Near-surface focus of new SEG president has Arctic application

Having moved from seismic exploration into other areas of geophysical research, the new president of the Society of Exploration Geophysics still helps to find oil. But the oil John H. Bradford detects now isn't deep in the earth. It's oil released by accident in frozen seas and trapped beneath ice-a possibility for which the oil and gas industry must be prepared as exploration and development expand in the Arctic.

Bradford, a professor in the Department of Geosciences at Boise State University, describes his primary research interest as "quantitative analysis of wave propagation based geophysical data, specifically seismic reflection and ground-penetrating radar (GPR)." Concentrating on earth systems no deeper than 20 m, Bradford by necessity makes greater use of GPR than of seismic methods.

"Pretty much our entire zone of application is the weathering zone," Bradford says, referring to the poorly compacted near-surface layer not conducive to propagation of the sonic energy central to seismic work. A seismic processing step called statics, in fact, essentially processes the weathering zone out of exploration seismic data by setting a reference plane for timing measurements below it. GPR thus provides a supplemental way to study the shallow subsurface.

Widely used to map underground objects such as pipes and cables, GPR has become an important tool of environmental remediation. And the method's ability to sense contaminants in water makes it increasingly useful in tracking oil releases below ice. But acceptance hasn't come easily.

Radar and ice

In the late 1970s and early 1980s, Bradford says, tests of GPR for detecting oil beneath sea ice weren't encouraging. Later, however, the former Minerals Management Service became interested in interpreting GPR signal attributes for the detection of oil beneath ice.

Boise State joined the effort in the early 2000s and became part of a consortium that conducted important research at the US Army's Cold Regions Research and Engineering Laboratory (CRREL) in Hanover, NH. An experiment involving CRREL's first controlled oil spill tested expectations that amplitude anomalies would indicate the presence of oil below ice.

"We were met with a lot of skepticism," Bradford notes. But results went beyond expectations: "We saw anomalies that we thought were outside containment cells." After the experiment, researchers discovered that oil indeed had leaked.

"Our funders were happy, and we were happy," Bradford says.

A reason GPR performance has improved since it was first considered for detecting oil beneath ice is improvement of survey equipment.

"The hardware has now advanced to the point that the fidelity of the signal is consistent and accurate enough to pull out the somewhat subtle attributes required for the oil detection problem," Bradford says.

Experiments at CRREL recently have tested ground-coupled, airborne, and under-ice survey techniques.

"Ideally, you'd like to apply multiple methods to the same problem in order to get to the best answer," Bradford says.

Another active area of research is the use of full-waveform inversion of radar data, which essentially integrates all the attributes and represents what the new SEG president calls "a pretty robust tool in targeted applications."

And "a research target" is adapting GPR methods to natural ice, which is more heterogeneous than the lab-grown ice sheets used in experiments.

Although the primary way to find oil beneath ice remains crews with augurs drilling holes, Bradford notes, "radar is out there as a tool now." Emergency equipment of Alaska Clean Seas, a spill-response consortium of operators on Alaska's North Slope, includes a GPR system.

Seismic analogs

Like many techniques used in the interpretation of GPR data, the attribute analyses tested at CRREL have seismic analogs. In addition to attribute analysis, GPR methods include prestack depth migration, amplitude variation with offset (AVO), and spectral decomposition-all common terms in seismic work.

"We can use many of the same tools and concepts to analyze radar waves and seismic waves," Bradford says.

But GPR isn't seismic. In a seismic survey, the acquired data are arrival times and locations of reflected sound. With GPR, reflections are of high-frequency electromagnetic waves transmitted from a radar antenna and recorded by a separate antenna.

Reflecting interfaces in seismic work are boundaries between rock layers of contrasting acoustic impedance, which depends on density and sound propagation speed. In GPR surveys, reflections are from materials of contrasting dielectric properties, which relate to tendencies to store and transmit electric energy. Reflecting interfaces can be soil horizons, the groundwater surface, boundaries between soil and rock, and objects placed underground by humans.

The high frequencies of radar pulses allow high resolution-the ability to distinguish between subsurface reflectors-but limit the depth of the tool's effectiveness. In CRREL experiments, GPR with a 500 Mhz antenna was able to detect a 1-2 cm oil layer in most scenarios and resolve a 4-5 cm oil layer.

Much of Bradford's GPR research concerns detection of nonaqueous phase liquids (NAPLs), which don't dissolve in water and are characterized as light or dense. Light NAPLs rise in water; dense NAPLS (DNAPLs) sink.

NAPLs can be distinguished in water because of their nonpolar molecular structure. Hydrocarbons, for example, are nonpolar. Water is strongly polar.

Because "radar waves are sensitive to the difference," Bradford says, GPR can detect DNAPLs that sometimes contaminate water, such as chlorinated solvents and degreasing fluids, as well as light NAPLs such as oil.

Slow acceptance

That the environmental clean-up industry has been slow to adopt GPR, Bradford says, is a source of "frustration among people who do this kind of work."

Geophysical analysis can help site wells drilled to find sources of contamination in water wells and wells drilled later for remediation. Bradford suggests a reason is that the background of environmental clean-up professionals tends to be civil engineering.

"They're not familiar with geophysical tools," he says.

Another difference in perspective is important not only to acceptance of the technology but also to availability of research funds: Oil and gas exploration is driven by profit; environmental clean-up is driven by penalties.

Although geophysical results tend to be seen as "soft data," Bradford notes, "You can do a lot of small-scale geophysics for $5,000-15,000."

He describes a clean-up project involving a 7-acre site that had been used for firefighter training at a US Air Force base and contaminated by NAPLs and DNAPLs. The clean-up effort included the drilling of 1,200 boreholes.

"You would think it would be well-characterized," Bradford says.

But the hydrology company performing the work still didn't understand the site and commissioned a GPR assessment, which conducted a pseudo-3D survey over the site and mapped a clay layer accurate to within 1-2 ft. The survey showed the previous clay map, used for the groundwater flow model, was incorrect. The pre-GPR remediation work, which cost "several tens of millions of dollars," Bradford says, missed the clay layer altogether in many locations.

"For fifteen thousand bucks and a couple of weeks we could dramatically improve on the result," he says.

Route 'not common'

Bradford describes his route into GPR and the study of shallow-earth systems as "not common" because he began his career in seismic research.

The holder of BS degrees in physics and engineering physics from the University of Kansas and a PhD in geophysics from Rice University, he worked during 1995-99 as a research scientist at the Houston Advanced Research Center in subjects ranging from spectral decomposition for seismic exploration to utility detection with GPR. He joined the Center for Geophysical Investigation of the Shallow Subsurface at Boise State in 2001 and served as director during 2006-09. His publication topics include hydrocarbon detection, hydrogeophysics, glaciology, and polar ecology.

Notwithstanding his focus on the near-surface, his broad research interests, and his focus on GPR, Bradford stays close to seismic methods.

"The state of the art in exploration seismology pretty much becomes the state of the art in GPR, so it's pretty important to keep those ties," he says.