Seismic Metadata Management: Optimisation with GIS

Seismic Metadata Management: Optimisation with GIS

Todd R. Porter*, M. Robb Isaac, Monty R. Martin
GeoSynergy Inc.
16225 Park Ten Place, Suite 805
Houston, TX 77084
Website:www.geosynergy.com

Abstract
As 3D seismic surveys continue to grow in size and complexity, logistics and planning play an ever more important role in determining the efficiency and profitability of a seismic operation. Meanwhile, the seismic metadata (information describing the seismic data) associated with these surveys is rapidly increasing in volume and variety, yet is still stored in disparate locations such as trace headers, co-ordinate files and paper and electronic observer logs. With the increasing popularity of time lapse seismic, this metadata can be considered to be used for the pre-plan information of a repeat seismic survey, and it is consequently more important than ever that it be efficiently gathered, stored, analysed and archived.

By combining these previously autonomous metadata sources with GIS layers (such as digital ortho-photos and CAD files of pipelines or wells) into one data warehouse, real inter-dependencies can be established through spatial, attribute, and temporal analysis, and proper cause and effect examination performed.

The industry has numerous powerful geological and geophysical planning, evaluation, processing, and analysis tools, however few operational tools exist to effectively manage today’s complex 3D projects. A system called MATRIX ® has been developed to address these requirements with an effective implementation of GIS, a flexible and familiar UI, and data connection technology. The net gains are improved efficiency and quality through effective information management, analysis, and dissemination.

This paper will present the issues critical to building such a system, and will examine case studies illustrating the benefits that can be derived therefrom.

Introduction
In land, transition zone (TZ), and mixed mode seismic projects, significant effort and expense is invested in designing and securing an exploration prospect. This includes mineral leases, surface permits, and any other field campaigns to collect data in efforts to better define the area of interest and reduce liability and exposure.

Seismic operations are becoming more difficult to conduct due to culture, permits, restrictions, divided interests, and complex acquisition methods used to improve quality of the final seismic product. Valuable information is gathered and compiled during feasibility, evaluation and design stages. Minerals, permitting, and pre-survey / hazard mapping information is provided prior to, and during the advance, survey, and drilling operations. These subsequent operations report incrementally on their respective progress as well. As these inter-dependant operations usually run concurrently, it can be extremely challenging to achieve production, quality, and safety objectives.

A system for managing, analyzing, and presenting information from inception to completion would therefore provide benefit to a wide variety of user groups involved in the project. From crew clerk to field geophysicist, seismic processor to operations manager, or drill push to safety auditor, an open implementation would enable these users to enter, access, report, and perform analysis to support decision making.

The Requirement
In order to conduct operations more efficiently, a comprehensive operations management system is required which captures the operational workflow. From this functional workflow, data are consolidated and joined to provide a comprehensive relational data structure.

With this consolidated information on a given project, access can be provided to a wide variety of clients for planning, coordination, and analysis by means of a Geographical Information System (GIS), open database architecture, data connection and load wizards, report and analysis builder, and web-enabling technology. A key requirement of such system is data dissemination.

A wide variety of user groups involved in a project experience significant information disconnects. Examples of these disconnects occur early, between the Client team and field operations, i.e. by not taking full advantage of permitting, minerals, design flexibility, hazards, and cultural information. Disconnects propagate during the field operations, i.e. the lack of information sharing between permitting, survey, drilling, and recording, the result of which can be trespass, proximity violation, accidents / exposure and resultant damages. Productivity and quality can suffer, as well. Finally, the production of an accurate geophysical acquisition database containing the seismic metadata required for data processing presents another disconnect in the overall process.

The Method
The system architecture utilizes GIS technology, COM (component object model) methods in a Windows 98/NT environment. In this “open systems” approach, including OLEDB, data sources and other third party products are attached to take advantage of external sources and functionality. GIS technology adds the spatial analysis capability, i.e. intelligent maps, and a geographic user interface to the databases.

