A project aims to optimize seismic surveys of seabed for oil prospection

An initiative of the Research Center for Gas Innovation will use new software technologies to shorten the lag between data capture and the availability of helpful information on offshore oil and gas exploration.

By André Julião  |  Agência FAPESP – A group of 65 researchers and developers will invest the next several years in the solution to a well-known problem in the oil and gas industry, which is how to profile the seabed more quickly in areas where there may be reservoirs and significant drilling potential.

The new project is under way at the Research Center for Gas Innovation (RCGI), a FAPESP-Shell joint venture under FAPESP’s program of Engineering Research Centers (ERCs). The RCGI is hosted by the University of São Paulo’s Engineering School (POLI-USP) in Brazil.

The focus of the project is the creation of open-source software that will shorten the lag between field data capture and the conversion of data into information for use by geologists and geophysicists in offshore oil and gas exploration. Currently, this lag can be as long as two years.

The project is the 46th begun since RCGI began operating two and a half years ago. It was announced at the opening of the Third Conference on Sustainable Gas Research and Innovation , on the theme “Transforming gas to transform the future”, held on September 26-27, 2018, at USP’s International Diffusion Center.

“This kind of analysis is necessary in any survey of oil and gas reserves, or even of carbon storage sites,” said Bruno Souza Carmo, a professor at POLI-USP and coordinator of the project, which is a partnership with Shell and Imperial College London.

In typical offshore seismic surveys, sound waves are bounced off the seafloor by air guns on a ship. Arrays of floating hydrophones towed by the vessel capture the sound waves as they reflect back to the surface, generating an enormous amount of data.

After processing, the data are converted into images, which are used by geophysicists and geologists to pinpoint the most likely location of oil and gas reserves and to estimate the difficulty of drilling in any particular area.

“It costs a lot in computational terms to process these datasets. Our idea is to create totally open-source software that can be used by anyone and can be adapted to advances in computing,” Carmo told Agência FAPESP.

“Computers are getting faster by the day. Hardware performance improves constantly, but rewriting code to get the best from new computers takes time, and one element of the project aims to make software more portable so that it won’t be necessary to rewrite algorithms whenever a computer is updated,” said Aly Brandenburg, Shell’s subsurface coordinator, geoscience program leader and liaison with the RCGI project.

New phase

Júlio Meneghini, Scientific Director of RCGI, explained that the new project will use existing hardware and software as programming tools. “We’ll also use artificial intelligence and machine learning because the sheer volume of data generated by seismic prospecting surveys is huge, and the algorithms we have today aren’t capable of processing it,” Meneghini told Agência FAPESP. “For this reason, we’ll develop tools for use with the new supercomputing architectures that are being consolidated as we speak.”

The presentations delivered to the conference covered RCGI’s other work in progress in addition to the new project. A novelty at this year’s event was the first set of results from the CO₂ abatement program begun in November 2017. The Center also runs programs in engineering, physical chemistry, and energy policies and economics.

The CO₂ abatement program alone comprises 16 projects on topics ranging from carbon capture from the atmosphere and carbon use in synthetic fuel and electricity production to carbon management, transportation and storage and the monitoring of methane leakage from wells into which carbon has been injected, among others.

“We currently have more than 300 people working at RCGI, and we’ll have about 50 other researchers in the new project. The numbers are rising, and scientific production will increase exponentially. Both FAPESP and Shell are demanding results at the knowledge frontier in all 46 projects. And we’re getting them,” Meneghini said.

The CO₂ capture and usage projects respond to demands both to reduce emissions because of global climate change and to create new energy sources.

“It could take decades to complete this energy transition [from fossil fuels to renewables]. If we look at history, we see that we’ve always been in an energy transition. The process is very slow,” said David Torres, Shell’s Vice President for Integrated Gas and CO₂ Abatement Technology.

The oil and gas industry is about more than burning fuel, Torres added. It includes the manufacturing of chemicals, fertilizer, plastic, and other products that do not yet have substitutes. As a result, even though oil production is expected to peak in 2030, Shell has set the horizon for its oil business at 2070, if not beyond.

“Production will start to decline [after 2030], but it will still exist. Oil will remain part of the mix [of energy sources]. So we have to think in terms of very distant horizons,” Torres said.

More information about RCGI: www.rcgi.poli.usp.br.