European Association of Geoscientists and …2004 EAGE Conference and Exhibition European...

2004 EAGE Conference and Exhibition

European Association of Geoscientists and Engineers Annual Conference and Exhibition, Paris, June 2004.

Figure 1 INT's 3D Viewer1

While the upstream software business evolves gradually, hardware is seeing a veritable sea change towards cluster-based visualization solutions. To enumerate a few: Schlumberger with GigaViz, IBM with Deep View, HP with Sepia, Paradigm with ResNav, Linux Networx/Fraunhofer with PV-4D and, if you want to do it yourself, you can talk to TGS (now Mercury) whose VolumeViz technology is under the hood of many third part implementations. Perhaps because of this, alliances between software and hardware vendors are popping up. Landmark is teaming with IBM on a variety of shrink-wrap solutions, typified by its Rapid Prospect Generation Engine. IBM is also allying with Schlumberger to sell ‘on demand’ computing (DCCoD) for example to perform multi-pass modeling with Eclipse and the new Cougar pre-processor. While HP is also working with Schlumberger to develop… shrink wrap solutions. Looking further down into the hardware stack – and indeed into the alliance space, Intel gets a special mention. Last year, most everybody had a ‘solution’ built to run on Itanium 2 clusters. This was the next big thing and would kill SGI, Sun, and especially AMD. Except that the Itanium 2 has been a flop and AMD’s Opteron has taken over the 64 bit market. Now Intel is rolling out a 64 bit Xeon chip while they fix the Itanium and is meanwhile doing the rounds, keeping its old Itanium customers ‘on message’ with healthy dollops of FUD2. Unfortunately, there was little cooperation between the plenary speakers (or more accurately, readers). This meant that they all came along with the same slides and discourse. Remember the days when oil was heading south of $25 – and we were told

1 All images are ©, courtesy of the companies concerned and gratefully acknowledged. 2 Fear, uncertainty and doubt.

Technology Watch Report 1 © 2004 The Data Room

http://www.int.com/


that low cost oil was ‘here to stay?’ Well now that oil is heading north of $40 the pundits are saying high price oil is ‘here to stay’. Whatever. Counting a few pages of the pre-registration names, we figured that around 20% of the attendees worked for oil and gas companies and NOCs. A much higher proportion of (potential) buyers to sellers than has been the case in the past. Having the EAGE located in Paris meant that there was an awful lot of talk of statistics and uncertainty. Hardly a presentation or product escapes this vogue, to the extent that folks seem almost more concerned about the error in the error than on the numbers themselves. Geostatistics and uncertainty have been applied to reserve computations for some time now. Now the same math is offered to support seismic velocity analysis (ERM.S) and sedimentology (Geovariances). Having to do multiple ‘realizations’ of models which are already very compute-intensive – even when run only once – means that it is wise to avoid a full frontal approach. One interesting strategy uses ‘experimental design’ to screen models for key input sensitivities (Cougar from the IFP). Even the good old history match is getting a re-vamp – with the application of statistical tests to determine the validity of different match hypotheses from the data itself (Scandpower’s Mepo). On the Society front, lots of philosophizing has come from the EAGE and SPE – whose president Kate Baker gave a talk at the plenary. All the orgs are coming under pressure to limit the number of conferences. Because conferences equate to revenue for the orgs, this is not happening very fast. In fact it’s not happening at all! Speaking of revenue, the EAGE is the only conference which requires payment from the press (including The Data Room/Oil IT Journal) to attend, which does not exactly endear them to us. But making the young things who distribute the Hart’s Show Daily pay for a day ticket? And refusing Hart’s reporter access to the plenary because he had the wrong sort of pass? That’s hard nosed!

Highlights Executive SessionIFP’s CougarInterviewsINTViewerLandmark’s Tanks and Tubes Lessons of a merger – TotalNeural nets and Committee MachinesPermanent monitoringPlenary SessionScandpower – Mepo



Contents Plenary ...........................................................................................................................5

Don Milne (EAGE president) ....................................................................................5 Kate Baker (SPE president) .......................................................................................5 Christophe De Margerie (VP E&P, Total).................................................................5 Olivier Appert (CEO, IFP).........................................................................................6 Andrew Gould (President, Schlumberger).................................................................7 Robert Brunck (President, CGG)...............................................................................7

Executive Session ..........................................................................................................7 Geoscience after the oil peak – Manuelle Lepoutre, Total ........................................7 Fields of Future – Peter Carragher, BP......................................................................8

Interviews.......................................................................................................................8 Dean Hutchings – President, Linux Networx ............................................................8 Mark Bashforth, Andreas Hatloy – Roxar .................................................................9 Murray Roth – Landmark. ......................................................................................12 Phil Neri – Paradigm................................................................................................14 Ali Ferling – HP.......................................................................................................15

Demos ..........................................................................................................................16 Cougar – Schlumberger/IFP ....................................................................................16 GigaViz – Schlumberger..........................................................................................16 Grid Computing – IBM............................................................................................17 Wireless devices in oil and gas – IBM.....................................................................17 Seismic Image Segmentation – Dave Hale, Landmark ...........................................17 Decision Space/Nexus – Landmark .........................................................................18

Data Session.................................................................................................................18 eEarth, an EU standard for borehole data – Jan Jellema, TNO ...............................18 Information aggregation – Nathan Balls, Petrosys ..................................................18 Lessons of a merger – Philippe Baldy, Total...........................................................19 Shared earth model and portals – Piantanida, ENI E&P .........................................21 Earth science decision making – Troy Wilson, Geosoft..........................................21

Exhibitors.....................................................................................................................22 Atos Origin, ISPoo3 release 10.2.............................................................................22 CGG, EYE-3D. ........................................................................................................22 dGB New fault and horizon tracking. ......................................................................23 Survey Optimization from enGenius .......................................................................23 ERM.S, SeisQuare and VSquare. ............................................................................24 PV-4D SeismicPRO – Fraunhofer Institute.............................................................24 GeoVariances, ISATIS 5.0. .....................................................................................25 Hampson-Russell .....................................................................................................25 INT, INTViewer/3D. ...............................................................................................26 IPRES, IPResource and IPRisk-Field. .....................................................................26



Knowledge Systems, DrillWorks 2004 R2.0...........................................................27 Landmark, Tanks and tubes simulation ...................................................................28 Magnitude, Microseismic monitoring......................................................................29 Mercury/TGS, VolumeViz5. ...................................................................................29 Network Appliance, SpinServer ..............................................................................29 Neuralog, NeuraDB and Neuralog ‘Plus’ ................................................................30 Neuro Genetic Solutions, Neural Networks and Committee Machines ..................31 Oedegaard – New release of Isis 3D inversion........................................................31 P2ES, Enterprise Upstream Suite. ...........................................................................32 Paradigm, ResNav....................................................................................................32 Petrosys, SDE raster-vector integration in version 14.4. .........................................33 Scandpower – MEPO history matching tool commercial........................................33 Sercel – Nomad 90 T Vibrator.................................................................................34 VS Fusion, VS3 VSP collaboration with Magnitude...............................................35 Weatherford, Clarion 4D permanent monitoring.....................................................35

EAGE Conference CD-ROM ......................................................................................36 Knowledge-driven shared earth models, Rainaud (IFP) et al. .................................36 Internet production database, Tchistiakov (TNO-NITG) et al.................................36 Listening to the waves, Taner (RSI) et al. ...............................................................36 Virtual reality for rock microstructure visualization – Kayser et al., Schlumberger..................................................................................................................................36 Managing subsurface work processes – Naess et al., Statoil...................................36 Integrating Skills, Processes and Technology – Naylor, Shell ................................37 Attracting and retaining technical personnel – Lloyd et al. (Independent)..............37 The Professional Society of 2020 – Baker, SPE and BP .........................................37 Value creation: an IM approach – Mehrabian (Schlumberger) et al. ......................37 Automatic Velocity analysis – Stinson et al. Data Modeling Inc. ...........................38 Storage of interpretation results – Aliverti, et al. ENI. ............................................38



Plenary

Don Milne (EAGE president) Milne announced that 3,500 were pre-registered for the signed up – confident that will be over 4,000 – a record3. The 200 exhibitors included Saudi Aramco and Total. Milne discussed the EAGE’s achievements in terms of measures vs. its objectives. These are ‘to promote, foster geoscience etc.’ The EAGE has sponsored 8 major conferences and workshops in EU and in North Africa (which was ‘a massive success’). The education program continues to expand with an agreement between the University of Bergen and Statoil for an E-learning program. The EAGE has acquired an e-Learning Portal which will be online real soon now. First Break is on the EAGE website and is now available in Russian. Membership is up by 10% over last year. A lot of work has gone into the website – which has become ‘an important marketing tool’. At Stavanger last year, EAGE agreed with SPE to hold joint workshops. The 4D seismic workshop was ‘an outstanding success’. In response from industry to reduce the number of conferences there will be a joint EAGE/SPE EOROPEC conference (used to be every two years). The EAGE and SPE are to ‘cross link technology content’ – allowing both to broaden content and help industry drive to reduce number of conferences. Of the relationship with the SPE, Milne avowed it was ‘not all smooth sailing, these are different cultures.’

