Digital Energy Journal- Ikon Science’s Ji-Fi now used by “leading US independents and supermajors”

Join | Existing Member Login

You are Home » News » View Article

Ikon Science’s Ji-Fi now used by “leading US independents and supermajors”

Tuesday, July 26, 2016

Ikon Science's 'Ji-Fi' or 'Joint Impedance and Facies Inversion' software is now used by 'leading mid-size active independents and supermajors in the US, and have provided valuable results on two of the largest oilfields in the North Sea' just one year after commercial launch

Ikon Science's 'Ji-Fi', or 'Joint Impedance and Facies Inversion' software, which aims to provide a much clearer understanding of the subsurface, is now used by "leading mid-size active independents in US," just one year after the commercial launch of the software.

It is also used on two of the largest oilfields in the North Sea, and one of the world's largest oil discovery in the last 10-12 years.

The software is used to 'invert' seismic data by creating an optimum model of rock properties, or impedances, which can reproduce the observed seismic image. These impedances are related to the speed of sound and the density of the rocks. With increased understanding of these properties, the geoscientist can better detect and predict the presence of high quality reservoir rocks, oil and gas in the subsurface.

'I've been in this business 37 years, and Ji-Fi is as close to the holy grail [of understanding the subsurface from seismic] as I've ever seen,' says Martyn Millwood Hargrave, CEO of Ikon Science.

Ji-Fi is designed to extract geological facies and petrophysical properties from the complex seismic signal and deliver directly a 3D description of the spatial distribution of facies and properties within a reservoir

'The thing that makes me excited is that we're right at the beginning of it. The more data you see and the more results you get, the more enthusiastic and competent you get.'

"It is going to make a big change to the way people explore and produce. This is an innovation the industry will make great use of, over the next 10 years."

"America has taken to it very quickly. They are quick to adopt new technologies.'

'Here [in the UK] we're mainly using it on service projects on producing fields and helping to de-risk drilling prospects," he says.

Tullow Oil, which financed the development of the system, had exclusive access to use it for a while. Tullow has used it on its TEN (Tweneboa, Enyenra and Ntomme) and Jubilee fields in Ghana.

"Like most people, Tullow tried it out on areas they had good well control and thought they knew what was going on," Mr Millwood Hargrave says. "Now they use it in new areas."

The system has been commercially available for since early 2015. "We've got 10 or 12 customers who weren't using it a year ago, now using it very happily."

The company has been working towards this for 15 years, and been working intently on it for 5-6 years, with some of the company's top technical staff, including company technical director Drs Michel Kemper, described as one of the world's leading experts in rock physics and inversion technology.

Partly as a result of developing the technology, Mr Kemper has been appointed an 'honorary lecturer' this year for the Society of Exploration Geophysicists, and will be talking about it at SEG events.

There was also innovative mathematical and algorithmic methodscontributed by the Commonwealth Scientific and Industrial Research Organisation (CSIRO) in Australia.

The software aims to make the work as uncomplicated as possible. Since Ikon Science is a software company rather than a research company, it takes seriously its role to "make things easier, more effective and more efficient, so things don't break," Mr Millwood Hargrave says.

The software is available with a wide range of different packages, including leasing the software rather than purchasing it, or purchasing a service.

Tullow

Initial financial support for the project was given by UK oil company Tullow Oil.

"Tullow Oil has a strong belief in technology partnerships," says Angus McCoss, exploration director of Tullow Oil, in a video recorded interview on the Ikon Science website.

"We've been working with Ikon Science for many years. Ji-fi is a brilliant idea. We backed the project, invested in the development with them, now they are launching the product."

"This product is clearly a major step forward in seismic inversion, it should really help our industry understanding of complex reservoirs."

What it is

Ji-Fi is a complex piece of software - but one way to explain it is to say, it is pre-programmed with algorithmic models of impedance and seismic reflection characteristics of general number of rock types.

For example, if you are looking for carbonates, you can do the seismic inversion on the basis that the rock actually is carbonates. If you end up with a good model, then it probably was carbonates. If you don't, then you know it wasn't carbonates, and you know that the rock doesn't hold what you are looking for.

"When you are looking for rocks there's only certain types of rocks that exist," Mr Millwood Hargrave says. "Wherever you drill in the world you're going to find the same rock types, there's only five or ten."

"In petroleum geology you're looking for certain types of reservoirs, carbonate reservoirs, clastics, sandstone reservoirs- or unconventional reservoirs - a mixture of sand or shale, that's kind of it. There are subtle gradations between them all."

