Thursday, June 30, 2005
Copyright © Las Vegas Review-Journal

Scientist denies falsifying Yucca Mountain data

Remarks in e-mails just `water cooler talk,' hydrologist tells Congress

"I have never falsified any documents regarding Yucca Mountain or any other project.'' JOSEPH HEVES GOVERNMENT HYDROLOGIST, TESTIFYING BEFORE CONGRESS

WASHINGTON -- A government scientist who wrote e-mails that ignited controversy over possible document falsification at Yucca Mountain told Congress on Wednesday that he did not alter reports or data.

Joseph Hevesi, a hydrologist who wrote computer models and gathered field data on water flow at the possible nuclear waste site, said provocative comments he sent to colleagues about research and quality controls on the site were "water cooler talk" that did not affect science he conducted.

Explaining one message, Hevesi said he was joking. Other e-mails, he said, were "raw emotional responses" that reflected work frustration but not malice.

"I have never falsified any documents regarding Yucca Mountain or any other project," Hevesi told a House subcommittee investigating electronic messages that rocked the repository program when they were revealed in March.

Hevesi's testimony failed to hand Nevada officials a smoking gun they hoped would reveal management lapses and flaws in science that they could use to sink the nuclear waste program.

Rep. Jon Porter, R-Nev., who organized the session as subcommittee chairman, said his panel's work was not done yet. He said some of Hevesi's answers were vague, and he planned for subcommittee staff to question the scientist further.

"The fact that Mr. Hevesi denied falsifying any data does not close the door on the investigation," said Porter, whose staff has interviewed other scientists about the e-mails.

Porter said he would refocus the subcommittee on the Department of Energy, which he alleged has refused, despite repeated requests, to provide him with documents concerning the e-mails and management's responses to them.

John Arthur, Energy Department deputy director for Yucca Mountain, testified Wednesday and told Porter the department did not want to interfere with a separate inspector general investigation.

Arthur could not say whether the department would respond to Porter's latest demand.

"I think all avenues lead to the DOE," Porter said after the hearing. He said he would subpoena DOE documents if the department did not comply in the next two weeks.

Hevesi's appearance and testimony under oath marked a turn in the e-mail controversy that has contributed to delays in the Yucca Mountain project.

Arthur on Wednesday said again that DOE will not file a repository license until questions surrounding the messages are put to rest.

Hevesi, a United States Geological Survey scientist in his mid-40s who is based in Sacramento, Calif., was subpoenaed to appear.

He was questioned for about 90 minutes by Porter and a half-dozen members of the federal work force and agency organization subcommittee. He was accompanied by an attorney, Scott Treadway of Indianapolis.

The witness said he had provided documents to the subcommittee through USGS but had been reluctant to appear because of a separate investigation being conducted by federal inspectors general.

"I was trying to focus on one situation at a time," he said.

Porter led Hevesi on a dissection of a dozen e-mail messages, part of a group written between 1998 and 2000 when the hydrologist was a USGS worker at Yucca Mountain.

Other subcommittee members asked broader, and in some cases friendlier, questions.

"I can only imagine what it is like being here to discuss a few e-mails out of 10 million e-mails and then be told these must be the epicenter of all that is important," Rep. Darrell Issa, R-Calif., told Hevesi.

"Do you think it is fair, considering the seriousness of the storage facility, or do you think, candidly, that we are looking for the proverbial needle in the haystack even if it is the shortest needle you ever saw?" Issa asked.

Hevesi offered explanations of e-mails containing passages in which he wrote of using "fudge factors" and of keeping two sets of files: one for himself and one for quality assurance auditors.

Hevesi said "fudge factor" was jargon referring to a simplified solution that served as a model placeholder until more detailed data could be inserted in its place.

"Scientists use fudge factors in their work all the time," he said.

On the dual files, he said one computer program he worked with could not use header information, and he had one set with identifying headers and one set without headers.

Despite the difference in the header information, the data was identical on both files, he said.

Hevesi said he was frustrated by quality assurance procedures, which he said in some cases were evolving as scientists were conducting their work.

In one message Hevesi wrote, "Wait till they figure out that nothing I've provided them is QA. If they really want the stuff they'll have to pay to do it right."

On Wednesday, he said that meant he had completed an engineering calculation before others decided it needed to be checked for quality assurance.

"It was very poor wording on my part in the e-mail," Hevesi said. He said he "had a reputation for being flippant in my e-mail."

Quizzed by Porter, Hevesi said he had only a vague recollection of deficiency reports concerning his work.

Arthur testified later that a January 2000 audit identified deficiencies with Hevesi's adherence to quality assurance requirements, including problems with software controls and the lack of a scientific notebook.

Although a DOE investigation has concluded that the e-mails do not suggest that science was falsified, Arthur said Hevesi's modeling "cannot be trusted today without reverification or replication of the specific work."