Kate Baker (SPE president) Kate Baker, SPE president, concurred with Milne on the need to reduce the number of conferences and to enhance synergy between the organizations. The plan is to link websites, combine meeting and focus, promote DISC, share membership and help students. The first fruits of this effort will be seen at next year’s EAGE meeting in Madrid. There is also to be an international conference co-hosted with the AAPG in Qatar, in November 20054. The organizations are to share short courses and avoid duplication of effort. The SPE’s mission is to aid information exchange and the society offers a ‘one stop shop’ for research papers from the SPE and other affiliate societies. Baker asked ‘what would it take to make for more enthusiasm, how can we help members deal with uncertainty?’ But she offered no answers to her own rhetoric.

Christophe De Margerie (VP E&P, Total) DeMargerie (VP E&P, Total) sees ‘a glimpse of blue sky in a gloomy environment.’ While oil prices ‘flambent5’ it should be remembered that in 1981, oil was at $80 in today’s prices. Between 1979 and 1986, oil was consistently over $50/bbl in today’s dollars. Issues and opportunities: ‘We believe that exploration as well as reservoir is priority’. It is not just about price and speculation, ‘we are approaching 100% production which is totally new’. Also environment is critical. If Venezuela had

3 As Milne spoke, there was a random showing of EAGE slides. Thus when he thanked Total &GDF for their sponsorship, an advert for an HP IPAQ incongruously appeared on the wide screen! 4 By our reckoning that means that the joint effort to reduce the number of conferences has resulted in a net increase of one conference per year from 2005. You can’t cut an organization’s revenue stream with a simple entreaty! 5 Literally, ‘have caught fire!’



happened in 2004 the oil price would have been well over $40. ‘Speculators don’t exist – or if you like, we are all speculators’. ‘We all underestimated India and China. – we were WRONG!’ There has been too much focus on western stock levels and little focus on international growth. Worldwide, BOE/capita/year is increasing ahead of population growth. In 2000, consumption was 9 bn. tonnes, in 2030 it will be 15 bn. tonnes according to the IEA. Most reporting shows new production backdated to year of discovery and so under represents what is done today. But discoveries have been less than production for a while. In the reserves stakes, the international majors are actually small players. It is the NOCs that have the lion’s share. Saudi Arabia still claims 260 bn. bbl. If reserves are not so much of a problem, access to them certainly is. Asia consumes 0.7 units, EU 3.2 and N. America 8.3. Asian growth is racing ahead. Which consumption model is Asia to align with? The industry is faced with the challenges of new E&P domains, unconventional resources, exploration in deep and ‘blind’ zones and how to make the most of existing fields. This requires experienced people, financing and addressing environmental concerns. While we need to be ‘more adventurous’ we must be careful about costs because extra cash ‘will not be in our hands for ever’. Adapt contracts to local investment needs as Total has done on the Hydra field in Argentine. Here 10km offset extended reach and submarine wells are used to revamp the field. Companies should foster open dialogue with all social players – ‘share the earth’ in a different sense. NOCs need to produce more even though they don’t need the extra cash – ‘that’s their role’.

Olivier Appert (CEO, IFP) Appert, CEO of the French Petroleum Institute (IFP) was previously with the International Energy Agency where he did projections and addressed the challenges of limited resources and climate change. Energy creates 80% of greenhouse gas emissions. By 2030 there will be 2/3 more energy used than today of which 90% will come from fossil fuel. Gas use is to double in the next 30 years. Nuclear will decline as old plants are decommissioned. The graph of CO2 in atmosphere shows exponential growth in the last century. Kyoto is a first step, but we must go further. In France the objective is to reduce total CO2 fourfold. What can E&P offer? Workflow and data integration represent a need for further technology development. We also need to address history matching and uncertainties which are ‘assessed and reduced with innovative technology’. Appert gave a plug for new history matching technology based on probabilistic analysis. Turning to C02 sequestration, Appert described the EU-funded Castor project (part of 6th Framework) to ‘reduce the cost of C02 capture’ and ‘validate concepts6’. 30 companies involved – IFP in charge. Casablanca field (Spain) re-injecting 2.5 million tons of CO2 from the Taragona refinery. Lindach field (Austria). Snohvit acquifer Norway – 0.75 million tons/year – will remove C02 from natural gas before liquefaction – for a 20 year period. K12b GdF –Proned Netherlands. See also IEA position paper.

6 Does gas injection really need government funded research? And who will benefit? Oil company producers will, with a reduced eco-tax bill. Oh, and the researchers of course!


http://www.actu-environnement.com/ae/news/480.php4

http://www.iea.org/dbtw-wpd/textbase/papers/2003/Org_Fossil_Fuels.pdf

http://www.iea.org/dbtw-wpd/textbase/papers/2003/Org_Fossil_Fuels.pdf


Andrew Gould (President, Schlumberger) Gould admitted that his thunder had been stolen by previous speakers. He too believes in a durable price rise, for the first time, the problem is supply to a sustained demand growth. Spend is shifting away from new production and towards ‘decline rate’ issues. We know little about decline rate and have no clear pattern of what it is. 70% of world production is from reservoirs over 30 years old. A high decline rate equates to a high investment. New fields decline more quickly (Rose & Assoc.). There has been a remarkable ‘decline’ too in F&D costs – thanks to new technology. Including downhole smart wells and intelligent completion. Geophysics used to be a language spoken only by explorationists, now 3D goes beyond exploration and is even used to optimize topside design. Gould could not resist a plug for ‘Q’ time lapse seismics which cause a revised well path on Statoil’s Norne field – world’s first Q on Q. An example of how operators’ ideas are ‘injected’ into service companies for development. Knowledge management is an issue with the large scale retirement of workforce. But for Gould, ‘production decline’ is where it’s at, we should ‘manage decline through technology.’

Robert Brunck (President, CGG) Brunck came armed with studies – from the BP statistical review (2003!), ASPO – Association for the Study of Peak Oil and the Carnegie Institute. The latter found that seismic spend has declined as a percentage of E&P spend. This is due in part to productivity, but also to a tough, very competitive market. Brunck spared us the ‘new business model’ plea, saying he had ‘no comment’ to make on competition in the seismic market, except to point out that ‘sustainable development’ for the seismic industry meant profitability. In turn, Brunck plugged CGG’s new EyeD – a revamping of CGG’s marketing effort. CGG’s 4D seismic over Total’s Girasol field is used to monitor gas injection through a ‘superb’ amplitude difference signature. CGG’s challenges can be very real – cables are cut with machetes and personnel threatened. CGG also has to contend with more and more local contractors. Here, CGG is both a supplier and technology ‘transferrer’. But yes, Brunck does hanker for a new ‘business model,’ the seismic ‘needs to progress as much on the surface as it has in the subsurface.’

Executive Session

Geoscience after the oil peak – Manuelle Lepoutre, Total Today focus is on technology challenge of deep offshore. Tomorrow will see similar need-driven technology breakthroughs. Companies have to improve recovery and discover more beyond current limits. It will be necessary to drill deeper (can have reserves > 6000m) working with drillers to solve problems in the foothills, beneath salt screens and in ‘non conventional,’ remote areas. Lepoutre plugged Total’s geoscientists for their own PSDM, predictive modeling and ‘global, basin-wide understanding’. Two years hence, stratigraphic processing and enhanced recovery with the I-Field will be realities. Already Total has performed real-time monitoring of a horizontal well in Girassol. 4D seismics was used to locate unchanged (therefore unswept) area.

