You are Home   »   News   »   View Article

Improving multi-disciplinary workflows with virtual models

Friday, July 14, 2017

SEG is running a joint industry 'Life of Field' project to improve integrated 4D workflows using virtual models - and is looking for more companies to join.

The Society of Exploration Geophysicists (SEG) Advanced Modelling Corporation (SEAM), the non-profit research and development arm of SEG, formed in 2007, is running a joint industry project to use virtual models to improve workflows, and along with the SPE, currently has a multi-disciplinary project happening in 4D, the 'Life of Field' Project.

To try to explain - one problem with seismic interpretation and reservoir simulation is that it is very hard to know how good your results are.

You can interpret the seismic, and generate nice computer earth models, and use these to build reservoir flow simulations. But since you don't know exactly what is in the subsurface, it is hard to know how good your seismic interpretation and reservoir models actually are. If they are completely wrong, they are worse than useless.

The idea is that by starting with a piece of 'virtual' subsurface that you know everything about - because it has actually been created by computer - you can then do 'virtual' seismic to see what seismic data you would gather if you did a seismic survey over it. Then you can put this seismic data through your normal seismic processing and interpretation workflows, and see if you end up with something similar to the virtual subsurface you started with.

'We know what the answer is, because we built the model,' says Mike Mellen, director of business development with SEAM.

Using methods like this, you can improve your algorithms and workflows, to develop the best way of putting together an interpretation of the subsurface which best matches reality.

You can keep tweaking (iterating) your workflows until you come up with a workflow that produces the earth model which is the same as the one you started with.

The project is also useful for education, helping people to better understand the dynamics of the reservoir and the interpretation process.

The seismic modelling includes a full range of high density seismic techniques, including full azimuth, long offset, anisotropic, elastic data in 3D. There can also be a gravity or electromagnetic response. These synthetic datasets can be paired down to simulate less dense, more routine surveys as well.

'We can see if there's something we don't understand about subsurface, if the algorithms are robust,' Mr Mellen says. 'A magic algorithm that integrates everything does not yet exist.'


The project aims to help develop better multi-disciplinary subsurface workflows, involving geophysicists, geologists, petrophysicists, rock physicists, reservoir engineers, and geomechanical specialists.

'Each of the different disciplines has their traditions, their rules of thumb and their assumptions,' Mr Mellen says.

'As they work more closely in building the model you find they have different assumptions, and sometimes they are inconsistent, these need to be resolved' he says.

Projects so far

SEAM has a history of running major projects to address 'industry grand challenges'. These are projects that are at a scale not easily managed by single companies or institutions.

The first project looked at deepwater subsalt imaging, running from 2007 to 2013, with 24 participating companies, and a $5.4bm budget. It included simulations of seismic, gravity, controlled source electromagnetic (CSEM), and other methods. The results were benchmark datasets for use in new algorithm development, education and training.

The second project, running from 2011 to 2014, looked at land seismic challenges, including unconventional shales, desert terrains and overthrust areas, with 22 participating companies. This project had a $5.2m budget. Again, the results were benchmark datasets to be used in further technology development and training.
The third major project, starting in 2014 and scheduled to complete in 2017 is 'Pressure Prediction and Hazard Avoidance through Improved Seismic Imaging'. With 11 industry participants, this project seeks to scrutinize pore pressure prediction techniques from basin scale to the wellbore. The goal is to develop reliable standardized methods to integrate seismic and other data to understand and avoid subsurface hazards prior to drilling.

A fourth project was created with funding from RPSEA and the US Department of Energy, a Time Lapse Pilot Project was undertaken in 6 months during 2016. As an extension of the Pore Pressure project, a smaller subsurface model extracted for use as a proof of concept in time lapse imaging. This project has completed and the data will be made available to the public in Q1 2017.

The current project, 'Life of Field' which runs from December 2015 to June 2019, includes reservoir simulation and geomechanical modelling in the virtual model, so bringing in a '4D' or time lapse element. Most importantly it brings multi-disciplinary and multi-company collaboration to the forefront in a manner uninhibited by proprietary data.

33 companies have participated in SEAM projects to date, including Anadarko, BGP, BHP Billiton, BP, CGG, Chevron, ConocoPhillips, COSL, Devon, EMGS, ENI, ExxonMobil, Geotrace, Global Geophysical Services, Hess, Ikon Science, ION, Landmark, Maersk Oil, Marathon Oil, Murphy Oil, Nexen, Oxy, PGS, Petrobras, Repsol, RSI, Saudi Aramco, Schlumberger, Sigma 3, Shell, Sinopec, Statoil, Talisman, Total, Tullow and WesternGeco.

It aims to bring together people from oil companies, service companies and academia.

Companies can join the project by paying a membership fee and putting forward their own technical expertise to join the project teams. These member participants make all of the technical decisions and select which geologic and production scenarios will be modelled.

SEAM is currently looking for new members for life of field project.
They can be from exploration and production companies, or service companies.

Getting to time lapse

The reservoir simulation study begins with a static virtual model of a complex piece of subsurface, and a complete reservoir description.

The static earth model will be representative of a number of 'high impact' reservoir types selected by the membership.

It will include stratigraphic, rock and fluid property variations at and below seismic resolution. Pseudo-wells and all relevant well measurements will also be generated.

Then the reservoir dynamics will be simulated, to see what it might look like at some point in the future, and what the various well and seismic data might look like.

As the reservoir in the virtual model is produced, the rock above and below will change in pressure (stress and strain changes) in different ways, which can be simulated, and which will also make a change in the simulated seismic recording.

Part of the study work will try to work out the optimum time to do repeat seismic surveys, which could be anything from months to years. Repeat seismic surveys are expensive. But the cost is justifiable if it will show how the reservoir is depleting in ways which can help identify a better pathway to maintaining production, for example enhanced oil recovery or drilling a new well.

The project may also model Enhanced Oil Recovery processes and/or CO2 sequestration. The participating members will decide.

The Future

The 'Life of Field' project is the latest in the grand challenges attacked by SEAM. In today's low price / high volatility oil industry, every effort needs to be made to insure that capital invested in understanding and developing oil and gas fields has a maximum impact on business decisions. 4D Seismic and field development decisions are not cheap, understanding the uncertainty in our methods and improving efficiencies and impact must be a central focus.

SEAM is currently investigating the next project(s) it will propose to the industry. A leading candidate topic is Unconventional Field Development, again incorporating multi-disciplinary integration including seismic, geomechanical modelling, and reservoir flow simulation but extending into microseismic data analysis, fracture simulation and induced seismicity hazards.

comments powered by Disqus


To attend our free events, receive our newsletter, and receive the free colour Digital Energy Journal.


Implementing an Enterprise Data Management strategy in Oil and Gas
Jane McConnell
from Teradata


Latest Edition Nov Dec 18
Nov 2018

Download latest and back issues


Learn more about supporting Digital Energy Journal