TRAIL — If and when pipeline crews set up on Bob Barker's property and begin drilling a 36-inch tunnel deep beneath the Rogue River near Trail Creek, Barker figures one of three things will happen and two of them are bad.

At best, the drill could systematically bore through the breccia bedrock and come out the other side in a successful crossing of the Rogue as part of a 232-mile pipeline project proposed to send natural gas from Malin to Coos Bay for export.

At worst, the drilling could cause the breccia to fracture, spewing a clay slurry at high pressure into the heart of the upper Rogue's last remaining wild spring chinook salmon spawning grounds two miles upstream from Shady Cove.

Somewhere in the middle is the possibility that the drilling fails to create a clean tunnel for the pipeline. If so, the draft environmental impact statement outlining this phase of the $1.74 billion pipeline construction proposal offers no backup plan for crossing the Rogue.

"It's a big what-if," Barker says. "I think they need to address the alternative if the drilling is not successful. They're just ignoring the issue."

The what-if game is at the core of the debate on whether Jordan Cove and Pacific Connector Gas Pipeline owners are right when they assert that this common method for running utility lines beneath major waterways will work here and whether the company is properly prepared should it not.

An engineering study done for the project concludes that horizontal directional drilling under the Rogue is "likely feasible" and should go off without a hitch as it has countless times under the Columbia River and as recently as 2007, when Avista drilled under the Rogue for a gas line at Shady Cove and again in Grants Pass a year later.

"This is the environmentally sensitive method for going under rivers," project spokesman Michael Hinrichs says. "Avista's been under there. We've done the tests. We believe we can go under there, too."

But it doesn't always work.

Similar drilling under the North Fork of Coos County's Coquille River in 2003 led to several "frac-outs," the term used to describe when fractures in the rock release a natural lubricant called bentonite into the surface water. On the Coquille, the bentonite coated spawning gravels and led to water-quality violations and fines.

"That's the concern, if you have one of these frac-outs," says Pete Samarin, an Oregon Department of Fish and Wildlife biologist studying the project. "It's impossible to tell what might happen. So, from that standpoint, we're preparing for the worst."

Samarin says his agency already has proposed that the pipeline either be elevated over the Rogue near Trail Creek to skip entering the riverbed altogether or shift the Rogue crossing upstream of Lost Creek Lake and away from wild salmon habitat.

Both of those have been turned down and only the underground option is addressed in the Federal Energy Regulatory Commission's study on the potential impacts of the project.

So the ODFW likely will request a "substantial" performance bond be taken out for this leg of the pipeline, although no specific amount has been discussed, Samarin says.

"That way, if it were to fail, there would be money immediately available to mitigate impacts to spring chinook spawning habitat," Samarin says.

In studies cited in FERC's draft environmental impact statement, any "inadvertent returns" of drilling fluid into the water likely would dissipate quickly because of river current. Any significantly damaged habitat would be removed and new spawning gravels could be added to the river, one study states.

If a frac-out did occur, salmon and steelhead in the Rogue likely would avoid the sediment plume as it moved downstream, then move back into their regular haunts when it dissipated, according to the study.

If cleanup measures were to cause more environmental damage, the bentonite might be left to get washed away naturally during later high flows, the study states.

"You plan for the worst and design for the best," Hinrichs says.

The actual design starts on Barker's property about 1,500 feet from the river's edge. The drill would work down at an 8 percent grade, reaching a maximum depth of 73 feet below the Rogue, with a 53-foot buffer of bedrock between the drill and the river's gravel and sand levels, the study states.

It would then exit at a 5 percent grade at about 3,050 feet from where the drilling began, the study states.

The same form of horizontal drilling is proposed to cross the Klamath and Coos rivers in this project.

Plans are to try up to three times for a hole worthy of holding the 36-inch pipe beneath the Rogue. Right now, there is no defined Plan B.

"It would just be another method for going across the river," Hinrichs says. "I can't really guess what that would be."

Barker guesses it would be two other options that surfaced in past environmental studies for this pipeline crossing — a pipe suspended above the river or a so-called "wet open-cut crossing."

The open-cut crossing entails building a temporary dam to expose half the riverbed for digging and pipeline installation, then rerouting the river to do the same in the second half.

That approach, which is planned for the pipeline's South Umpqua crossing, involves a major disturbance of the river channel and could be worse than elevating the pipe, Barker says.

"Still, you're putting a 36-inch pipe over the top of the river," Barker says. "I don't think people will be very excited about that. It's a liability. It presents safety concerns. It's an eyesore."

Barker wants FERC to require that backup methods for crossing the Rogue be completely vetted before any construction begins because he believes the pressure would be too great to move forward too quickly should the Rogue crossing hold up the entire project.

"It seems to me that issue should be added at the front end, not the back," he says.

Reach Mail Tribune reporter Mark Freeman at 541-776-4470 or by email at mfreeman@mailtribune.com.