Q&A


http://www.westerngeco.com/content/about/press_releases/2003/news_05022003.asp

http://www.peakoil.net/


Schlumberger – Seismic technology is actually mature already. How about oils paying the cost of higher spatial sampling? Lepoutre – I don’t agree that seismics is mature – these techniques are not really implemented yet. We do not yet acquire HR seismics systematically.

Fields of Future – Peter Carragher, BP Tomorrows fields will be deeper, hotter, harder to image, and will have lower frequency response. Access to and timing of discoveries will be ‘unpredictable’. Current workforce is to renew itself. We should ‘be fit’ and continue to innovate. IHS Data shows steady decline in field size (except Kashagan and deepwater) since 1970 with poor replacement for oil (good for gas.) Depth to top pay is rising.

Depth to top pay (IHS Energy data) P 5 P 95 1970 2,000 feet 4,000 feet 2000 4,000 feet 18,000 feet

Chris Heath’s Survey of Geoscience Skills in the oil industry showed need for non technical workers and IT. According to the survey, the oil industry does not need sedimetomolgists, structural geologists, core and many other traditional skills. Biostratigraphy, petrology and remote sensing are ‘obsolete’. Carragher is not so sure. Challenges to the Field of the Future (FoF) are onshore, Arctic deepwater low quality reservoir – poor seismic attributes. How should we approach the FoF? Like youngsters playing soccer – all following the ball in a herd – or like Wayne Gretsky who ‘skates to where the puck will be’. There ensued a BP/Magic Earth spectacular with images of sandstone distribution along strike slip faulting, log patterns from biostratigraphic data and one shot strangely showing Mars imagery. Intriguing time lapse gravity data from Prudhoe Bay showed an ‘amazing breakthrough’ 10 micro-gal resolution made possible by 1cm GPS positioning. Risk analysis is also critical – it may be necessary to hold competing models of the reservoir for a prolonged period during exploration before one is shown to hold true.

Interviews

Dean Hutchings – President, Linux NetworxHutchings – We sold our first cluster to the US government in 1997 already running on Linux. Our first commercial sale was to Brookhaven National Labs. The company started in 97 as board manufacturer. Then our customers asked to have boards assembled and the cluster business was born. Clusters are not just a bunch of PC’s – managing them requires special skills. ClusterWorx provides a complete management solution for clusters – system admin, app management, job scheduling, resource management and prioritization. All greatly increases level of use. Cluster management has been the key to successful migration from big iron. For oil & gas this means better price performance than SMP systems. Our key accounts are GX Technology, Shell (Rijswick) R&D. What about companies like CGG which just uses ‘commodity’ Dell boxes for the clusters? Companies need to take a step back and see how much managing clusters is costing them. For most computational users, the value is created through a relationship with


http://www.aapg.org/education/vgp/skillssurvey.pdf

http://www.spe.org/cgi-bin/viewpaper.cgi?paper=00087662.pdf

http://www.spe.org/cgi-bin/viewpaper.cgi?paper=00087662.pdf

http://www.linuxnetworx.com/


an independent software vendor (like Schlumberger, Landmark). We can then think about creating the best infrastructure to run clients’ apps. So clusters are not commodity hardware? Hutchings – You should ask those who buy clusters. Both ExxonMobil and ChevronTexaco let their system administrators run MPI7 and were disappointed with the results – they should have bought a managed system. What of those oil companies who have made ‘skunk work’ clusters – buying boxes on eBay? Hutchings – Yes, Daimler Chrysler made a Grid out of 12 laptops. But how does that help their business? Today even early adopters should be concerned with total cost of ownership, productivity etc. We test our systems and code in Salt Lake City before delivery. So they run out of the box. One system, the sixth fastest in the world, with 2,800 Opterons and high speed interconnect was installed at Los Alamos. This was 6 days from delivery to doing science. How do you count the number of processors in a cluster? Hutchings – It’s defined by the scalability of an application. At 80-90 CPUs most systems max out as CPUs don’t have enough to do. What about the new ‘on-demand’ computing paradigm? Hutchings – That’s what we are doing at Los Alamos. Some jobs run on 2,800 CPUs – some on 16. There is a shift in focus from the number of CPUs/$ to productivity/$. Interconnect is also crucial. Oil and gas often uses Gigabit or hi speed Ethernet but there are much faster options; Quadrics or Infiniband. Multicast software upgrades allow you to radically change configurations according to work load. This resource/node allocation lets you do your re-engineering at night and GigaViz interactivity during the day. The new pre-stack focus represents great potential for cluster use. See also Fraunhofer – cluster-based visualization delivering PV-4D to oil and gas à la Gigaviz. What happened to the Itanium? Hutchings – The Opteron has grabbed the Itanium market. Intel has reacted by providing a 64 bit Xeon. Opteron has great memory access. Itanium may take off in 12-24 months. There will be a 2,200 node Xeon 64 at the DoD in September. The world is moving towards central computing and data servers, with distributed clients. Same for visualization where the key is configurability, multi-case provision, scheduling, upgrades. Contact - Guy Guertitz [email protected] - www.linuxnetworx.com

Mark Bashforth, Andreas Hatloy – Roxar Bashforth - Roxar has changed dramatically in the last 12 months. I joined in Feb 2003 from IT (hardware – non oil sales and marketing). Roxar used to be very product-focused, not problem focused. So we are changing to a customer-centric focus, solving problems and addressing needs. We are also updating our portfolio, staying innovative but also emphasizing integration and ease of use. Where do you fit in with the major vendors?

7 Message Passing Interface – protocol for cluster administration – see also MPI Forum.


http://www.lam-mpi.org/

mailto:[email protected]

http://www.linuxnetworx.com/

http://www.mpi-forum.org/


Bashforth – In terms of integration and multi-platform openness, the oil and gas lags. So we are taking advantage of this. The reality is that the two dominant players don’t want openness – so there is an opportunity here for Roxar. Hard to achieve if S&L don’t want to play? Bashforth – This is not rocket science. OpenSpirit is a good example of what can be achieved – except that Landmark doesn’t want to play. Also Roxar’s own Open File Format – OFF can be used for data transfer, or the RMS Open API. This shows that a small company can achieve such things. Our new software reflects ease of use and our commitment to customers.

Figure 2 New release (V7.3) of IRAP RMS

Hatloy – I re-joined Roxar in August. There have been three major releases in one year. With V 7.3 of IRAP, our portfolio is expanding. Our offer goes from mapping to flow simulation. The Integrated simulator is used as screening tool. What commercial impact has the built-in simulator had? Hatloy – It has been good for current clients – and is bringing us new clients. Bashforth – This has been a great success for a company of our size. Statoil and BP are moving to integrate the Tempest simulator. Tempest is also moving into flow simulation. But there is still a strong demand for a stand alone simulator. So what exactly is in RMS FloSim? Bashforth - A Black oil simulator plus some bells and whistles. And what is new in 7.3? Hatloy – We have revamped the mapping, which is very fast; a 10 fold speed improvement. Also new is 3D well correlation model with horizontal and highly deviated wells. Models can go from simple to very complex as field matures. We are moving further to seismic integration. Bashforth – The other news is that our ‘right time’ RT product hardware division is evolving its ‘smart field’ DACQUS product with new inter-application synergy. This will be integrated with the RMS portfolio. We have also been doing WITSML RT visualization for Hydro. Is WITSML really deployed on rigs – or just in a test environment?



