ACADEMIA

Research and Markets has announced the addition of the "Technology Watch from The Data Room - 2009 Society of Exploration Geophysicists, Houston" report to their offering.

The 2009 Society of Exploration Geophysicists (SEG) Convention was one of the most interesting trade shows we have attended in recent years. The State of Energy R&D session vacillated between reports of R&D successes and complaints of underfunding. Ming (RPSEA) noted that Norway's offshore R&D spend matched that of the US. In per capita spend, Brazil beats the US ten fold.

While funding may be short, imagination is not as witnessed by Gus Berkhout's (Delphi) notion of an autonomous dragonfly and ray fish robotic seismic sensors!

Geophysicists are interested in applying their techniques in novel fields such as carbon capture and storage (CCS) and geothermal exploration. In a special session on CCS, Raymond Lee Orbach (University of Texas at Austin) noted the need for, and high cost of, flue gas capture and storage. Capturing CO2 from power plants and compressing it consumes 2530% of a stations power output and is not a very popular idea amongst electricity generators. Scalability to gigawatt levels is uncertain, while early underground storage tests have led to the phenomenon of the NUMBY, i.e. not under my back yard.

The SEG is the oil and gas trade show for high performance computing (HPC). In several HPC events there was debate as to whether Moores law has really ended. The seismic processor is now confronted by a multitude of increasingly complex architectures. These include multicore CPUs, graphics processing units (GPU) used for number crunching and more esoteric architectures such as FPGA2s. The commonality across the new multi core world is that there is no standard or easy way of porting and optimizing legacy code to the new highly parallel architectures. As Robert Clapp (Standford) put it, the HPC industry is running scared and there is as yet no answer to the multi core development problem. This situation is likely to be exacerbated as plans for exaflop computers with perhaps millions of cores are realized.

On the seismic processing front, recent developments in non coherent acquisition and processing offer an alternative route to high resolution imaging, without the exponential increase in trace count and compute resources implied by the mainstream. A related notion, again from Berkhout, is to see multiples and reflections from the underside of boundaries as all adding to the illumination process and using them, rather than seeking to eliminate them. According to Felix Herrmann (UBC/SLIM), new matrix solving math means that we are on the cusp of a breakthrough in seismic imaging. Aramco also described a new processing technique for statics that removes the need for fiddly model building, deriving everything statics, velocity model and image from the data itself. Mark Thompson (Statoil) traced twenty years of hugely successful ocean bottom seismic recording which has led Statoil to increasingly densely sampled seismic carpets, a digital fiber optic oil field and (more) autonomous seismic nodes connected to the seismic compute cloud.

On the development front we noted use of Nokia's Qt porting tool. This is used to port an existing code base typically from Unix/Linux to Windows. Qt provides a safe porting route without the lockin that a port to .NET would imply.

Finally, while there are geophysical applications and jobs in both CCS and geothermal, these need to be put into context. A cubic meter of rock holding oil or gas contains 40 to 400 times the energy of the same rock holding hot water putting a cap on the cost of applicable geophysical surveys.

Highlights

• State of Energy R&D Special Session
• US underfunding R&D (RPSEA)
• Incoherent acquisition and processing (Delphi)
• Future of HPC (LANL)
• Matrix math breakthrough for seismic imaging (SLIM)
• New era for land seismics (Aramco)
• HPC Special Session
• Geothermal Special Session
• Kelman iGlass data management system
• Geographix Discovery 3D/Windows 7 release
• Direct Hydrocarbon Indicator best practices (Exxon)
• Two decades of ocean bottom recoding (Statoil)

Key Topics Covered:

1 SEG Forum State of Energy R&D

2 SEG Forum - The Road Ahead

3 IBM HPC event

4 Exhibitors

5 Invited Session

6 High Performance Computing (HPC) Special Session

7 Geothermal Special Session

8 Standards - SEGD

9 The Data Room - Technology Watch

For more information visit http://www.researchandmarkets.com/research/d20b48/technology_watch_f