Visualization of Multidisciplinary
Time-lapse Subsurface Data
Graham Brew, Bill Hanson, and Rick Schnell, Dynamic Graphics, Inc.
Summary
Time-lapse seismic acquisition and life-of-field seismic monitoring initiatives are becoming increasingly commonplace. These data can be critical to the decisionmaking process for reservoir surveillance and development of mature assets. Consequently, asset teams are under increasing pressure to incorporate these new temporal data into workflows as quickly as possible. One of the best ways of rapidly reconnoitering and analyzing these data is through comprehensive visualization solutions.
The example of temporal seismic data highlights the wider requirement for visualization solutions that pervade the entire multidisciplinary team. There has never been a greater need for spatial integration and analysis of all available data. Accurate and timely development decisions require that all relevant data, from first seismic to latest production statistics, be available to all team members as quickly and easily as possible. Ideally any visualization solution should be accurate, spatially consistent, quantitative, and with the lowest achievable barriers to entry. Current ad-hoc solutions (e.g., screen shots and slide show presentations) do not achieve these objectives.
Presented herein is a software solution for these ever increasing demands of the visualization environment. The solution is compatible with a wide variety of common geoscience and engineering data types, and has considerable latitude of import flexibility, requiring only limited data conversions and migrations. The visualization solution is centered on a fully geo-referenced interactive 4D viewer. The viewer is capable of simultaneous display of time-lapse seismic data, geologic models, well data, instantaneous and cumulative production data, 4D reservoir simulations, and much more. Moreover, the solution also includes links to ″non-spatial″ business data (e.g., financial spreadsheets) that are equally important in the investigation. Our proposed solution satisfies many of the essential requirements of a fully functional cataloging and visualization environment that can unify the analysis across the multidisciplinary asset team.
Introduction
Subsurface geoscience and engineering data, of increasing variety and increasing volume, are being acquired at accelerating rates. Not least among these data, time-lapse seismic acquisitions and life-of-field seismic initiatives are becoming increasingly commonplace. These trends are accelerating the rate at which coincident subsurface datasets are obtained, and shortening the time available for meaningful analysis.
All these subsurface data, both static and temporal, are paramount to accurate and focused decision-making for optimum field development. Hence the rapid, simple, and complete integration of these data into the workflow of the multidisciplinary asset team is an economic necessity. To fully leverage team synergies, the geologist, geophysicist, and reservoir and production engineers should all have access to the data from each other′s disciplines. Moreover, all these data should be easily accessible from an individual′s desktop, and should be visualized and interrogated in a unified 4D geo-referenced space.
Presented herein is a software solution for these ever more demanding visualization requirements. This visualization solution consists of a fully geo-referenced interactive 4D viewer (Figure 1). The environment is capable of simultaneous display of time-lapse seismic data, geologic models, well data, instantaneous and cumulative production data, reservoir properties, 4D reservoir simulations, drilling hazards, and many other data types essential for fully-informed decision making. The software is compatible with a wide variety of common data types and has considerable import flexibility, requiring only limited data conversions and migrations. Furthermore, the links between the viewer and the source data are dynamic, meaning that the material can be automatically kept up-to-date.
Most importantly, the viewer environment is extremely easy to use and removes many of the barriers to entry normally associated with new software.
The remainder of this abstract highlights the types of temporal and spatial data typically required of a useable solution. We also include examples from the North Sea where use of this type of visualization tool to integrate time-lapse seismic data with reservoir simulations has allowed critical development insights for mature assets.
Spatial Data Types
As we have seen, the demands of the visualization environment are data-driven. The following is just a partial list of the types of spatial data that the visualization environment should be capable of rendering:
- Seismic data volumes in time and depth: To include, simultaneously, multiple seismic attributes calculated from 2D, 3D, and 4D seismic volumes.
- Seismic interpretations including horizon and fault picks. Also seismic velocity data and supporting data.
- Well locations, rig locations, well path surveys: To include mechanical data, perforation details, casing points, positional uncertainty, drilling targets.
- Well logs, curve data, VSP′s, synthetic seismograms, televiewer images, core photos, flow tests, biostratigraphic data etc.
- 3D structural models and property models that include faults and horizons, along with their supporting data. Possibly multiple structural realizations encompassing model uncertainty.
- Annotation⁄cultural data: To include lease boundaries, facilities, coastlines. Also remotely sensed data including aerial⁄satellite images, digital elevation models, potential field geophysical data, and surface geologic maps, and outcrop laser scans⁄photographs.
Temporal Data
The efficient development of subsurface resources is very much a 4D, rather than a 3D, problem. Hence any visualization system should be fast, efficient, and flexible in displaying a wide variety of temporal data. The timevariant data utilized by the asset team could include:
- Reservoir simulations: history matches and predictions with a potentially huge variety of reservoir properties; to include all supporting data.
- Time lapse seismic data: Many gigabytes of data, collected possibly as frequently as every several months in life-of-field seismic situations. Potentially dozens of coincident seismic amplitude cubes.
- Time lapse seismic attributes: Calculated from the multiple seismic data cubes. Various combinations of attributes, particularly the examination of differences between adjacent time-steps, can lead to thousands of seismic attribute maps calculated for the various reservoir horizons.
- Production data: Oil⁄gas⁄water production, injection, fluid compositions, pressures, GOR, water cut etc. Both cumulative numbers and instantaneous rates.