Bashforth – I’m not sure about rig site deployment. We are testing our WITSML implementation with Baker Hughes. What’s the timeframe for rig site modification of software? Bashforth – This is quite a big task! We already have fast cycle time : 12-18 months release cycle and 3 year roadmap. But our quality focus slows things down some. Is there a field project where RT is happening? Bashforth – Bids have come up, but there’s a lot of hype. Statoil’s Valhall 4D seismics and real time screening of the waterflood is a reality. Huldral is Statoil’s test bed for RT operations monitoring. What about computing with clusters? Bashforth – We don’t but clients can using remote execution. Actually we see more demand for storage but the move to 64 bit Linux/Windows will be a major paradigm shift. Sun and SGI have priced themselves out of the market. Roxar has partnered with Microsoft on 64 bit computing – which will split into two camps – Linux and Windows. RMS 7.3 is a big step forward with first official Linux version (32 bit). What of the Itanium? Bashforth – There are issues with Itanium at the present time. So we will probably go with Opteron. There is also a groundswell of PC-based technology such as Direct X – leveraged in a new fracture solution. What else is of interest in the PC world? Bashforth –COTS8-type collaboration data mining such as Cognos although this is not deployed by Roxar. Data access is very primitive in the industry still …I can’t tell all… Windows or Linux? Bashforth – Both. We will not force our customers’ hands – probably there is more rapid take up in Windows. Hatloy –The industry has a long way to go. GUI, ease of use, deployment and access to distant data. We have something in the pipeline for next year… Market share? Bashforth – IRAP/RMS has largest market share for reservoir model – over 50%. What’s the killer software combo for your clients? Bashforth – Some consultants work with RMS and Excel spreadsheets. This can be done with a scripting language to move data to Excel and into Crystal Ball or @RISK.

8 COTS : common off-the-shelf



Figure 3 New release (V6.1) of Tempest Simulator

What are Tempest’s selling points? Bashforth – It was developed by the simulation community based in Oxford UK. Contrary to widely held belief – Eclipse does not rule the world. Technology and IT-wise, RMS is roughly equivalent to Eclipse, but Tempest has a cost advantage. Dave Ponting came from the Eclispe team; Tempest was developed by ex-Eclipse folks. Today, Tempest is mainly sold in Russia but we are taking it to other markets. It has been successful with consultants because of its ease of use and speed (due to a Fortran to C++ port). Contact - Jeanette Halford [email protected] - www.roxar.com

Murray Roth9 – Landmark. Oil ITJ—The announcements of the Prospect Generation Engine and Field Development Engine sound great. What are these and how much is re-packaging of existing solutions? Roth—OK, let’s cut to the chase! The big change is the expansion of OpenWork’s scope from G&G to engineering with the OpenWorks Engineering Data Model (EDM). This is leveraged in products like Asset View and Well Planning—making new workflows possible. These are underpinned with synchronization and data management tools for working with both Unix/Linux-based OpenWorks databases and EDM on Windows. This has involved a new focus on IT/hardware components and we are working with Sun, SGI, Intel/IBM to offer complete ‘shrink-wrapped’ IT systems. Oil ITJ—Shrink-wrapped? Roth—We have been working with IBM and Intel on IT ‘templates’—addressing workflow bottlenecks such as seismic processing. Systems are tuned for load balancing and to assure data management across the workflow. These can include Myrianet switches from NetApp and United Devices’ Grid computing solutions. The

9 This interview appeared in the June 2004 issue of Oil IT Journal.



http://www.roxar.com/

http://www.oilit.com/


Grid is very applicable to reservoir modeling—you don’t need big NUMA machines any more. The flexible IT model, as deployed in Abu Dhabi, is part of a global network of Landmark Asset Management Centers leveraging our on-demand agreement with IBM. Oil ITJ—Where else are these available? Roth—These need to be locally available, we don’t want to separate data from the CPU. IBM has centers in Abu Dhabi, Poughkeepsie and France. Oil ITJ—These are Itanium-based? Roth—No they use Xeons. Oil ITJ—What’s the ‘best thing’ chez Landmark since Magic Earth? Roth—Auto Imager is a good candidate for that categorization. This is new automatic seismic processing technology, automating velocity analysis. We partnered with Calgary-based Data Modeling Inc. to develop these image-driven techniques. One 80 million trace land 3D survey, which would have taken six weeks in a traditional workflow, took two days with no human intervention. The technique is also great for pressure prediction and AVO studies. The integration of processing with interpretation has now been accepted by our larger clients—particularly ProMagic’s use of GeoProbe in processing. Oil ITJ—Is ProMagic a killer app? Roth—ProMagic sells well to GeoProbe customers—big oils and NOCs. We have also seen interest in Well Seismic Fusion reflecting the changing role of AVO analysis, accessing pre-stack seismic data during the interpretation process. We are working closely with Statoil in this area. Oil ITJ—What is Landmark doing in knowledge management these days? Roth—We have had some success with our Team Workspace portal but we try to avoid the ‘portal for portal’s sake’ mentality. Clients get better results starting with a data management focus. Here, Open Explorer has been replaced by Power Explorer, along with WOW for QC which now offers thumbnail displays of data. But really, data management will never be a ‘shrink-wrap’ application. Oil ITJ—Still using Java? Roth—Yes, in the context of a heterogeneous platform. Java gives platform independence that matches the state of the industry today. DecisionSpace runs on both Windows and Linux. We use .NET in isolated engineering applications but get more flexibility and less risk with Linux. Oil ITJ—What’s Dave Hale up to; how did the atomic mesh work pan out? Roth—He is presenting some interesting work on atomic-mesh derived ‘tanks and tubes’ which are used to perform a simple simulation of subsets of the reservoir—to high grade modeling options. Oil ITJ—Will you be productizing atomic mesh? Roth—Some European customers are looking to form a consortium around this. The tanks and tubes work may prove a quick win for the technology—in the seismic to simulation workflow. The technology is leveraged in the new DecisionSpace Nexus —next generation unstructured simulator. This uses a tetrahedral grid, and reservoir simulation is modeled along with surface facilities. Nexus was a joint development with BP.



Figure 4 Visualization of subsurface and facilities in Nexus.

Phil Neri – Paradigm

Figure 5 Paradigm’s Reservoir Navigator leverages disk caching for multi-GB datasets.

There is a move in the visualization marketplace to the desktop. Using crafty techniques, visualization is possible on a laptop – with 1:20 or 1:40 memory to data ratios. This compares very favorably with established techniques as used in Voxelgeo, and Magic earth – which require all data in memory. Can roam very large regional data sets. Also added 2D and 3D interpretation. Released on IA32 Linux, Itanium 64, Solaris, AMD Linux. Geolog now runs native on Windows – got rid of Nutcracker. Paradigm also got Coherency through Core RTD division acquisition. Direct X? Neri – Wait and see – not sure this is right for data environment. Version 2.0 of Reservoir Navigator uses a cache in memory to buffer data. The ratio of data volume versus buffer size has been tested at factors of up to 40 (so a 500MB buffer will address 20 GB of seismic). While Reservoir Navigator can view seismic



traverses and slices, blend multiple attributes or co-visualize them, support interpretation, etc.. it does not offer the advanced features of a full voxel-based solution such as VoxelGeo (opacity, sub-volume detection, etc..). Reservoir Navigator includes Paradigm's multiple-resolution format for seismic volumes and grids. Users can specify up to four sub-sampled volumes (or maps), and switch between them. Activities such as the preliminary interpretation of major markers can thus be conducted on the sub-sampled seismic and grids with very high throughput, with the ability to switch to higher resolutions for complex areas. Reservoir Navigator is the common 3D window into most of Paradigm's Geoscience solutions. Including Geolog. Epos 3.0 introduces access to seismic through the ‘brick’ format, offering equal response times in all directions. Also new are Paradigm’s new seismic section interpretation and mapping tools, SeisEarth and iMap. Contact - Phil Neri [email protected] - www.paradigm.com

Ali Ferling – HP HP has partnered with Schlumberger Information Solutions to port software to Linux. The idea is to move towards a commercialization agreement to deliver a true ‘plug & play’ upstream computing environment. HP’s Project SEPIA builds on High Performance Computing work done for the US Department of Energy (the ASCI Views project) using COTS technology for high end visualization test. The techniques are applicable to both simulation and seismic visualization. Sepia V3 will be out real soon now and promises to be 10 times faster for 25% price of current solutions. The cluster-based solution is claimed to be 6 times faster than high end SGI. Scalability is easy because you can start small, then add nodes. Oil and gas is one of the most demanding markets for visualization. The solution was designed for Itanium 2. But because of current performance issues with Itanium, HP is testing on Opteron and will deploy on Itanium. Reservoir simulation on clusters? To date the problem has been with middleware tools. Now with SCALI MPI this will change with scalability and performance. Still limited to 16/32 processors per cluster because of algorithms. Seismics is embarrassingly parallel and can use 1,000s of nodes in a cluster. Contact Ali Ferling - [email protected]. Website www.hp.com



http://www.paradigm.com/


http://www.hp.com/


Demos

Cougar – Schlumberger/IFP

Figure 6 The IFP’s Cougar, now marketed by Schlumberger.

