Seemingly overnight, lead pollution has become one of Utah's chief environmental headaches.

Lead contamination in yards near Midvale's Sharon Steel is second only to what seems certain to be dangerously high levels discovered in the bed of Bingham Creek in West Jordan and South Jordan.As the concentrations in the dry creek bed - up to 30,500 parts per million - were only confirmed a little more than a week ago, it's too soon to try and analyze the hazard. It is probably a threat to health.

But in Midvale, soil samples taken late in 1989 turned up one reading of lead at 4,000 parts per million in a vacant lot, with one home's lead level reaching 2,500 ppm - far beyond the EPA's "action level" of 500 ppm, above which it will take action to clean up the contamination.

These readings have been known long enough for environmental physicists to have time to analyze them.

Lead-impregnated dust blew into yards, businesses and homes when windstorms lifted the contaminated tailings left behind at the defunct Sharon Steel Mill in Midvale. The Environmental Protection Agency is trying to force companies involved with Sharon Steel to pay for the cleanup.

So far, two companies - UV Industries Inc., and the Liquidating Trust (Sharon Steel) - have agreed to pay millions of dollars in cleanup costs. The third defendant in a government civil suit, Atlantic Richfield Co., is resisting.

During opening arguments of the liability trial against ARCO, which began in U.S. District Court on Tuesday, the Justice Department's W. Benjamin Fisherow said, "The United Sates asks you to prevent a risk of harm to children. We are entitled to the fullest protection that the court can provide."

But would the lead actually cause any harm to Midvale residents?

On that subject, Daniel M. Allred, lawyer for ARCO, drew attention to a disturbing gap between government predictions and reality.

Based on the concentrations of lead in the soil at Midvale from the Sharon Steel mill's tailings, federal experts estimated that children there might have blood lead levels up to 30 micrograms per deciliter of blood, he said. The upper range of the federal estimates would tend toward lead poisoning.

The federal government could have checked Midvale residents' health by having physicians draw blood samples. But it didn't.

Instead, defendants in the suit hired Dr. Robert Bornschein of the University of Cincinnati to study blood-lead levels. The results were startling.

"We're pleased at the levels we've seen," Bornschein said at a press conference in January.

Rather than the threatening 30 micrograms, the study found an average of 5.2 micrograms, with the high range close to 10.

By contrast, a control group in a national study - kids not exposed to any unusual lead pollution - had an average of 14.7 micrograms. Salt Lake City residents have an average of 16.7.

In fact, Allred said, the level in Midvale is lower than detected by other tests elsewhere in the United States. He told this paper that he wishes Cincinnati's children had levels that low.

Lead is a pervasive danger. With the phasing-out of leaded gas, blood-lead levels throughout the country are dropping. But it's still everywhere. Throughout the world, centuries of industrial pollution dumped an incalculable amount of the mineral.

In London, lead concentrations in the dust on the sidewalk is about as high as the level of lead in the Sharon Steel tailings themselves, Allred said.

According to his argument, the government's prediction of lead levels in the bloodstreams of local residents should be a way to calibrate its modeling.

Does the government's claims of risk make sense, he wonders, "when the blood levels are so low?"

That's a chilling question, one that brews up a lot of other dark doubts.

Just how reliable are the government's risk assessments? Given the amount of trust we must place in its approval of everything from drugs to pesticides to automobile safety features, it's a real worry.