Michel Boufadel began a recent presentation about Marcellus Shale drilling with a photo of the ruptured Exxon Valdez oil tanker spilling into Alaskan waters, a disaster whose remnants the Temple University engineering professor has been studying for years.

He flipped to a photo of himself and some graduate students standing around a pool of oil in a hole in the sand of an Alaskan beach.

"Everyone assumed in 1992" that the oil from the spill had been properly remediated and was "going to disappear," he said. "Yet it is still there. That is the problem with groundwater pollution. It doesn't go away that fast."

Dr. Boufadel is one of the scientists who study the rocks, water and people directly affected by Marcellus Shale drilling who cautions that everything from the way the rock breaks underground to the way contaminated water travels through an aquifer has not been - or cannot be -thoroughly considered.

Much of the attention about the environmental risks of natural gas drilling in the Marcellus Shale has focused on the potential for hydraulic fracturing to contaminate drinking water aquifers.

According to the industry and both state and federal regulators, there has never been a confirmed case of contamination being caused by the fracturing - a process of injecting millions of gallons of water, sand and chemical additives underground at high pressure to break apart the rock.

The industry takes a narrow view of what such contamination would mean, limiting it to what they say would be an impossible instance of the toxic mixture migrating through the new cracks caused by the fracturing operation, up a mile of rock, and into a drinking water aquifer.

But legislators and federal regulators are increasingly looking at hydraulic fracturing as more than the isolated act of breaking apart the gas-bearing rock; they see it as part of an interconnected series of often hazardous steps, from trucking and storing toxic chemicals on a well site to disposing of the fluid laced with salt, metals and radiation that comes back out of the wells.

In March, the Environmental Protection Agency announced plans for a study of hydraulic fracturing that would consider all of those factors - the whole life-cycle of a well.

Kathryn Klaber, the head of the Marcellus Shale Coalition, a Pennsylvania-based gas drilling cooperative, said the industry supports the EPA study, as long as it does not halt or slow the pace of drilling.

"I don't think you have to stop something in order to study it," she said.

The industry also points to a previous EPA study of hydraulic fracturing that did not find any instances of the practice causing water contamination, but which was limited to only one type of hydraulic fracturing, in coal-bed methane wells.

"What we're missing is that definitive, absolutely unquestionable, science-based, non-politically influenced study," said Dr. Anthony Ingraffea, a Cornell University engineering professor who has two decades of experience working on computer simulation of hydraulic fracturing in oil and gas wells. "And that is what everybody is hoping that the EPA will do."

'What can we live with?'

Many scientists, including Dr. Ingraffea, acknowledge that there are limits to the usefulness of the EPA study, no matter how ambitious the final scope, because it must be completed by 2012, a relatively short amount of time.

"There shouldn't only be one study or awaiting the EPA study," said Dr. Boufadel, who advocates for risk-assessment studies tailored to individual communities near gas drilling. "There should be 10 or 20 studies. That would allow the truth to come out."

He proposes studies that would measure and assign a value to the relative risk of living among a certain number of wells, compressor stations, pipelines, wastewater impoundments and the other infrastructure necessary for drilling and production.

Evaluating risk, he said, is "how insurance companies make decisions. That's how we, as people, make decisions.

"It is not yes or no. It is what can we live with."

Asked if he knew of anyone conducting a study like that he said, "No. Not to my knowledge."

Dr. Boufadel also suggests that several practices that are standard in Pennsylvania for measuring contamination from drilling are questionable.

The weight of any water contaminated with the salty waste produced by Marcellus Shale wells will cause it to sink in an aquifer, he said, below the reach of conventional measuring tools, like groundwater monitoring wells.

"We really need more advanced models than we are talking about now," he said, or the state will risk misjudging the scope of a contamination incident until a "whole aquifer is polluted."

'Nobody knows; no one can know'

Conrad Dan Volz, director of the Center for Healthy Environments and Communities at the University of Pittsburgh, also argues that science has been missing in much of the consideration of long-term or cumulative effects of shale gas exploration.

He lists a number of elements of the drilling process that require further study, and plans to begin researching some of them this summer in southwestern Pennsylvania. His work will include baseline testing of rivers and comparisons of drinking water wells in areas full or free of gas drilling.

"The question is, why didn't we do the science beforehand on this?" he said.

"What we're really bad at - and we have the tools to do this - is anticipate problems. And I don't see where anyone has done much anticipatory work."

Even the most straightforward assurance about the hydraulic fracturing process - that aquifers are protected from fracturing by thousands of feet of layered, solid rock - is not as certain as the industry insists, Dr. Ingraffea, of Cornell, said.

Although he does agree that the chance of contamination through those layers is minuscule, he also knows from experience that the work to predict and measure where fractures go is necessarily inexact, and the rock "unfortunately" is not solid or impermeable.

To say that hydraulic fracturing contamination through direct communication with an aquifer is impossible is "nonsense," he said. "To say that it is inevitable is nonsense.

"We're dealing with a highly probabilistic underground system, where nobody knows, no one can ever know, exactly the geology that's down there, exactly the geometry of what's down there."

Add the very remote risk of fractures causing direct contamination, to the larger risks of well casing failures and human errors on the surface and the total probability of failure during Marcellus Shale gas production "starts looking, to me, high," he said. "Very risky."

Gas drilling companies have financial incentives to avoid mistakes, he said, but the experience of Marcellus Shale exploration so far - what he calls "ground truth" - has been a series of mistakes followed belatedly by attempts at improvement.

"They could have done this totally differently if they weren't in a hurry," he said.

Contact the writer: llegere@timesshamrock.com