Demo showed use of Petrel to drape attributes onto seismic horizon for data qc. Geostatistics are used to distribute parameters before upscaling the model and exporting to Cougar. Here parameters are defined for the sensitivity analysis. Nine parameters and 32 runs – to be done on IBM/ Platform Computing LSF cluster in IBM’s HPC center in Montpellier. A machine crash somewhere interrupted the demo. The IBM ‘On Demand’ solution uses the Open Source Globus toolkit for Grid Computing. Cougar output plots field oil prod total vs. time in days. Tornado plot shows sensitivity. Cougar runs Eclipse black oil simulator 32 times. 32 ‘experiments’ run on different data deck with parameters selected using ‘experimental design’ – a statistical technique for unbiased selection. Cougar offers pre/post processing for statistically significant driving of Eclipse and ‘analysis filtering’ for risk analysis. Next, non-sensitive parameters are de-activated in the data deck and re-run for optimization. The result is a ‘response surface model,’ (RSM) a 5D polynomial in the results space. Simulation runs on a ‘proxy model’ built from a interpolated surface over the true model data. Currently you can’t visualize RSMs, this will be available in the next version.

GigaViz – Schlumberger After our less than enthusiastic comments re GigaViz (GV) in our report from the AAPG, we were invited (summoned?) back for another look. GV is not just a visualization system. It puts 2D/3D inline, cross line ‘all in one place’. Schlumberger claims GV is the best, fastest tracking tool in the business. You can interpret a 40GB volume on a laptop. All graphics processing is done on the cluster and pixels that change are sent to the client. The idea would be for an oil company to install a big graphics processing cluster – and give G&G folks low end machines. A demo with a 1400 km. sq. dataset showed that GigaViz does do Magic Earth type probes with fast pan and zoom. With a single seed it is possible to pick pretty well all the faults, although such unsupervised horizon tracking is far from perfect. GV offers arbitrary cross line, fault plane tool, random bendy surface/line slice through data. The amplitude tracker uses wavelet processing and there is a function for picking in very bad data.



How does GigaViz fit in the Schlumberger portfolio? SIS line-up

Petrel modern, smart system Eclipse, Geoframe, GeoViz – legacy software for big data sets Inside Reality, collaboration-focus VR

GigaViz core technology will be added to other SIS products above. Long term, GV was acquired for the technology. Short term use in GeoFrame and write back to GF. GV will be released as stand alone product with OpenSpirit. The guts of the tool (rendering, opacity, attribute handling) will form the basis of next generation seismics. This is ‘where SIS is heading’ with Project Ocean (.NET development). Today uses Qt for Linux/Windows compatibility. Also used by Western Geco for seismic data review and QC of very large datasets. Shell is going for the technology ‘big time’. Contact - Logan Dent [email protected] - www.sis.slb.com

Grid Computing – IBM Demo started with a major crash – not sure if this was software or the projection system – nobody seemed very concerned. Jaouad Alkhaliki said that grid computing is important to Shell because it offers access to unlimited resources. Alkhaliki, citing Shell’s Jaacob Buur said that for Shell, the Grid is the foundation of the on demand operating environment (ODOE). Web Services also ran in the ‘self configurable’ Grid. ‘Changing hardware is not enough.’ The Grid looks out to IBM deep computing on-demand offering (DCoD). Deployers build a ‘re-usable software wrapper’ around legacy code. PSDM a good candidate for Grid and is part of IBM’s Grid@petroleum offering with guaranteed service levels, storage and computing on demand. Dynamic provisioning lets you add/remove resources as required. Globus also ran as did the IBM Dynamic compute factory on demand, as did grid@shell where a ‘wrapper provides Globus-based API. Elsewhere, IBM were vaunting the merits of its ‘Upstream Value network and Upstream Ecosystems10’. Available in Poughkeepsie, Montpellier and Houston (RSN).

Wireless devices in oil and gas – IBM IBM’s Symbol offering supports the intelligent oilfield with links to the sensor world. Data from the sensor network and RFID also ran. IBM new ‘sensors and actuators’ EBO – emerging business opportunity is linking assets to sensors. Wireless SCADA? Yes, have done pipeline monitoring for ChevronTexaco. But IBM is not a SCADA provider – more of a system integrator. Leveraging IBM Websphere MQ series with publish and subscribe messaging. Smart sensors and ‘smart dust’ will filter and correlate events so as not to ‘drown’ data events.

Seismic Image Segmentation – Dave Hale, Landmark A continuation of Hale’s work on seismic meshes. The seismic ‘image’ is segmented into a mesh of connections – splitting image along faults – with recursive cutting of

10 IBM is in a category of its own in the bullshit bingo sweepstakes!



http://www.sis.slb.com/


http://www.perkigoth.com/home/kermit/stuff/BullshitBingo/


the mesh. Hale showed the segmented image to a reservoir engineer who said– ‘those are our tanks’. The engineers were using tanks and tubes (see Glyn Williams, Mike King 2002(?)) to simulate reservoirs. Tanks are polygonal reservoir blocks and tubes equate to transmissivity. The mesh of connections can be simulated at different scales – to ‘get the basic plumbing right before working on the details.’ Uses imbricated tanks to do up/down-scaling. Tested on a major upscaling exercise, built a model with one tank per wellbore. Proved very fast compared with a fine grain model. Put injector in one tank and producer in another. Simulator gave good qualitative results and allowed for transmissibility of faults to be tweaked. Simultaneous seismic interpretation and fluid flow modeling.

Decision Space/Nexus – Landmark Fine grain geological model out of DS PowerGrid is starting point – then ‘in situ intelligent coarsening applied using ExxonMobil’s Residual Optimization method and flow information for upscaling. DP PowerModel allows view/simulation of Nexus results along with the gathering network – fully coupled implicit solution of subsurface and network. Also see surface imagery (of BP’s Wytch Farm field.) Lots more DS tools rolled into demo and VIP Data Studio used to fiddle factors – adjust relative permeabilities etc. Also ran OpenWorks Engineering Data Model.

Data Session About 40 present.

eEarth, an EU standard for borehole data – Jan Jellema, TNO 50% EU funded research program with partners BGS, TNO, BGR (DE) to develop a multi-usage geodata exchange format. The project sets out to free-up cross EU data access by solving language issues. The claim is that this will enable efficient start-up of cross border projects. Issues – in Holland there are ten types of peat, access is free in some countries, in others (UK) data costs money, in Poland, you need written permission from the Ministry. Analog projects cited are WITSML and the Australian XMML (CSIRO) – a mining industry initiative that builds on GML “taking off now”. Will facilitate eCommerce via a multi-language, multi-country portal to national databases. 35 data fields selected for core exchange format. PDM access and mobile computing also ran! Said to stimulate third party products.

Q&A How much is this project costing? € 1.5 million11.

Information aggregation – Nathan Balls, Petrosys Entirely commercial presentation of Petrosys – see Technology Watch report from the 2003 PESGB Data Management.

Q&A (Herb Yuan POSC chairman) – What of data standards?

11 Again, one has to question the wisdom of putting public money into an XML schema which, in the unlikely event it is ever used, would benefit a handful of large civil engineering contractors.



Petrosys is a member of POSC and PPDM – but our work is client-driven, we can’t wait on standards. Isn’t some standards discipline a good thing? No doubt – but here the issue is interfacing with legacy data stores. What of duplicates across legacy stores? The tool was designed for data clean-up through query. What of OpenSpirit compared to Direct Connect? Petrosys is not OpenSpirit-enabled. There is overlap, but customers are happy with the PetroSys solution.

Lessons of a merger – Philippe Baldy, Total Total merged with Fina in 1999 and Elf in 2000. Baldy asks “what’s next?12” The result was technical geoscience data stored across four different systems. All three companies used Schlumberger’s LogDB with well reference data in Finder – but there were significantly different customizations. Elf had seismic data on nearline robots and on shelves, Total was shelves only. A major remastering project was underway with Elf at the time of the merger. Physical data was gathered and indexed (DocEP for the 676,000 reports, DocXplo for the 2.4 million ‘other’ items). Migration took 11 man years of effort over a 14 month period.

