March 17, 2012
Advanced Search
Home / Issue Archive / 2009 / February #2 / New Interpretation Software Helps Avoid Dry Holes

№ 2 (February 2009)

New Interpretation Software Helps Avoid Dry Holes

The most important question any E&P company must answer is – How Can I Avoid a Dry Hole? The answer – focus on risk reduction. Evaluating possible risk before drilling is much more economical than drilling a non-producing well.

By Indy Chakrabarti

Share it!

To maximize the likelihood of success, organizations need to update to the current technology. As a result, interpretation software, based on the now prevailing Windows™ platform, has migrated from merely “easier to use” to “critical for success”.


With this in mind, SMT’s KINGDOM 8.3, built natively on Windows, introduces three new capabilities that help interpreters reduce risk, around the world. These capabilities focus on exploiting unconventional sources more quickly, exploring new fields more cost-effectively and maximizing on-going production. The science needed to conduct this analysis includes AVO Conditioning, Forward Modeling and Neural Network Based Log Section Replacement.

AVO Conditioning

With KINGDOM’s 8.3 AVO Conditioning capability, SMT is the first company to bring advanced AVO processing capability to the generalist interpreter. This enables even geoscientists not specialized in AVO or processing techniques to reduce noise in pre-stack data. This means interpreters can often avoid the time and expense of sending data to a third party to be re-processed because of unclear images. Instead, interpreters can now conduct their own AVO conditioning, which is an important component to detecting sources of unconventional gas.


During the last decade, development of unconventional gas reservoirs has become an area of growing interest and opportunity in Russia. And now, with AVO conditioning built inside SMT’s core interpretation tools, prospecting in unconventional gas fields can be made more economical and completed in much faster time frames. In addition, with SMT’s new office in Moscow, Russian customers will receive the same high level of technical support as SMT’s 2,500 global customers.


Reducing risk in evaluating subsurface reservoirs requires that highly accurate image processing techniques be applied to acquired seismic data. Due to the historical complexity of these processes, they have been performed either within the exclusive domain of 3rd party processors or they have required that processing software be created specifically for those who have specialized experience in processing data. Because of the historical requirement for specialization and the complexity of the processing tools, especially pre-stack processing for AVO conditioning, the general interpreter has had limited, if any, access to such tools. Yet, it is exactly this kind of capability that is most needed to avoid drilling costly dry holes when searching for unconventional reserves.


With KINGDOM 8.3, interpreters now have access to these critical pre-stack processing tools for AVO conditioning and pre-stack analysis, which enable them to do basic processing of data to answer questions such as:
When should I use de-convolution?
When should I use muting and rebinning?
How can Super Gather analysis improve my image?


Forward Geologic Modeling
Another critical need for risk reduction is to conduct ‘what if’ analysis as it relates to well data in new fields. KINGDOM 8.3’s new 1d Forward Modeling (1DFM) provides the ability to conduct scenario-based analysis by changing assumptions on lithology and fluid composition in oil and gas fields.


‘What-if’ analysis provides interpreters the ability to understand the environment and subsurface where there is a lack of “real” data from wells. Such analysis and geologic modeling is particularly critical in the exploration of new oil and gas fields.


With Russian oil production predicted to peak around 2010, new fields and new reserves will be even harder to recover since they are in colder, more remote regions with underdeveloped infrastructure. As a result new interpretation technologies, like forward modeling, can advance exploration prospects and plays to help improve recovery and production levels.


Using existing borehole information, interpreters can evaluate where fluid and lithology compositions may be slightly different. By looking at the ‘what-if’ scenarios, interpreters can determine what the reservoir response patterns may be in nearby sections of the field. The ability to conduct this analysis enables interpreters to more quickly identify reservoirs because they have a model of what the response may look like in similar areas of interest.


To complete the picture, KINGDOM’s 8.3 1DFM provides scenario-based analysis of both pre- and post-stack data, further reducing the likelihood of drilling dry holes.


Neural Network Based Log Section Replacement
Another major requirement in reducing risk when exploring the subsurface is understanding downhole information. Well logs are the key source of this information. However, in many cases, aging well log data can have sections or even entire types of well logs missing, which makes it difficult for interpreters to fully evaluate a particular part of the subsurface.


Neural network technology can “fill in” missing information with high levels of accuracy. Using KINGDOM’s Neural Log Replacement capability, log sections can be reconstructed, automatically, by network based processing that uses data from similar wells, related log curves or different sections of the same log. In fact, in internal benchmarking this predictive technology has shown to achieve an average of 88 percent accuracy.


This new technology analyzes all known information about a specific part of the subsurface from the rest of the log and other similar logs and reconstructs the lost data.


What sets this technology apart from past solutions is that there is no need to exhaustively “train” the simulator before it can begin to produce valuable output. SMT’s solution is essentially self-training. The neural net looks at existing data, and after culling it down to the relevant information, attempts to generate the best prediction with minimal intervention from the geoscientist – making it faster and easier to use.


The benefit of using this technology is to help maximize oil and gas production in both older and declining fields where outdated logs with missing information often exist. Being able to use interpretation tools to improve the odds of finding remaining pay allows exploration and production companies to squeeze more out of aging fields.


E&P companies understand that avoiding dry holes is the key to success in the coming year. Fortunately, new tools are now available that can significantly reduce the risk of this outcome. The advanced technologies discussed here integrate geophysical and geological data analysis to help interpreters reduce structural and stratigraphic risk – ultimately increasing exploration and development success.

Share it!
Copyright © 2008 Eurasia Press, Inc. (USA). All rights reserved.
Web programming by Iflexion
Copyright © 2008 Eurasia Press (www.eurasiapress.com)