seismic data processing division that uses visualization technology as one of its core products. Suhail Butt, geophysical support manager for PGS’ Cairo office, holds over 20 years’ experience in processing and imaging. Alastair Lewis, the depth imaging manager for EAME/CIS based in Weybridge, has over 15 years’ experience, and has been with PGS since 2010. Together they explain PGS’ outlook on visualization technology.
Schlumberger provides many services to the oil and gas industry, including seismic data processing and visualization technology. Clark Chahine, technical marketing manager—geophysics, at Schlumberger Information Solutions, gives some insight into Schlumberger’s approach.
TGS, headquartered in Asker, Norway, also provides seismic data processing and visualization technology. Jim Howell, director, global imaging sales, explains TGS’s work on visualization technology development.
Left: PGS’ holoSeis shows the depth residual at selected offsets (red seed points) has been picked at each sampled xline and interpolated by the visualization system.
Right: holoSeis display of Viper data.Images from PGS.
OE: Could you tell us what visualization software and depth imaging technology you use for pre-salt formations?
PGS: The key visualization tool used to construct our depth imaging velocity models is a proprietary product called PGS holoSeis, which is developed and maintained by our in-house software group. In addition, PGS holoSeis is the front-end to accompanying technologies such as the PGS hyperBeam platform that can deliver rapid cycle depth imaging using our Beam imaging algorithm.
SLB: The Petrel E&P software platform enables 3D and 2D visualization of any geological information, including pre-salt formations.
TGS: TGS uses PRIMA software for time processing, visualization and interpretation, and ImageZ software contains the tomography and migration algorithms used during depth imaging. Both of these packages are TGS proprietary software. A module known as PrimaViz provides full 3D workstation visualization and interpretation capabilities.
Left: Angola, Velocity update from Image Guided tomography. Right: Angola, Velocity update from conventional tomography. Images from TGS.
OE: How does the quality of seismic acquisition affect visualization and does the visualization software optimize the seismic acquisition data quality?
PGS: Visualization is not affected by the quality of seismic acquisition, however, good visualization tools can better determine the quality of data acquired and processed. The exception to the above statement is in today’s ever demanding environment of data acquisition where data volumes are increasing rapidly and where multiple data gathering programs are conducted over the same field prospect. Integrating all this data into one “seamless” experience is a huge challenge.
SLB: The quality and type of the seismic data acquired influence the visualization of a geological object, especially in a complex salt environment. The Petrel platform allows users to load and visualize any seismic data type. It features processes to clean the signal and extract crucial seismic measurements directly into the project, such as seismic attribute and post and pre-stack processing tools. This helps enhance the image beneath the salt and accurately delineate salt boundaries.
TGS: Receiving seismic data that was acquired with acquisition parameters suited for the geophysical (e.g. correct sampling, good signal to noise ratio, proper fold, sufficient bandwidth) and geological objectives of the area is always beneficial, but the proper handling of the data during time preprocessing (e.g., de-noise, de-multiple, broadband processing, regularization) is critical to optimizing the quality of the data for both imaging and visualization. The visualization software would not optimize the acquisition data quality, but properly acquired and processed data would look better through all stages of imaging and interpretation.
OE: How do the technologies work, what algorithms and inversions do you use?
PGS: PGS holoSeis does not contain processing or inversion algorithms beyond simple functions such as attribute smoothing, editing and QC analysis. It does, however, integrate into the wider PGS processing package where all the algorithms required for seismic processing, imaging and inversion reside. As datasets become ever larger, it is no longer efficient to process using desktop workstations and therefore the batch algorithms are optimized for cluster technology, allowing the user’s desktop to focus on visualization demands only. However, seamless integration between data visualization and batch processing is essential and PGS hyperBeam is the ultimate culmination of that integration effort.
SLB: There are various algorithms and inversion processes used to enhance the image in this environment. Using a combination of seismic attributes, such as spectral or structural decomposition, based on unique edge detection attributes, users can blend results and visualize geological features more clearly. Using pre-stack data, in context, within the same environment as interpretation is undertaken, enables asset teams to get the most out of their seismic by looking directly at the source data. For example, by stacking on the fly and visualizing in 2D and 3D simultaneously. users can discretize the noise and remove it to enhance the image.
TGS: Specific to imaging around salt, the technologies applied during imaging and visualization would be using the optimal anisotropic depth imaging and migration approaches that solve the imaging problems specific to any given dataset. Approaches throughout the depth model building and final migration stages normally include titled transverse isotropic modeling, dip guided or image guided tomography, TTI Kirchhoff, TTI reverse time migration (RTM), and potentially TTI beam migration. Subsalt updating via RTM delayed image time analysis (DIT) is also used. Orthorhombic modeling and migration (Kirchhoff, RTM, or Beam) might also be utilized.
OE: What advantages do these visualization technologies bring to pre-salt formation characterization?