0% 20% 40% 60%

Effort

SupervisionMappingQCData LoadingData Matching

Topographic and navigation data was stored in five systems and collected to Finder.

0% 20% 40% 60%

Effort


Total’s LogDB now holds 120,000 references, and nearly one terabyte of data – it is believed to be one of the largest in the world. Log data move proved a lot easier – all data from and to LogDB.

12 Speculation in the financial press is that it might be Shell!



0% 20% 40%

Effort


Key issues

The common database (LogDB) and shared rules for migration made the merge easy.

Data transfer/load a small part of the whole project. Data match and clean-up very important. A large footprint of pre-merger history and culture is preserved in the

corporate database. The best QC does not mean error-free! The final QC stamp will be given by

users. This has led to a new information management approach for Total, focusing on how to improve and ease future acquisitions. A survey found that 60% of corporate data and 30% of interpretation data was re-used. Subsidiaries ‘know’ their data better than HQ. Don’t duplicate effort. Leverage cross border assets (North Sea, Middle East). Baldy announced Total’s upcoming GADAM Project – Geoscience Affiliate Data Management. Affiliates are responsible for their own and non-operated data. HQ is responsible for ‘orphaned’ data, new ventures and corporate.

Portal Access Finder LogDB eSearch

Data capture in subsidiaries One TB data not official yet – physical document

store from Schlumberger/Iron

Mountain. In the future there will be a fully-integrated system, based on Finder, connected to document systems and OilField Manager, FieldView, DCS and FVIM for automated data capture. The Geoscience Web Portal will integrate PetroVision, OpenWorks, GWIS, IRIS21, BD Petro and DocXplo. GADAM is rolling-out in Indonesia, Nigeria and Angola. Still need to solve synchronizing with HQ databases. An RFI has been issued for the Portal. A ‘community of cultures’ will enforce rules and create a partnership between dedicated local data managers and HQ. Need overlap and cooperation with other domains – drilling, production etc.

Q&A What was the business impact? We know that management is keen – but we’re not sure that financial cost justification numbers are really true! Total is saving money here, but an overall business case was not performed. A huge amount of time and money is being saved by improved data availability.



DecisionPoint is used in subsidiaries – what about HQ? We are in the process of issuing an RFI for the HQ portal. But HQ needs are not necessarily the same as subs. Data use is different. Why is interpretation information use not higher than 30%? Interpreters may not trust previous work, may want to dig deeper into the data or go into more detail.

Shared earth model and portals – Piantanida, ENI E&P ENI’s project is for a shared earth model across ‘distributed’ people, subsidiaries and HQ. 3D Data sharing spans the application workspace (GeoFrame, OpenWorks, Tigress, FlowMap, Petrel and Eclipse). Issue is that subs don’t know what ENI best practice is. Apps work on shared data with uniform data discovery interface. Self service data management – download simple data types. A service desk provides complex data and project population. Full remote access is offered from ENI Norge, London and Ravenna over an ASP link. 2D is enabled by Citrix Metaframe, 3D by MIT ThinAnywhere. ENI is experimenting with a new Grid computing tool from Softricity, the SoftGrid System Guard. Data sources include Diskos seismic (Stavanger), Iris21 (Milan) EDMS (Milan, for studies), @Concession and EDIN/EDGE (IHS Energy). The Portal gives access to data in Oracle and non Oracle (e.g. Recall via ODBC – which was pushed by ENI). Also ran, PowerExplorer XML services and DecisionPoint XML services to the Schlumberger DMC, C&C Reservoirs, and ENI Corporate datastore. From the Portal, users can drill down to Finder – with tabular data access. HQ offers Recall ‘self service’ data management. Query through CDA, Petrobank or order with FTP delivery. See paper from Shared Earth session. OpenSpirit used wherever data types allow. OpenSpirit has been extended to include a zonation model. A specific Tigress – Petrel zonation tool was shown. Best practices are stored in the technical ‘Know-how’ Portal as workflows for reservoir model building. Project directory uses Lotus Domino. Project Diary synchronizes with k-base of best practices. This is based on E&P OnLine (Accenture) and SAP Portal 5.0. XML used to sync data and launch apps. Workflow management includes description of tasks, best practices, database access with drill down to specific tasks such as facies identification. Click on task and app fires-up. The Portal is now considered mature and is being adopted by ENI E&P. The SEM tools and Know-how Portal are still in the pilot phase.

Q&A What has been the business impact? The time taken to prepare a study time has gone down from 2 weeks to one day. Quality is up. How many people involved? Six or seven.

Earth science decision making – Troy Wilson, Geosoft Geosoft has been doing mapping and data delivery for the mining industry for 20 years. In mining, one in 20 prospects turns into a mine. Much data is lost, time wasted – knowledge lost and companies have to redo work. There has of late been a data management renaissance – with renewed interest in a central database. Due to rising mineral prices – huge increase in bringing old prospects to life. Issues with data


http://www.softricity.com/home/index.asp


access, productivity. Geosoft offers an ArcInfo-based solution and another data access layer – DAP (geoDAP?) – see website. Spatial catalogue and web Browser – Open API.

Exhibitors

Atos Origin, ISPoo3 release 10.2.

Figure 7 Well velocity analysis in ISPoo3

Sattlegger (which has now become Atos Origin13) has released version 10.2 of its ‘interpretive seismic’ package, ISPoo3. ISPoo3 comprises tools for 2D and 3D depth migration, modeling, velocity processing, mapping, reservoir volumetrics and synthetic seismogram computation. A new velocity processing module assists the user in finding the best fitting velocity model for migration and modeling. Contact - Helmut Egbers [email protected] - www.atosorigin.de

CGG, EYE-3D. CGG is playing catchup with I/O’s marketing machine and has rebundled its 4D/4C, high res and field instrumentation into a shiny new ‘EYE-D’ offering. According to the PR ‘The EYE-D offering brings together CGG’s suite of enhanced seismic solutions for application throughout the life of the reservoir’. CGG claims a lead in high-resolution, 4D and seabed seismic. Contact - Christophe Barnini [email protected] - www.cgg.com

13 Sattlegger was acquired by CCI GMBh, itself bought by Sema Group. This was later acquired by Schlumberger before being divested to Atos Origin.



http://www.atosorigin.de/


http://www.cgg.com/


dGB New fault and horizon tracking.

Figure 8 Simultaneous horizon and fault tracking from dGB.

dGB’s OpendTect tracker tracks faults and horizons simultaneously to ensure geometrical consistency. The software module is open source and part of dGB’s OpendTect initiative. Contact - Herald Ligtenberg [email protected] - www.dgb-group.com

Survey Optimization from enGenius 3D survey design and optimization software. Used to arbitrate between different ways of shooting survey – answers question like ‘how to extend survey?’ ‘How much extra time needed?’ ‘How shoot with port feather?’ etc. CGG has acquired the software for all its vessels and it is also used by Veritas. Contact - Stephen Fleming [email protected] - www.survopt.com



http://www.dgb-group.com/


http://www.survopt.com/


ERM.S, SeisQuare and VSquare.

Figure 9 VSquare’s spatial quality index shows anomalous velocity picks.

VSquare is a geostatistics-based spatial QC tool for seismic velocities. The software uses a geostatistically derived spatial quality index to supervise stacking velocity data sets. Contact - Samuel Chalus [email protected] - www.erms.fr

PV-4D SeismicPRO – Fraunhofer Institute

Figure 10 PV-4D uses Linux cluster for seismic display.

PV-4D turns Linux clusters into high end dynamic visualization system. Rapid access to terabytes of stack or pre-stack data with high speed rendering and stereo display of multi-volume data. Cluster management in association with Linux Networx. Up to 256 rendering nodes supported. Gigabit, Myrianet or Infiniband interconnect. 32-bit Xeon or 64 bit Opteron.



http://www.erms.fr/


Contact - Franz-Joseph Pfreundt [email protected] - www.pv-4D.com

GeoVariances, ISATIS 5.0.

Figure 11 Plurigaussian simulation in Isatis 5.0.