Data Model
In order to manage the complex aspects of today's 2D, 3D, and emerging 4D surveys, a functional data model was designed and implemented. The design accommodates multi-component recording operations, i.e. 1C -> 4C, full point indexing, grid logic (as shown in Figure 2), multiple coordinate categories, point status, and all related attributes. This core data model is the foundation categories, point status, and all related attributes. This core data model is the foundation for geophysical operations specifics.

The data model implemented is Hybrid, made up of a core data centric store, and the ability to join (relate) non-centric sources. Referring to Figure 1, this flexible architecture enables the User to access external databases, which are likely administered by others carrying out specific functions within the project. Leveraging OLE-DB, legacy databases are incorporated into the project, and prior investment in these sources is preserved.

The data model can be described as dimensional, via indexing and point category. Indexing, according to the SPS3 format definition is as follows; "Shotpoint: to be one (1) for original shot within the grid cell denoted by fields two (2) and three (3) (this is the line station identifier), and be incremented by one (1) for each subsequent shot within the same grid cell." Similarly, "Receiver: to be one (1) for original positioning of a receiver group, and be incremented by one (1) every time the receiver group is moved or repositioned, even when put back to any previous position.” The model uses a counter, to track each new occurrence of a source and receiver point. Even the best survey designs are subject to change during operations, and today, a large percentage of stations are moved at least once, and sometimes many more times throughout than span of the survey for many operational reasons. Therefore, it is imperative that the data model supports the storage and management of these events.

As Figure 2 illustrates, indexing is a key function. The underlying grid logic is imposed for conformance to rules regarding station numbering, renumbering, half and non-coincident source / receiver grids. Station renumbering occurs when source and receivers are offset from original design and to relate to their new grid cell location. All relationships must be preserved, i.e. the survey point relation to an original design point. Point categories have been developed to accommodate design, survey, observed (at shot time), instrument, and target. The target category holds the grid logic; therefore conformance can be checked for any category. Figure 3 is a simple example, illustrating these relationships; where the "+" symbol is the target rule for point placement, and lines show relations.

Query Mechanism
The query mechanism is the gateway to analysis, from simple to complex. With the hybrid model described earlier, and the ability to perform relational threading to any Open data source, a wide variety of questions or queries can be built. By combining the spatial and attribute data, analysis and results can be presented in map views. These maps are the most effective communication medium for geophysical operations. Spatial operators are available for point, line, and polygon objects. Source and receiver points are obvious examples of point objects, pipelines, roads, waterways, etc., examples of line objects. Permits, lakes, city limits are examples of polygon objects.

Queries may be built with spatial, attribute, and temporal information. Temporal (time/date) data may be considered as an attribute that is critical in generating production statistics and forecasts. As attributes are loaded into the core database, event time tags may be produced if not available in the data content. As well, task codes may be assigned to track the most current status of any source or receiver point operation, and its history. Current status is highly valued information for daily operations planning and reporting.

Let's review general classes of operators.

Where
Proximity - within a distance of
Buffer - within a zone surrounding an object
Within An Area
Intersecting
Touching

When (exact)
Current (accumulative)
Interval (daily, weekly, monthly)
Valid, Expiration (periods)

How
Who
What
Why

Combinations of these general operators produce very comprehensive analysis.

Figure 4 illustrates the result of a very simple query. The User is concerned that heavy rain is forecast for this region of South Texas. The terrain is very flat and flood warnings have been issued. Unfortunately, receiver stations have been surveyed, and layout of receiver groups completed. These land geophones may have to be replaced with marsh phones. How many stations are within 200 feet of any waterway, and are these indeed geophone types? The result is displayed and a list generated, complete with line, station, receiver type, crew, deployment time and date. The Coordinator can then act accordingly.

In the next query example below, the User has just received hazard survey data from the survey crew. All the producing wells in the area have been identified, and a proximity analysis performed. The spatial query results identify all design source points with planned charge sizes of 5 lbs within 300 feet of producing wells as shown in Figure 5. A subsequent list is generated for design modifications.