- Well data: Date of drilling, plus potential changes in the status of the well bore (e.g. from producer to injector, or from producer to abandoned) that span the entire life of the field.
- Mechanical data: Which perforations were opened and when, time for pump on⁄off, work-overs, flow tests, choke settings, and other well events.
- Well logs: Multiple well logs may be collected from a single wellbore during reservoir development.
Non-Spatial Data
As discussed above, the subsurface development team will likely use a huge quantity and variety of spatial 3D and 4D information. Equally important, however, can be nonspatial information, or information with only ″soft″ spatial associations. These data are critical to the decision making process, and rapid access to these data by the team within the multidisciplinary workflow is essential. Examples of these types of data could include:
- Background geological information⁄reports
- Photographs
- Graphical presentations
- Well review documents⁄written reports
- Mechanical schematics
- Risk Assessment⁄financial analysis⁄forecasts
- Visual Montages as used in review meetings
- Other business information and databases
Furthermore, these data are likely to be found in a variety of non-spatial formats such as presentation slide sets, spreadsheets, text documents, images⁄PDF files, websites⁄intranet sites, or innumerable database formats.
The solution we present provides mechanisms for cataloging and tracking this information by associating these non-spatial data into a broader ″data registry.″ In doing so, these data can be accessed in the same way as any other data type. The data can then be displayed within the 4D environment, or in their native applications. As such, these key ″decision support″ data can be accessed, reviewed, and analyzed as readily as any true spatial object.
Archiving and Workflow Documentation
The proposed visualization solution permits dynamic links to be established to all the data sets that are required in multidisciplinary analysis. One beneficial side-effect of this arrangement is that this data registry can also be used to instantly archive all the data as used in the decision making process. This archive could include all the data and the native datasets, as well as the more generic data formats used in the visualization.
Furthermore, the registry can be used to record critical points in the workflow, and to capture associations between related, dependant datasets. For example, the process can document which set of well logs were used to pick the formation tops, and subsequently which set of tops were used in the construction of the geologic model.
In conclusion, the archiving capabilities can provide a thorough ″audit trail″ of the reservoir development workflow, decision-making, and reserves estimation.
Multiple Model Spaces
The visualization environment as presented is clearly capable of displaying large volumes spatial information of considerable variety. When numerous data sets are displayed concurrently, the display may become confusing, and difficult to interpret if not handled correctly. Furthermore, datasets that are spatially coincident can commonly have different vertical units. Consider, for example, a time-migrated seismic cube (scaled in seconds), or a magnetic anomaly map (nanoTeslas) that need to be displayed alongside a reservoir model (meters).
To accommodate these requirements, the proposed visualization environment allows multiple ″model spaces.″ These are duplications of the z (vertical) space stacked above one another and spatially referenced in x and y. Such displays can make individual data components much easier to visualize, and allow for different vertical units in the different spaces (Figure 2). Furthermore, this method allows for the interactive tracking of features between the model spaces (e.g., simultaneously tracking a structural horizon and a seismic amplitude cube).
Schematic Temporal Visualization, North Sea Example
The example below shows visualized multidisciplinary data can yielding valuable insights for asset development. These insights would be difficult to comprehend, and may indeed be missed entirely, if the various spatial data were not displayed and animated simultaneously. (N.B. the software displays a complete continuum of animating time steps; only three time points are shown in Figure 3.)
July 1999
Figure 3 shows: seismic amplitude cube (1993 vintage), seismic attribute (difference) map calculated from two different seismic amplitude cubes and displayed on the topsand surface, well data (colored according to the status of the well in July 1999), and a reservoir simulation prediction colored by water saturation, filtered to >18% Sw. Numerous other data that were used in the decision making process have been removed for clarity in this static display.
Note the large red seismic attribute anomaly in the lower right. This seismic response indicates strong pressure support⁄water injection in this (upper) portion of the sand. Yet the reservoir prediction model shows only low levels of water saturation in the sand at this time point.
February 2002
Note the change in well status (color) by this time point, including the addition of a water injector (blue) into the upper sand. The reservoir simulation now shows increasing water content, but still not at the level indicated by the time-lapse seismic attribute.
Further seismic attribute analysis of a second, lower sand (not shown) shows anomalously low water entry to that lower sand. Hence the team concludes that the perforations to the lower sand in the injector well are blocked, leading to elevated levels of injection into the upper sand. This unexpected blockage explains the discrepancy between the reservoir simulation (prediction) and seismic attribute response (observation).
September 2004
New seismic data have been acquired since the previous time step. Hence this temporal image shows a new seismic cube (2004 vintage) and a new seismic attribute difference map. The mismatch between seismic attribute and the reservoir prediction is lessened, but still requires a reevaluation of the reservoir prediction model. Also shown (via 3D symbols) are production and injection data statistics for this time step. Examination showed that water injection pressures into the upper sand had spiked, thus supporting the hypothesis that blocked perforations were impeding water injection into the lower sand.
Conclusion
Better, faster access to critical data is vital for fully informed decision making by today′s multidisciplinary asset development teams. We have presented a visualization solution consisting of a fully geo-referenced interactive 4D viewer. This visualization environment is capable of simultaneous display of very wide variety of static and temporal data, both spatial and non-spatial, critical to asset development. When all these data are made rapidly and seamlessly available on team members′ desktops, then the full value of all data can be extracted, and better decisions result. In the example shown, only through the visualization of geophysical and reservoir engineering data were discrepancies between the datasets observed, thus leading to critical development decisions.
