VANCOUVER — Top bureaucrats in Ottawa have muzzled a leading fisheries scientist whose discovery could help explain why salmon stocks have been crashing off Canada's West Coast, according to documents obtained by Postmedia News.

The documents show the Privy Council Office, which supports the Prime Minister's Office, stopped Kristi Miller from talking about one of the most significant discoveries to come out of a federal fisheries lab in years.

Science, one of the world's top research journals, published Miller's findings in January. The journal considered the work so significant it notified "over 7,400" journalists worldwide about Miller's "Suffering Salmon" study.

Science told Miller to "please feel free to speak with journalists." It advised reporters to contact Diane Lake, a media officer with the federal Department of Fisheries and Oceans in Vancouver, "to set up interviews with Dr. Miller."

Miller heads a $6-million salmon-genetics project at the federal Pacific Biological Station on Vancouver Island.

The documents show major media outlets were soon lining up to speak with Miller, but the Privy Council Office said no to the interviews.

The Privy Council Office also nixed a Fisheries Department news release about Miller's study, saying the release "was not very good, focused on salmon dying and not on the new science aspect," according to documents obtained by Postmedia News under the Access to Information Act.

Miller is still not allowed to speak publicly about her discovery, and the Privy Council Office and Fisheries Department defend the way she has been silenced.

But observers say it is indefensible and more evidence of the way the government is undermining its scientists.

"There is no question in my mind it's muzzling," said Jeffrey Hutchings, a senior fisheries scientist at Halifax's Dalhousie University.

"When the lead author of a paper in Science is not permitted to speak about her work, that is suppression," he said. "There is simply no ifs, ands or buts about that."

The Harper government has tightened the leash on federal scientists, whose work is financed by taxpayers and is often of significant public interest — be it about fish stocks, air pollution or food safety.

In one high-profile case reported by Postmedia News last year, Natural Resources Canada scientist Scott Dallimore had to wait for "pre-clearance" from political staff in the minister's office in Ottawa to speak about a study on a colossal flood that swept across northern Canada at the end of the last ice age.

Researchers, who used to be free to discuss their science, are now required to follow a process that includes "media lines" approved by communications officers, strategists and ministerial staff in Ottawa. They vet media requests, demand reporters' questions in advance and decide when and if researchers can give interviews.

Environment Canada now even has media officers in Ottawa tape-recording the interviews scientists are allowed to give.

Yet transparency as well as open communication and discussion are essential to science, Hutchings said, and Ottawa's excessive control over communication is "really poisoning the science environment within government."

"An iron curtain has been draped over communication of science in the last five to six years," he said.

The Privy Council Office and the Fisheries Department said Miller has not been permitted to discuss her work because of the Cohen Commission, a judicial inquiry created by the prime minister to look into declines of the famed Fraser River sockeye salmon. She is expected to appear before the commission in late August.

The Privy Council Office has "management responsibility" for the commission and decided Miller should not give media interviews about her study because of the ongoing inquiry, said PCO spokesman Raymond Rivet.

"Fisheries and Oceans Canada is conscious of the requirement to ensure that our conduct does not influence, and is not perceived to be attempting to influence, the evidence or course of the inquiry," department spokeswoman Melanie Carkner, said in a written statement.

Hutchings doesn't buy it, saying he finds it "inconceivable that the Cohen Commission would have viewed the communication of brand new scientific information as somehow interfering with its proceedings."

To Hutchings, the muzzling of Miller is "all about control — controlling the message and controlling communication."

The government released 762 pages of documents relating to the Miller study to Postmedia News. Many passages and pages were blacked out before they were released.

The documents give a glimpse of the way media strategists, communication specialists and officials control and script what government scientists say — or, in Miller's case, do not say —about their research.

The documents show the Fisheries Department wanted to publicize Miller's study, which raises the spectre of a mysterious virus killing huge numbers of Fraser River salmon before they reach their spawning grounds.

In November, two months before Miller's findings were published in Science, Fisheries Department communications staff started preparing "media lines."

The lines said Miller's findings "demonstrate unequivocally that salmon are entering the river in a compromised state and that survivorship can be predicted based on gene expression more than 200 kilometres before salmon reach the river."

Miller's team has not yet identified a culprit, but her Science study said one possibility was a virus associated with leukemia, which can be transmitted from fish to fish.

Reporters from Postmedia News, CBC and many other media, including Time Magazine, asked to speak with Miller after receiving the Jan. 9 notice from Science.

The documents show DFO communications staff firing off a series of "URGENT" emails as they tried to get clearance from Ottawa for Miller's "media lines" and the OK for her to speak with reporters.

They eventually got approval from DFO's deputy minister and the federal fisheries minister's office but then had to go "to PCO for sign off," the documents say.

"You need to write a note for hot-button approval," Rhonda Walker-Sisttie, director of DFO public affairs and strategic communications in Ottawa, told the Vancouver communications branch by email, advising them to use the "PCO template for media requests."

As the reporters' deadlines loomed, Terence Davis, DFO's Pacific regional director of communications, implored Ottawa to clear Miller to talk.

"If we are unable to set up a technical briefing or interviews for later today, the opportunity for DFO to gain the profile we would like for Kristi's work may be lost or very much diluted," Davis said in one email.

"We are pushing hard," Walker-Sisttie assured the Vancouver communications office.

Then, weeks after the department learned Miller's findings were to be published in Science and several days after 7,400 journalists were notified about the study, the PCO decided not to let Miller talk about her findings and their significance.

"PCO has decided that we can only respond in writing," Walker-Sisttie reported from Ottawa. Another explained: "Kristi was not approved to provide interviews."

The reporters, who the documents show were baffled and miffed by DFO's inability to get Miller on the phone or on camera for interviews, filed stories based on her highly technical Science report and interviews with some of Miller's colleagues at the University of B.C.

Miller is still not allowed to speak about the Science report, which she wrote in a Nov. 12 memo "reflects only a fraction of what we know."

But Miller will finally be able to discuss her work in late August, when she is scheduled to testify at the Cohen Commission.

Hutchings said government communication strategists are likely now busy telling Miller: "Here is what you can say. Here is what you can't say. Here is what we want you to stick to. Don't talkabout this."

"I'd be amazed if she is not receiving such quote, unquote 'advice,' " said Hutchings.