No query discussion would be complete without presenting a land permit example. One of the most difficult aspects of surveys in continental USA is managing permits; minerals, landowner, and tenant data. Aside from exposure to hazards or encroachment / damage thereof, establishing and maintaining good landowner relations is critical to the success of any size project. Therefore, many attributes are associated with the permit management data, and proper spatial registration of land tracts is challenging. Figure 6 shows the result of a "no permit" query, to identify which points will be dropped, moved, etc., and avoid any damages related to conducing operations on that land tract. This specific example shows all dynamite source points moved out of the town site. Note that the relationships are maintained between the original design, and the "grid rules" move to a new acceptable location.

Another difficulty with data management systems is providing the correct amount of data. A user may be interested in a specificregion for conducting certain operations. By spatially selecting an area via the map view, a simple query reports back only pertinent information as shown. This way, work assignments can be generated with specific and accurate data. This is an example of data focus for each unique user group on a project.

Reporting and trend analysis take full advantage of the temporal attributes in the data. Managers are interested in production statistics, which can be built for each work group and time interval on the project with time/date type queries. This is useful for cost tracking and forecasting.

4D / Archival
An effective means to archive, retrieve, and reuse the 3D survey information is becoming more important with data exchange / resale, multi-client AMI's, and 4D time variant surveys. The 3D survey is an expensive and valuable asset and can be managed as such in digital, rather than hardcopy form. Projects can be safely archived and retrieved on demand for review by prospective buyers or project managers planning the next 4D epoch survey. Complete and comprehensive analysis can be performed to satisfy quality and logistical concerns without giving away any subsurface information, or the "real" asset.

Results
Various degrees of derived benefit of this technology have already been realized. Most apparent has been communicated from User groups involved directly and indirectly on projects, that information is more readily available to assist their planning and execution of tasks. Therefore, a primary goal has been achieved. As part of producing final deliverables, the generation of a set of correct, fully indexed SPS files which are "seismic processing ready" is possible. By proper management throughout the project, geometry errors can be minimized, and hopefully eliminated thus improving the seismic data processing turn-around.

A number of other objectives are realized with GIS / database integration, such as;

Improved project management,
Improved 3D survey quality by making use of the best and most current information,
Improved efficiency, overall optimization, cost savings,
Improved community relations, and Reduced exposure (trespass, damages, liability),
Improved safety through hazard recognition and demarcation,
Enabling management of 2D, 3D, 4D, single and multi-component acquisition methods,
Enabling management of dynamite, VIB, HFVS, airgun, impact, stacking source types and methods,
Workflow and business model capture,
Comprehensive map and report production for each client group,
Comprehensive and up-to-date reporting, scheduling and cost control analysis,
SPS for on-demand geometry analysis, and final seismic processing,
Project archival, and direct import into other applications, analyses, and downstream processes,
Project analysis, what-if scenarios and “post-mortems” through statistical variance analysis,
Improved client group communication via Internet Map Server technology.

Conclusion
Currently, the industry has numerous powerful geological and geophysical planning, evaluation, processing, and analysis tools. However, limited operational tools exist to effectively manage today’s complex seismic acquisition projects. These requirements can be effectively addressed with an effective implementation of GIS and COM technology as shown. The net gains are improved efficiency and quality through effective information management, analysis, and dissemination.

References

"Building Applications with MapObjects V2.0", ESRI, 1999, ESRI, Redlands, CA.
"Command & Control of Seismic Operations", 1999, Todd R. Porter*, GeoSynergy Inc., John Archer, Grant Geophysical Inc., Presented @ SEG99, Houston, TX.
"MATRIX Functional Specification & Development Document Series", 1999 GeoSynergy, Inc., Houston TX.
"Shell Processing Support Format For Land 3D Surveys", as adopted by the SEG in 1993, SEG Technical Standards Committee, Shell Internationale Petroleum Maatschappij B.V., The Hague, The Netherlands.