New features in ISATIS 5.0 include a redesigned variogram map, plurigaussian simulations for modeling complex reservoirs (channels, reefs, bars ets.) and new interfaces to LAS, SEG-Y, ODBC data. A direct link to embedded Isatis applications in Gocad and Beicip Franlab’s RML. Contact - Brigitte Hogan [email protected] - www.geovariances.com

Hampson-Russell



http://www.pv-4d.com/


http://www.geovariances.com/


Figure 12 Hampson Russell Colored Inversion Operator

Hampson-Russel was previewing its upcoming ‘CE7R1’ release which will update the entire Hampson-Russell product suite. Highlights include a new colored inversion window for STRATA, a well log database, GeoView, a product launcher and a new 3D viewer. Another new product, developed along with ENI, uses a stochastic approach to quantify fluid probability with AVO. Contact – John Coffin [email protected] Website - www.hampson-russell.com

INT, INTViewer/3D.

Figure 13 INTViewer now commercial.

INT is branching out into end-user targeted software with the commercial release of INTViewer for pre-stack and attribute seismic visualization. INTViewer offers a ‘simple workflow’ for pre-stack data manipulation – ‘replacing an entire suite of tools from (say) Landmark’. Provides horizon picking tool with upload of horizons to OpenSpirit apps. INTViewer is designed to efficiently visualize and interact with very large multidimensional datasets. Both prestack and post-stack seismic traces can be accessed through any key in the data such as inline, cross-line, CDP, offset, realization, etc. Easy navigation along any direction in the data is possible through menu shortcuts. A transpose utility is available for accessing 3D data as time slices for display as a map or in a 3D display. A new Wellbore Schematics tool has been developed with ChevronTexaco – using WITSML data exchange. Contact - Paul Schatz [email protected] - www.int.com

IPRES, IPResource and IPRisk-Field.




http://www.int.com/


Figure 14 IPRisk petroleum risk evaluation tool.

IPResource is a resource/reserves database for company reporting (Annual Reports, SEC and Key Performance Indicators). Developed using Business Objects – originally as add on to Schlumberger VOLTS, but now works with other production systems. IPRiskWELL integrates the disciplines of reserves estimation, production profile generation, drilling and economics in one model. IPRisk is a new petroleum risk evaluation tool for use at well, field and portfolio level. Used by ChevronTexaco, ConocoPhillips, Marathon, Norsk Hydro and Statoil. Contact - Erlend Glommen [email protected] - www.ipres.com

Knowledge Systems, DrillWorks 2004 R2.0.

Figure 15 Knowledge Systems DrillWorks 2004 R2.

Drillworks 2004 R2 introduces a geopressure database to collect well-and basin-level geopressure and wellbore stability data. Drillworks allows operators to perform



http://www.ipres.com/


uncertainty analysis on geopressure predictions, understand the uncertainty in calculations, and aid decision making for critical well planning, mud, and casingprograms. Contact - James Webster [email protected]

Website - www.knowsys.com

Landmark, Tanks and tubes simulation

Figure 16 ‘Tanks and tubes’ simulation from segmented seismics.

3-D seismic image segmentation (see Hale Demo above) is used to build simple as geologic layers and fault

ic es obtained

odels.

reservoir models with a small number of flow units, suchblocks. These are the ‘tanks’ which are connected with ‘tubes’ of variable transmissibility. Reservoir properties, such as pore volumes of tanks and transmissibility of tubes can be adjusted to test the impact of different seisminterpretations on fluid flow. Numerical experiments suggest that propertifrom such coarse models can be used to constrain properties for more detailed mContact - Dave Hale [email protected] - www.lgc.com



Magnitude, Microseismic monitoring

Figure 17 Microseismic monitoring with Magnitude (and VS-Fusion)

Magnitude performs microseismic monitoring of gas storage facilities, heavy oil, tar sands production and hydraulic fraccing. Sweep efficiency, frac success and water flooding can be monitored during lifetime of field. Contact - Christophe Maisons [email protected] - www.magnitude-geo.com

Mercury/TGS, VolumeViz5. TGS (now part of Mercury Computer Systems) announced version 5 of VolumeViz, its volume rendering technology for large data sets. VolumeViz is used by most all upstream software vendors (Landmark, Schlumberger, SMT, Jason, Paradigm, Roxar etc) and many majors to visualize 100GB plus datasets. TGS comes as a C++ library with Qt support for the GUI. VolumeViz is ‘thread safe’ and can be multi-threaded across multiple processors and 32 or 62 bit graphics engines. Contact - Laurent Coureau [email protected] - www.mc.com

Network Appliance, SpinServer The Network Appliance SpinServer solution is a flexible, high-performance grid storage implementation that scales to hundreds of file servers within a single SpinServer cluster. Specifically, a SpinServer cluster can grow to encompass as many as 512 servers and a total of 11,000 TB of storage. This scalability enables SpinServer to provide sufficient aggregate bandwidth to support a grid computing environment encompassing thousands of Linux compute servers. Storage pools contain one or more virtual file systems (VFSs). From the perspective of an NFS client, there is a single NFS export, which represents the entire global file system that is exported by the SpinServer cluster. Behind the scenes, the global file system is actually implemented by the collection of VFSs within the SpinServer cluster. Contact - Vaughn Miller [email protected] - www.netapp.com



http://www.magnitude-geo.com/

http://www.tgs.com/

http://www.mc.com/


http://www.mc.com/


http://www.netapp.com/


Neuralog, NeuraDB and Neuralog ‘Plus’

Figure 18 NeuraLog+ automated digitizing.

Neuralog has re-launched its flagship automated digitizing product NeuraLog, as Neuralog+. Neuralog+ brings a complete overhaul of the user interface to Windows multi-document environment and interactive digital log editing. LAS data features include customizable templates, depth shifting, baseline correction and TVD conversion. Raster calibration export and a ‘Virtual Light Table’ assist in QC. NeuraLog+ V6.0 will incorporate an image straightening function for skewed logs. Neuralog Plus has speeded paper log digitizing up by a factor of 10. Contact - Kenneth Land [email protected] - www.neuralog.com



http://www.neuralog.com/


Neuro Genetic Solutions, Neural Networks and Committee Machines

Figure 19 Porosity Modeling by human (left) and neural net (right).

Generic neural network application can be applied to any data types – e.g. well logs. Offered as consulting service. Same university background as Decision Team. An expert interprets porosity from log data and ‘teaches’ the neural net. Committee Machines hook multiple neural networks together to enhance results. Can be used to synthesize missing well log data. Contact - Sepp Steinlechner [email protected] - www.bestneural.net

Oedegaard – New release of Isis 3D inversion Ødegaard’s new generation of ISIS simultaneous 3D globally optimized inversion provides the ‘first genuine method’ for converting all AVO (amplitude versus offset) seismic data into high resolution rock properties. The new inversion technology improves the accuracy of both acoustic impedance and Poisson’s ratio/shear impedance to give useable estimates of density. Other Isis innovations include the ability to parameterize for any combination of acoustic and shear wave impedance, Poisson’s ratio and Vp/Vs data. Contact - Dave Davies [email protected] - www.oedegaard.com




http://www.oedegaard.com/


P2ES, Enterprise Upstream Suite.

Figure 20 Tobin Upstream Enterprise P2ES is working on a project to combine the recently acquired Tobin LandSuite with its vertical ERP system Enterprise Upstream. So far it is ‘integration by brochure’. Contact - Maggie Seeliger [email protected] - www.p2es.com

Paradigm, ResNav.

Figure 21 Paradigm’s Reservoir Navigator leverages disk caching for multi-GB datasets.

Version 2.0 of Reservoir Navigator uses a cache in memory to buffer data. The ratio of data volume versus buffer size has been tested at factors of up to 40 (so a 500MB buffer will address 20 GB of seismic). See interview with Phil Neri above. Epos 3.0 introduces access to seismic through the ‘brick’ format, offering equal response times in all directions. Also new are Paradigm’s new seismic section interpretation and mapping tools, SeisEarth and iMap. Contact - Phil Neri [email protected] - www.paradigm.com



http://www.p2es.com/


http://www.paradigm.com/


Petrosys, SDE raster-vector integration in version 14.4.

Figure 22 Petrosys 14_4 features direct display of SDE raster images.