PGS: The advantage these visualization technologies bring is the ability to interpret the key geological markers and review the spatial consistency/integrity of the interpretation in real-time. In addition, new auto-tracking technologies enable the interpreter to pick a few “seed” locations on key events and allow algorithms to scan the volume to make the interpretation over the entire volume. In challenging areas such as pre-salt, this strategy, while providing a good initial guess, is not often good enough for a final interpretation, but the ability to rapidly view the data and make adjustments is essential.
SLB: Salt structures can take any kind of shape, therefore it is very important to be able to visualize the pre-salt formation from every direction and angle possible. In Petrel, for example, the 3D visualization is native, so you are able to fly through your pre-salt formation to visualize the full complexity of a reservoir.
TGS: TTI modeling and migration exhibit better imaging in complex geologic areas with steep dips, both the pre-salt sediments and sub-salt are better imaged. Image-guided tomography provides a more stable, structurally consistent velocity depth model. DIT is used to tune the image both at the salt base and in the sediments below. Orthorhombic modeling and migration may be used in the more complex areas that exhibit azimuthal anisotropy (fracture and stress zones).
Example of visualization utilized during fast track pre-stack depth modeling and migration in frontier area: Sediment flood migration with supra salt velocity overlay. Data is from the TGS Olho de Boi multi-client 3D, deepwater Campos Basin.
Image courtesy of TGS and Dolphin Geophysical.
OE: How long have you been using these visualization technologies and can you give us example(s) of pre-salt fields where the technology was used?
PGS: PGS holoSeis has been used for over a decade to visualize data in all hydrocarbon-bearing regimes in the world, onshore and offshore. Recently, Brazil, Gulf of Mexico and West Africa have seen the most consistent and concerted exploration activity and successes, but not without their own imaging challenges. PGS works for all major oil companies and have visualized data from all producing oil provinces and frontier exploration areas.
SLB: It has been in use for more than five years. The Petrel platform is used in the Gulf of Mexico, offshore Brazil, Angola, and the Barents Sea, as well as in onshore projects in North Africa and Europe for depth imaging studies and to interpret pre-salt structures. It is used by independent, major, and national oil companies alike. The platform optimizes pre-salt visualization, interpretation, modeling, and decision-making.
TGS: Prima and ImageZ software has been in use for over a decade, although new or updated interpretation and visualization modules are released regularly by the TGS research & development group. These technologies have been used in Brazil (Campos and Santos Basins), Angola (Blocks 35, 36, and 37), and over the majority of the central Gulf of Mexico and deepwater fields where TGS or their proprietary clients have data.
OE: How is the imagery interpretation done after the image is produced and how long does the process, from receiving the data to producing and interpreting an image actually take?
PGS: The time taken from receipt of field data through to imaging and final interpretation can take 6-18 months. The length of time is a function of the data volume and geological complexity – which can impact the complexity of the image processing workflow and the time required to interpret. Unfortunately, neither is an exact science, which can often lead to undesired project delays while un-foreseen challenges are overcome. In addition, the regular improvement in imaging technologies implies interpretation can be significantly affected in some cases. As a result, interpretation in challenging geological regimes evolves over time as more and more uncertainty in the image is resolved with ever better imaging tools.
SLB: The interpretation and the processing of the image (seismic) is an iterative process. To get the best image, interpretation needs to be done to apply depth-imaging algorithms and get a clearer image, then a new interpretation is done and another depth imaging step is run, and this goes on until we get the best quality possible. The interpretation is done in the Petrel platform relatively quickly, depending on the size of the area to interpret, then it is seamlessly sent to Omega geophysics data processing software to get the new depth image even sharper and more accurate. When the data is acquired offshore, a first pass processing is done on the boat, which provides the first image. Then the iterative loop starts. This can take a few weeks or months, depending on the complexity of the subsurface. But today, because the tight link between the interpretation and visualization undertaken in the Petrel platform and the Omega processing software, this time is hugely reduced.
TGS: Once the very first depth images are available, the interpretation begins and becomes an iterative process until the imaging objectives are achieved. The length of the process depends on the size of the processing area, density of the acquired data, and complexity of the area from an imaging and interpretation sense. Large regional multi-client 3D surveys can take up to one year or more to process and interpret from start to finish, while smaller prospect level reprocessing projects could take roughly half of that time.
Example of PrimaViz 3D visualization. Image from TGS.
OE: Is the interpretation done in real-time with the client involved?
PGS: Interpretation of the final image is almost exclusively done by the client. The exception is during the velocity model building stage where interpretation required to control model compartments can be conducted in real time. It is typical for a client to send interpretation and imaging specialists to PGS’ offices for up to one week, during which time the client knowledge is integrated into the model building process. Interpretation updates can be then migrated in real-time and the impact upon imaging assessed.
TGS:The client is normally always involved in the interpretation, as they have the best insight into the geologic details of the area. The picking can be done in real-time either by the interpretation group at TGS (with client input) or by the client.