Geoscientists visualize in 3-D, but spatial analysis has not been as widely used in production operations. Existing exploration and production (E&P) applications have been limited to proprietary platforms and complex data interfaces that undermine the ability to integrate financial and operational data.

Support for interactive analysis and real-time alarms and alerts has been limited to drilling applications, with no consideration for production operations. Where there has been operational analysis capability, these solutions have been optimized, black-box applications requiring specialized client tools and non-standard proprietary data formats not well aligned with standard IT architectures.

Complex oilfield operations
Use of 3-D data management, analysis and simulation is in the early stages of transforming

Figure 1. The PetroTrek Bubble Maps tab by Oracle Spatial partner iStore allows the user to graphically show relative production in areas of interest. You can click to see multiple types of data. Production volumes are shown in this example.
exploration and production. This is brought on by the growing availability 3-D spatial data in the formation, well bore and surface facilities. Sensors are now able to generate real-time spatial data that can be analyzed throughout the E&P value chain. This explosion of data and innovative 3-D solutions is generating a new set of requirements that should be addressed by software vendors:
• Support for large 3-D E&P data sets: terabytes of data, millions of objects;
• Support for open 3-D data standards for E&P (e.g., WITSML, PRODML);
• Fusion of real-time sensors, E&P applications and operational information with traditional 2-D vector layers and network models;
• Server-based spatial data management combined with Web delivery (the only viable approach);
• Spatial analysis: ability to conduct traditional GIS queries on 3-D
volumes (e.g., nearest neighbor, within distance, network trace, polygon overlay);
• Transactional updates: ability to add, delete and update 3-D objects — both the structures and attributes synchronously through distributed Web approach;
• Enterprise integration: ability to integrate 3-D structures with other centralized or distributed enterprise or Web content;
• Versioning and long transaction support: manage 3-D projects in version mode for long durations; and
• Object-level data security, access control, encryption, authentication and auditing.

These requirements are helping to define an evolving E&P spatial architecture for managing large scale, 3-D volumes. This architecture suggests subtle enhancements at all levels of the solution stack, including: data servers, rendering and analysis servers, Web clients, and 3-D data structures.

Advances in 3-D data management using commercial relational database management systems (RDBMS) provide the foundation for a scalable, high value storage and analysis platform for managing 3-D surface models, 3-D seismic analysis, 3-D earth models and reservoir simulation models. The technology is being complemented by the emergence of PPDM (Public Petroleum Data Model), a 3-D spatially enabled industry data model that describes E&P data from multiple sources. The PPDM data model and Oracle Spatial 3-D are ideal for comparing and overlaying 3-D data from these different sources. This server-centric approach overcomes the high cost of traditional 3-D visualization solutions by allowing end users to incorporate commercial, off-the-shelf and open source visualization tools on an open database platform. It also provides a secure transactional environment for handling very large scale geologic and simulation models.

These geospatial data management capabilities are positioned to now take
3-D data types beyond the domain of specialists and into mainstream oilfield operations.

Overview of spatial databases
Spatial databases support storage, indexing and spatial analysis of 2-D vector data. As spatial databases have evolved, they have incorporated support for additional data types and models including georeferenced (raster) imagery, network connectivity, persistent planar topology and linear referencing.

Database vendors have also expanded their geoprocessing capability to include most of the spatial analysis functions found in GIS tools (e.g., point in polygon, polygon overlay, spatial aggregation, network routing, geocoding, dynamic segmentation). An important value proposition of spatial databases is their openness, scalability, security and reliability. In addition, the leading GIS, CAD and image processing tools extend this robust platform to enhance data creation, editing, analysis, dissemination and archiving. Spatial databases have also been instrumental in advancing the recent concept of enterprise GIS and server-side geospatial processing, which has become a foundation for most enterprise and Web-based solutions.

By extending the 2-D features of spatial databases to handle 3-D data, it becomes possible for a new generation of 3-D solutions to blend E&P data, operational information and financial data. The native support for 3-D data models by the database kernel offers fast indexing, search and retrieval and opens the door for cost-effective emergence of new 3-D analytical applications.

Value proposition for 3-D spatial databases
The use of 3-D CAD tools has been a mainstay of architecture and engineering for many years. These relatively specialized, thick client tools, with their own file-based data structures, have generally used vector engines to support the design of man-made objects. However, petroleum engineers and geoscientists are increasingly incorporating different data types and oilfield work processes. The growing use of satellite imagery, 4-D seismic data, 3-D earth models and 3-D reservoir simulation is leading to the rapid, accurate and realistic 3-D modeling of large-scale oil fields. Spatial maps in 2-D and 3-D are useful for offshore and onshore pipeline operations and for comparing geologic, geophysical, engineering and operational data. In addition, the dynamic aspects of real-time drilling can be modeled and represented to support more realistic reservoir simulation. The ultimate goal of this effort is the accurate 3-D reservoir characterization of an oil field so that production and profitability can be optimized.

To support these new data models, traditional spatial databases have been extended to support 3-D data. For example, the Oracle Spatial index uses an R-tree index that supports up to four dimensions. This index can manage each dimension of a 3-D object, while still allowing index of one further dimension (e.g., time, measure). When creating a 3-D R-tree index, a bounding volume is created around the 3-D object. This bounding volume around the object equals the bounding volume around the solid. The index is helpful in conducting very fast search and spatial analysis over large 3-D scenes — capabilities that are not generally practical with traditional E&P applications.

Visualization requirements
The availability of large 3-D exploration data sets, deep offshore wells with 3-D directional surveys and complex production facilities are driving the need for integrated 3-D rending tools. To visualize 3-D objects it is necessary to use programs that actually retrieve 3-D files or spatial data objects and render them. Numerous 3-D CAD, and GIS tools perform 3-D visualization but none have the ability to perform spatial integration via mathematical algorithms. Many of these commercial and open source tool providers are extending their support for 3-D databases. The requirements for these 3-D visualization tools are: 1) support for large 3-D volumes and objects; 2) tight integration with operational and ERP databases; 3) the need to support interactive spatial analysis on 3-D volumes.

Examples of large 3-D E&P applications
Geographic search, data integration, visualization and simulation. 3-D simulations of petroleum reservoirs are enabling more realistic and comprehensive analysis derived from massive subsurface and surface data sets. The use of 3-D models by Google Earth and Microsoft Virtual Earth now enable non-expert users to walk through 3-D “virtual reservoirs” allowing unprecedented realistic views of seismic, geologic, petrophysical and engineering information. The combination of 2-D surface maps, 3-D earth models, smart wells and automated surface facilities create a collaborative, interactive visualization tool to optimize production. Additional value can be realized by incorporating financial and supply chain information.

Drilling platform design and visualization. Large offshore platforms and FPSOs frequently combine multiple data sources such as CAD models, 3-D walk-throughs, process simulation and photogrammetry to design large “super models” with an inherently high degree of accuracy so they can be confidently used in modeling extreme offshore conditions.

Summary
As the availability and diversity of rich 3-D geospatial data grows beyond geoscience, the costs of these solutions are dropping. Advances in handling 3-D data models in an RDBMS provide an affordable, scalable, and high performance storage and analysis platform for 3-D geospatial analysis. This server-centric approach to managing large, continuous 3-D geospatial solutions allows the integration of E&P applications, operational information and financial data. By building on open standards and integrating diverse sources of data, petroleum asset teams can focus on new ways to increase profitability and reduce costs.