Technology: 'Disruptive' Innovation in the Subsurface...

The last big innovation or disruptive breakthrough in subsurface interpretation was almost 20 years ago when Geoquest and Landmark 'invented' the 3D seismic interpretation workstation. Since then, improvements have been incremental and the systems have fallen increasingly under the control of IT and supply chain managers who like low risk, competitively priced contracts with major suppliers.

In the meantime, the amount and diversity of data available to sub-surface specialists has outpaced the ability of their systems and workflow processes to manage, integrate and interpret it. For example in:

Exploration

1. Many basins are now assessed, and nearly all wells are nowadays drilled, on the basis of a 3D seismic survey which in its basic form allows sophisticated stratigraphic and structural interpretations, tied to available well logs.
2. Beyond this, lithology and fluid prediction depends on a wide range of derivative seismic attributes - near- and far-stacks, AVO gradients, elastic impedance, phase etc etc.
3. Non-seismic geophysics has its place, especially when such data can be understood and interpreted (integrated with) the geological framework provided by a 3D seismic interpretation i.e. the output of 1. and 2. above. Electro-magnetic surveys and Full Tensor Gravity Gradiometry are especially significant in this respect.

Reservoir Management

1. Most fields will now have a static reservoir description based on all available well data (logs, cores) and a sophisticated interpretation of a 3D seismic survey (deliverables similar to 1. and 2. under Exploration above).
2. Reservoir dynamics are observed by wide-ranging production monitoring, from continuous downhole PLTs to regular 4D towed-streamer seismic to 'as-often-as-you-want' permanent seismic monitoring (either on the surface or downhole).
3. These static and dynamic facts and interpretations have to be integrated into, and with, the reservoir engineers' simulators, sometimes 2D but most often (massively) 3D.

I have left out of these lists several things, for example passive seismic (which some would say offers, via attenuation measurements, information on reservoir permeability) and micro-seismic (the observation of seismic events associated with 'fraccing'), also newer downhole methodologies such as cross-well seismic, micro-gravity etc.

The result is that a team of subsurface specialists - whether working on a basin, a prospect a discovery or a field - can in principle access very large amounts of different types of data, each requiring its own conditioning and analysis, before attempting to integrate these many strands into a coherent interpretation, certainly in three dimensions and possibly four.

I say 'in principle' and 'attempting' because in reality - as stated earlier - the amount and diversity of data available to sub-surface specialists has outpaced the ability of their systems and workflow processes to manage, integrate and interpret it.

Can we see any solutions on the horizon?

First of all, we can recognise that there are new computer systems which can handle much bigger amounts of data - see this article from the Economist.

Second, and at the other end of the analysis process, visualisation is a good way to finally look at a lot of data with many attributes.

So far so good!

But there is a lot missing in between!

The analysis and interpretation of any one sort of data tends to depend either on a lowest-common-denominator seismic interpretation workstation or on specialist software operated by specialists. For example, a Major oil & gas company will have several hundred folk who can interpret a 3D seismic survey on a workstation (1. above), no more than a handful who can undertake the analysis to deliver any desired attribute (2. above under Exploration). I can count on the fingers of one hand the geophysicists I know who could undertake 1., 2. and 3. above.

Of course, companies surmount this problem by having a team of petrotechs working on any one project. However, any one project may be unique in terms of the data available and the analyses undertaken, meaning that the work processes and software used will be chosen a la carte and - pursuing the restaurant analogue - with every dish cooked by a different chef to a secret recipe!

Clearly, the adoption of global data standards has a lot to contribute here, avoiding what an ex-BP (SOHIO Petroleum) colleague referred to as 'the Department for Re-Formatting', and applications that can truly enable and integrate diverse analyses would seem to have a rich future (but are there any out there?).