Petrosys has extended the coverage of data types that their mapping system accesses with the inclusion of raster images from Arc SDE data store. This builds on work previously done to integrating SDE vector data with other E&P map data layers. Contact - Nathan Balls [email protected] - www.petrosys.com.au

Scandpower – MEPO history matching tool commercial

Figure 23 Scandpower’s MEPO history match tool.



http://www.petrosys.com.au/


MEPO runs as a front end to the reservoir simulator to test out multiple hypotheses to

ch ently

[email protected]

optimize history matching prior to full-scale modeling. Multiple parameter optimization avoids correlation errors when single parameters are optimized sequentially. Mepo uses Baysean analysis to check the validity of a match. Sutechnology promises an ‘objective’ match to production data. Scandpower is currrunning pilot studies on a 22CPU HP cluster running Linux. The pilots test different optimization methods included in MEPO with a link to the Schlumberger’s Eclipse simulator. Other simulators which have been tested include 3DSL (Streamsim Technologies), Fronsim (SLB), MORE (Roxar), as well as Scandpower’s own IPOSand a core simulator from Total. Contact - Ralf Schulze-Riegert raWebsite - www.mepo.com

Sercel – Nomad 90 T Vibrator

Figure 24 Sercel’s biggest and best vibrator.

) all-terrain tracked vibrator The new 90,000 pounds force (400 kNContact - www.sercel.com.




VS Fusion, VS3 VSP collaboration with Magnitude.

Figure 25 Microseismic activity detection in multiple wells

CGG/Baker Hughes joint venture VS Fusion has extended its offering to include microseismic monitoring in cooperation with Magnitude. Contact - Neil Peake [email protected] - www.vsfusion.com

Weatherford, Clarion 4D permanent monitoring.

Figure 26 Weatherford's in-well seismic monitoring station.

Weatherford launched its ‘Clarion’ 4D permanent in-well seismic monitoring at the SEG last year. Weatherford’s aim is to make seismic more ‘P’ (production) than ‘E’ (exploration). The Clarion permanent multi-component system is being trialed in the



http://www.cgg.com/


Izaute gas storage field in southwestern France. Both time-lapse (4D) and Vertical Seismic Profiling (VSP) surveys are being recorded by a permanent five-station, three-component array. Extended microseismic monitoring is also being recorded. The study is used to map gas-water contact variations using VSP and to correlate microseismic events to gas injection and production activity. Contact - Christine McGee [email protected] – www.weatherford.com

EAGE Conference CD-ROM This CD of the proceedings is possibly the worst ever. There is no search facility. The software doesn’t recognize arrow key navigation and ignores double mouse clicks. Result is that it takes seven mouse clicks to navigate to a paper - when you know exactly where it is. This is then thrown up in a new instance of Adobe Acrobat - so when you are through reading you have to start over clicking and cussing. Masochists can apply for a CD on www.eage.org.

Knowledge-driven shared earth models, Rainaud (IFP) et al. The paper describes an ‘alternative approach’ for Shared Earth Model (SEM) building which captures interpretation metadata via a ‘Geo-Ontology’ and ‘abstract descriptor’, the Geological Evolution Scheme (GES). Annotated geological interpretations can be shared between applications and stored for re-use. The EpiSEM project was funded by the EU.

Internet production database, Tchistiakov (TNO-NITG) et al. An internet database containing fluid flow simulations from over 32,500 ‘parametrically distinct’ models of shallow marine reservoirs was created as part of the EU-supported SAIGUP. The project aims to quantify the influence of sedimentology, structure and up-lift.

Listening to the waves, Taner (RSI) et al. Uses joint time frequency analysis with unsupervised neural networks to produce seismic lithology maps. Automatic event recognition and classification simulates human hearing perception.

Virtual reality for rock microstructure visualization – Kayser et al., Schlumberger Uses Schlumberger’s Inside Reality Virtual Reality software to visualize the internal structure of cores. Densely spaced microfocus computer tomography (µCT) images are digitized for display in the VR Cave.

Managing subsurface work processes – Naess et al., Statoil. Statoil’s experience from the Heidrun Field has underlined the importance of the single common database for subsurface data. Initially this was through the SCORE project, now augmented with ‘rigorous rules and guidelines’ for data capture. Data volumes and model complexity mandate a comprehensive data organization and management policy. Faults are uniquely labeled and captured for subsequent work processes including geological models, simulation models, well plan prognosis,



http://www.weatherford.com/

http://www.eage.org/

http://www.fault-analysis-group.ucd.ie/Projects/SAIGUP.html


drilling reports etc. ‘Firmly defined’ work processes for commonly performed tasks have also been standardized. Such techniques underpin remote drilling operations from Statoil’s new Onshore Support Center in Stjørdal.

Integrating Skills, Processes and Technology – Naylor, Shell Shell’s new operating model for E&P has exploration organized into a global business, with ‘unambiguous, single point accountability for performance, portfolio and resources.’ This model contrasts with the autonomous asset-based approach. ‘Competence based assessment’ is the basis for professional staff development and for project team building. Staff support is provided by web-based tools for ‘knowledge sharing and enquiry,’ leveraging the expertise of the global exploration community. Technology is key – including basin and charge modeling, rock property data, seismic imaging, quantitative interpretation (QI) techniques, and volume interpretation of structure and stratigraphy. Cross-discipline integration is facilitated by 3D visualization and virtual reality centers. The Real Time Operations Centre ‘leverages global expertise’ to enable rapid decision making even in remote locations.

Attracting and retaining technical personnel – Lloyd et al. (Independent) Industry is ‘graying’ while in-house training programs and research centers are gone. In 10 years, ‘most of the people attending this conference will be gone.’ Recent studies show students steering away from the oil and gas industry because it is seen as ‘low tech’, ‘environmentally unfriendly’ and offering an ‘uncertain career future’. To meet such challenges an ‘educational-industrial continuum’ is required. Companies need to manage future human resources as they do the subsurface. Universities should extend traditional degree programs to meet industry needs. Remuneration ‘while important assumes a secondary role’.

The Professional Society of 2020 – Baker, SPE and BP Baker started out by defining what ‘integration’ means – in the poultry industry! The SPE has been pondering such matters in its long range plan for the industry. Here, illuminating insights such as ‘gas production will grow faster than oil,’ ‘hydrocarbons will dominate the energy mix,’ ‘the move towards global standards will continue slowly’ and ‘the internet will grow rapidly.’ The implication of all this for professional societies is that ‘the bulk of professional society members will come from outside the US and EU.’ Real time data will ‘reduce the decision-making cycle time to hours and minutes.’ Asset team members will need ‘robust related job knowledge as well as a core discipline expertise.’ One major issue is whether we continue to be societies of individual members (as SPE), or of local societies (as AAPG) or of companies (as IADC). Baker believes that meetings are key – ‘people are gregarious; they like to come together.’ The development of virtual meetings is likely to impact professional societies’ revenue but ‘if we are to stay relevant in these days of restricted travel, we have to become a player in the virtual or distance-meeting realm.’

Value creation: an IM approach – Mehrabian (Schlumberger) et al. Measuring the return on investment (ROI) in technology is hard. Metrics are evolving that allow ‘rigorous calculation’ of value added and of the effect on discoveries. Business impact is reduced cycle time, improved quality of strategic decisions, and



the ‘value of standardized processes.’ Data management can have a large impact on exploration success – avoiding data loss and errors. Recently, the time to load a 30Gb 3D seismic was reduced from 10 days to 9 hours.

Automatic Velocity analysis – Stinson et al. Data Modeling Inc. Automated high density migration velocity estimation computes velocities at every CMP and time location. The Auto Imager approach was demonstrated with 3D real and synthetic examples and tested against the models. A 32 node Linux cluster ran the SEG/EAGE 3D Salt Model in 11 hours (60,000 velocity profiles). Results are ‘as good as or better than’ the human picked velocities.

Storage of interpretation results – Aliverti, et al. ENI. ENI’s Shared Earth Model (SEM) project aimed at providing G&G specialists with an interoperability framework where interpretation results are stored, indexed and restored at a future date. Open Spirit provided some ‘live’ data connections. Elsewhere, the ENI framework includes tools for data import, export and reformatting. A ‘Results Database’ (RDB) was developed by ENI to support most of the G&G interpretation data types. Results are documented with the geoscientist’s name, the date and comments. The RDB integrates with ENI’s Electronic Document Management System, for document storage and retrieval. Results are viewable over the ENI intranet.


European Association of Geoscientists and …2004 EAGE Conference and Exhibition European...

Documents

Transcript of European Association of Geoscientists and …2004 EAGE Conference and Exhibition European...