April Strzelczyk cradled the mallard in her arms and clipped a metal identification band on the duck's orange leg, while fellow wildlife technician Patrick Hagen took blood and other samples from the bird.

After the exam near a stream in Brooklyn Center, Strzelczyk released her grip, and the greenhead flew off.

The duck was among 432 that were captured and tested this winter in the Twin Cities in hopes of unraveling a mystery: how the highly pathogenic avian flu came to devastate Minnesota's domestic turkey operations last year, wiping out 5 million birds.

Wild waterfowl, known to be carriers of flu, were suspected as the source of the outbreak. Ducks and geese now are migrating back to the state, which means poultry producers will keep wary eyes on their flocks.

But despite extensive testing of waterfowl since the epidemic, the mystery remains — and has perhaps deepened. After sampling more than 6,000 wild waterfowl in Minnesota since last spring, officials have found none infected with the H5N2 bird flu, the strain that hammered the poultry industry here.

"It has everyone scratching their heads," said Tom Cooper, U.S. Fish and Wildlife Service regional migratory bird chief.

The cause of the outbreak may never be known, said Chris Jennelle, a research scientist with the Minnesota Department of Natural Resources' wildlife health program.

"At this point, it's hard to nail down a smoking gun," he said. "We know waterfowl are reservoirs for avian influenza."

But officials tested enough waterfowl to give them a 95 percent chance of detecting the highly pathogenic virus in less than 1 percent of the waterfowl population. The result: None was found.

"It's very possible that waterfowl didn't play as large of a role as people think they did," Jennelle said.

Nationally, more than 47,000 wild birds have been tested since 2014, and just 100, nearly all waterfowl, were found with the highly pathogenic strain of virus.

Two Minnesota birds have been found with the virus, a Cooper's hawk and a black-capped chickadee, birds not normally associated with bird flu. That news deepened the puzzle.

Jennelle said researchers might expand the search. Some have suggested looking at seagulls. "They have been found to hang out on farm fields, and actually on turkey farms," Jennelle said.

"The problem is, how do you catch them [for testing]?" he said.

Songbirds, rodents, skunks and other mammals could be possibilities, too.

Testing metro ducks

After thousands of birds were sampled last year — including wild turkeys and ducks shot by hunters — officials decided to continue surveillance over the winter. Hagen and Strzelczyk work for the University of Minnesota and began collecting and testing Twin Cities-area ducks in January.

The goal was to sample 500 ducks; but the early arrival of spring weather meant the lakes thawed and trapping ended last week.

Dr. Pat Redig, a professor of bird medicine and surgery at the university and co-founder of the university's Raptor Center, is overseeing the two-year study. He is among those who are puzzled by the epidemic.

"We cannot explain how it happened," he said.

The likelihood is that humans somehow picked up the virus on boots or equipment and spread it into barns full of thousands of vulnerable turkeys, he said.

"There was a breach in bio-security somewhere," Redig said. The poultry industry has since ramped up its security measures.

Redig said he proposed testing ducks that winter in the Twin Cities on waters that generally don't freeze to learn if some of those ducks were infected.

"I knew this population of resident ducks in the Twin Cities area was a constant factor year after year," he said. "And it's not an unreasonable assumption that these ducks are representative of other ducks in the state. We don't have to drive great distances to test them; they are right here."

"You also get this huge concentration of ducks in the Gulf of Mexico, and you get a mixing of ducks from all the flyways, where disease could be spread. Then they come back up here. We wanted to see if there's anything in returning spring ducks."

The migration already has begun, and some of the recently captured ducks are migrants.

While tests on two-thirds of the 432 ducks haven't been completed, those that have been done haven't turned up the H5N2 virus.

"There are 144 different types of influenza virus, and most aren't highly pathogenic," Redig said.

The H5N2 virus generally doesn't kill waterfowl, which have evolved over eons with various strains of viruses. That means there are no piles of dead waterfowl to be found.

"The virus just doesn't leave any footprints out there," Redig said.

But domestic turkeys are sitting ducks for the virus.

Next year, Redig is hoping to expand the winter duck testing beyond the Twin Cities.

Despite the mystery, he is optimistic that last year's poultry devastation won't be repeated this year.

"I think the industry has biosecurity at a much higher level than it was, so if we do have an occurrence, it will be isolated," Redig said.

Doug Smith is a retired Star Tribune outdoors writer. Reach him at doug.smith23@charter.net.