"If the program finds something that 'smells' like a carbonate reservoir and has rock properties of a carbonate reservoir - it probably is a carbonate reservoir."

'And if you know what you're looking for, you can reduce the complexity of the inversion problem."

"It is basically a way of calibrating seismic data using rock physics, and a very sophisticated mathematical model," he says.

The results of the seismic inversion are detailed information about the rock structure, and perhaps information about fluids.

You can get a rough starting idea about what rock you are likely to encounter from regional information, outcrops or nearby well logs. The well logs don't necessarily need to be close by - they can be from wells hundreds of km away.

Another way to explain Ji-Fi is to say that you start off with separate 'low frequency' or simplified models of the subsurface, covering any rock you expect to find, such as shale, water bearing sand and gas bearing sand.

This is much more powerful than developing a single low frequency model covering all phases, as interpreters were previously trying to do.

Previously, rock physics was usually being used as a separate process to the seismic interpretation.

'It is a combined problem," Mr Milton Hargrave says. 'You've got a problem for rock physics, and problem of doing the inversion of seismic data. We focus on both of those problems (at the same time).

Seismic inversion

Other than Ji-Fi, there are two main techniques for 'inverting' seismic data to get a model of the subsurface.

The first is 'model based simultaneous inversion', where you build a model of how you think the subsurface actually looks like, and then gradually improve it.

The second is geostatistical inversion, which uses statistical techniques and advanced mathematics to work out what the subsurface most likely looks like.

The problem with both techniques is that you usually need to have a lot of wells in the region under study.

Model based simultaneous inversion is a "fast technology which is good - but doesn't give you a reliable result unless you've got a lot of wells to constrain the inversion," says Michel Kemper, director of technology with Ikon Science.

If you have a well in the region, you can directly measure rock impedances with a well log tool, rather than trying to guess or estimate it from seismic data.

Geostatistical inversion "also needs a lot of wells, typically, it is very difficult to set up and only very skilled people can run it," he says. "It can take weeks or months."

"We felt some innovation was needed in this area."

Benefits

With Ji-Fi, less subjective interpreting is involved, and less assumptions need to be made during the process. You are not working as though the rock is simpler than it actually is.

The 'geological rules' can be acknowledged during the process, so you won't end up with something which makes no geologic sense (such as if you have water bearing sand above gas bearing sand in your model.

All of it leads to a much clearer understanding of the subsurface. It is particularly helpful when there are thin reservoir beds, small reservoir pools, or complex geometry. It may also show up reservoir formations that you didn't know about.

The process does not require having wells in the region under study. If you do have wells, you can use the well data at the end, to see if they give a value for rock impedances similarly to your study.

All of this helps make better estimates of oil in place and work out the best place to drill.

It fits an exploration pattern around the world where operators are moving from focussing on exploration to focussing on working out how they can get more oil out of their existing reservoirs at the lowest possible capital costs.

The process enables companies to integrate previously known regional and local geological models and analogues (places with similar geology) as part of the inversion process.

The end result is a high resolution, quantitative prediction of rock and fluid type and properties.

It could also be explained as 'bridging the gap between geology and geophysics".

Because Ji-Fi is not completely dependent on wells, it can be used in frontier exploration (where no wells have been drilled before).

You can also develop a number of different facies models and test them. For example you can do an 'amplitude vs offset' analysis and see if your actual AVO data matches what your facies model would give if it matched reality.

Ji-fi also creates a framework which could be used to add in other types of data, including wide azimuth seismic, 2D seismic, time lapse and PRM monitoring data electromagnetics. Ikon Science is working on ways that these types of data can be combined.

Case studies

The system was used over in one North Sea project, starting with trends derived in a regional North Sea study, with no direct well calibration involved.

The results of the project matched the facies and rock properties where they were known from well logs.

Another test was a giant North Sea Palaeocene oil field, where Ji-Fi came up with a result which was similar to the conventional methods (simultaneous inversion and geostatistics), but with a 'fraction of the time, and involving only light work on a handful of wells', Ikon Science says. The work could have used regional information instead of well data and ended up with just as good results.

It was used by one oil company in Ghana, and revealed significant additional oil volumes which have acceptable porosity and permeability values, and which showed fluid characteristics which were similar to those seen in the main oil-bearing reservoirs.

Ji-fi was used by Woodside Petroleum off the Northwest shelf of Australia, to help predict fluids and lithology near to existing wells, with work needed within 3 weeks.

The project was run on seismic data for Australia which is often available for open use. The oil company wanted facies models consistent with the wells and the geology, and investigate potential in open acreage.

Associated Companies
» Ikon Science

CREATE A FREE MEMBERSHIP

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