Speaking at the annual meet of the professional petroleum data management association, PPDM, Bruce Smith, member services manager, investigated the data management requirements of the booming hydraulic fracking business.
While fracking has been used by the oil and gas business since the 1950s, the shale gas boom has increased scrutiny of the activity with, as Smith reports, a slug of new requirements from the regulator.
Texas and other states revised hydraulic fracturing rules during 2011, Colorado and Pennsylvania are to require disclosure of frac fluids in a ‘queryable’ database and in Canada, British Columbia is to require online registration of frac information starting in January 2012. The Interstate Oil and Gas Compact Commission (IOGCC) and Ground Water Protection Council (GWPC) have set up a website for frac fluid reporting. As of October 2011, some 5400 wells have been registered by 44 companies. Reports, in PDF format, contain voluntarily-supplied information on frac fluid chemistry.
The GWPC’s own ‘risk-based data management system’ (RBDMS.NET) is used by some 20 US regulators for management of well, production, injection, surface facilities, permitting and more. The RBDMS has been extended with a frac data model covering treatment dates, chemicals, water sources, samples and disposal techniques. RBDMS includes fairly comprehensive modeling of stimulation activity and lifecycle frac fluid use across multi stage treatments including CAS and MSDS chemical tags. Another significant contribution is Energistics’ Witsml ‘Stimjob’ object for exchange of post-stimulation data between service company and operator. Operators’ requirements may be more detailed, with a need for ready access to treatment information from offset wells such as proppant, number stages, pressure and horsepower data. The current PPDM well treatment table offers summary-level data on most of the above, but, Smith concluded, ‘There have been sufficient changes in frac reporting and in information supplied by service companies that PPDM should take a fresh look at the model.’
PPDM CEO Trudy Curtis suggested that the subject could be treated in the context of a wider investigation of the data management specifics of non conventional resources in general. These include tar sands mining, coal bed methane and shale gas/oil. Some of these issues are already addressed in the upcoming 3.9 release of the PPDM data model.
The subject of ‘non conventional’ data management also cropped up in another presentation from Bruce Smith on well identification—an extension of the earlier ‘what is a well’ project. In complex wells with long multiple completions and environmentally-sensitive treatments, the unambiguous naming of well bores and laterals has added pertinence. PPDM is working to update the API well numbering standard and bring it into the modern world.
Chuck Smith (Divestco) introduced the petroleum education task force (PETF), a group of data management specialists from, inter alia, Total, Hess, Chevron, BP, RasGas and Devon, who are working to build an education program and to ‘capture domain knowledge before it is lost.’ The PETF is working to ‘align’ PPDM education with the DAMA data management book of knowledge and Common Data Access’ competency map. A framework for PPDM certification is under development as is ‘Dacum’ (developing a curriculum) - based content. A new online course on US land survey systems will probably have launched by the time you read this. Live classroom sessions are planned covering architectural principles, ‘what is a well,’ geochemistry and PPDM implementation.
Steve Cooper’s (EnergyIQ) presentation went ‘beyond the well,’ with an overview of three projects that extend PPDM’s traditional subject area. One company has leveraged PPDM ‘spatialization’ to support handheld field data capture. These have been tied in to upper level PPDM modules including facility, equipment and HSE incident. Spatial data can be mapped into PPDM’s ‘SP_’ tables, even though many deployments store spatial data in the GIS system. 95% of the required spatial attributes were mapped directly to PPDM 3.8 with the rest added as new tables in what Cooper described as a ‘robust, fully-integrated’ solution.
Other projects studied involved oil and gas meter management and AFE/drilling cost tracking. Cooper concluded that PPDM is a flexible and powerful data model. The ‘Component’ tables provide excellent support for integration projects—allowing complex relations to be made across equipment, sites and support facilities. However, it takes patience and a thorough understanding of the problem to build out the model correctly.
Neuralog’s Robert Best advocates ‘standardizing’ your PPDM implementation. Best showed how this was achieved in a recent joint development undertaken by Neuralog, OpenSpirit, Volant, Oilware and several clients. Standardization works at different levels, on natural key values for reference data, consistent use of API well numbers (for the US) and a global UWI for international work. Denormalized latitude/longitude data can be embedded in the Well table for performance. For multiple sets of data, the Active_Ind flag should be used to indicate which is the preferred set. Likewise, a records management strategy requires careful thought—again maximizing the usefulness of the natural key. While binary large objects (Blobs) can be stored in PPDM, most implementations use linked files. Here both original (source) location and current database copy locations need tracking in the records management module. Best recommends using the OGP Geomatics. To date, vendors take different approaches to ‘standardizing’ PPDM. Best suggests that standardizing the standards will provide better inter-vendor interoperability.
Earl Amankwah outlined how IHS provides support for multiple PPDM versions. This proved necessary after IHS migrated its own data services to 3.7—while several clients’ older releases required support for in-house applications and scripts. Implementing backwards compatibility proved tricky. Challenges of code consolidation, primary key changes and new PPDM columns in 3.7 have been solved thanks to coordination with clients’ teams and IHS ‘proven’ methodology. Download the presentations from PPDM.
© Oil IT Journal - all rights reserved.