Ros Davidson reports on the seismic breakthrough at BP making waves throughout the Gulf of Mexico
Geophysicists at BP’s Subsurface Technical Centre in Houston have achieved what they say is a “major breakthrough” in advanced seismic imaging. The two-part novel technique has already allowed scientists to identify additional oil resources underneath salt formations better than ever before.
At the recent Offshore Technology Conference (OTC) in Houston, BP’s regional president Richard Morrison said that the new sharper images would allow the UK-headquartered company to identify as much as 1 billion additional barrels of new “possible resources” beneath the belt of salt domes which underlie the Gulf. Morrison said the technique would be applied to four of the supermajor’s deepwater fields: Atlantis, Mad Dog, Thunder Horse and Na Kika.
A few days earlier, BP formally announced that the breakthrough had led to the discovery of an extra 200 million barrels of additional resources in the eastern area of Atlantis, a field originally discovered in 1998. The BP-operated field lies in the Green Canyon area, at a water depth of more than 2,100m, and is the company’s deepest moored floating platform in the Gulf.
Wood Mackenzie had previously estimated Atlantis’ recoverable reserves as 744 million barrels, suggesting that better seismic imaging may have a profound effect on the company’s reserves, and its production. New resources discovered at Atlantis alone could be worth about US$2 billion, according to unofficial estimates.
Pillars of salt The breakthrough involves acquiring more data and then using a new and proprietary algorithm to process it much faster than before. “This degree of success has not been accomplished before in the industry, imaging a field below a complex salt formation with an automatically derived subsurface model,” John Etgen, BP’s distinguished advisor for seismic imaging, told InnovOil. The enhanced imaging is also applicable in other situations where the seismic image of a reservoir is obscured, for example by volcanic formations, or in shallow gas-bearing fields.
Although seismic imaging remains one of the industry’s most standard tools, formations such as salt domes routinely complicate the imaging process by distorting or obstructing the acoustical waves. A salt dome acts as an acoustic lens, refracting and scattering the sound waves, but the presence of such domes is often associated with hydrocarbon reserves – leaving seismologists and drillers with a challenge to pinpoint these resources accurately. BP’s breakthrough has two components: First, the way in which BP’s geophysicists collected data was improved. They expanded the amount of information collected by approximately doubling the length of the ‘offsets’ – the distance between the location of the receivers on the ocean floor and the seismic source vessel – used during the imaging process.
Typically, the length of offsets that had been used in Atlantis was 8-10 km, but that was upped to 25 km, said Etgen. Longer offsets have also now been used at Thunder Horse, which is estimated to contain about 1 billion barrels in recoverable resources.
Longer offsets had not previously been used so successfully. Before the advent of full waveform inversion (FWI), there was no clear consensus that longer offsets would be sufficiently valuable, given that they raise the cost of seismic acquisition, he explained. “So, in a way, it was just a historical thing that we didn’t know how to get effective use from them, so why bother to pay for them. Now that we know how to use them, they seem to be very important and beneficial,” he said.
FWI builds a model of subsurface properties, such as the speed at which sound waves propagate, and can thus predict what is underneath an obstruction, such as a salt dome, and produce high-quality subsurface images. FWI and the synthesised data it produces – which is matched to data recorded in the field – have been used widely in oil exploration for more than 10 years.
BP had previously invented and was the first company to deploy wide-azimuth towed-streamer (WATS) technology to illuminate better and image below complex structures like salt by capturing data from wider angles; however, it seems that the new method may provide even greater detail.
Altered images To turn the collected data into more accurate images, BP scientists then used a new algorithm developed at the company’s Subsurface Technical Centre. The algorithm was applied to seismic data run at BP’s Centre for High Performance Computing in Houston, one of the largest supercomputers in the world dedicated to commercial research. The supercomputer has a processing power of 6 petaflops – almost 6,000 trillion calculations per second – and takes up about 14,000 square feet (XX sq metres) at the Centre, which was opened in 2013.
The new automated algorithm, in conjunction with other processing algorithms that the supermajor had employed previously – enhanced the FWI process and allowed data that would normally take a year of analysis to be processed in just a few weeks. BP says it was thus able to accelerate its development recommendations for Atlantis considerably.
Etgen could not comment on what exactly makes BP’s new algorithm better – it remains a proprietary technology – but did add that it was developed by a scientist just two years out of graduate school at Stanford University, 32-year-old Xukai Shen. The technology was first used at Atlantis a little more than a year ago, with the first “intriguing” results delivered in early 2016, he added.
Feedback from other members of BP’s leadership has been similarly glowing; in a statement the company’s upstream technology director Ahmed Hashmi commented: “The new technique has produced the best images of this reservoir that we have ever seen.”
The quest for improved understanding of reservoirs is never-ending. Once a field is discovered, geophysicists are always trying to improve their imaging so that drilling can be done with confidence. Etgen concedes that the initial seismic image of Atlantis was poor, largely because of the salt dome, and that the reservoir’s architecture could not be understood properly. Consequently, drilling was always liable to be inaccurate in parts of the field.
“Really, this latest breakthrough is about the efficiency of how we do our business,” explained Etgen. ”We often know the upper boundary [of a field’s size], and the real goal is to improve the recovery factor,” so much so that the technique is to be applied to other areas of operation. BP global upstream chief executive, Bernard Looney, recently noted: “This technological breakthrough has essentially allowed our team to find a new oilfield within our existing Atlantis field. Given the overwhelming success of this project, we are now deploying this technology across BP’s global operations.”
Oil and gas fields are not always suitable for this kind of imaging, for example if there is insufficient field data, but evidently the operator sees the potential for seismic improvements around the world. The technology is now being lined up for fields elsewhere in the Gulf of Mexico. First up is Mad Dog, which has an estimated 5 billion barrels of oil in place (OIP). It will then be used for fields in Azerbaijan, Angola, and Trinidad and Tobago, where salt domes have also obscured images in the past. In addition, BP has leases for fields that contain shallow gas-bearing formations in the Caspian Sea and off Trinidad, which are also prime prospects for the new technology.