Three years ago a very popular snowy owl spent much of that winter in Anoka County, entertaining birders and photographers alike. The bird was surprisingly loyal to its location in the city of Ramsey.

The bird, which researchers named Ramsey, became part of an ongoing study of snowy owl movement, in which owls are fitted with tiny transmitters that track location.

Ramsey, an immature male, was tagged on Jan. 26, 2014. In late winter he moved north, his last known location being in southern Saskatchewan.

The transmitter he wore sent location data via satellite to cellphone towers. The research team collected the information when the owl, in effect, phoned home.

Ramsey was last heard from on April 26 of that year. Cell service is sketchy in Saskatchewan.

It's not unusual for an owl to go silent, says Scott Weidensaul, one of two men leading the owl-tracking project, known as Operation SNOWstorm, based in Pennsylvania.

"We expected that a significant majority of the birds we tagged would not check in subsequently," he said. "All the owls we tagged in 2013-14, our first year, were juveniles, and young birds have a poor survival rate," Weidensaul said in an e-mail.

Transmitters also fail, solar-powered batteries drain dry, birds never come within range of a cellphone tower, and some owls die.

Owls that go quiet do not hinder the research, though.

"What we really want are the winter-grounds data that we get after we tag a bird down here," he said. Where do the birds go when they move south out of breeding territory, and how long do they stay?

Without electronic tracking, no one would know. Most birds stay in touch for a few weeks or months.

A new bird was tagged in North Dakota in January 2016. Her reports lasted four months.

Twenty-two owls were given transmitters that first winter, including Ramsey, the only owl tagged here. The number has grown to 43. Three birds remain in contact.

This winter, snowy owls have appeared here in the usual small numbers, occasional and scattered. One was even seen in downtown Minneapolis in mid-January, perched near the corner of 3rd Street and 10th Avenue S.

The study found more owls than expected, Weidensaul said. There are very good numbers of owls in Canadian prairies, he said. Hunting must be productive. The birds have not moved south.

Voles, the birds' first prey choice, cycle in numbers, feast to famine.

"We don't really know what drives the specifics of vole irruptions, other than a good breeding season the previous summer," Weidensaul said. Irruptions, in the ecology world, are a sudden increase in an animal population.

Last winter, he said, "there were tremendous numbers of voles on Amherst Island in Lake Ontario, for example, and dozens of owls. This year, the rodent cycle is at a low point there, and the owls are largely absent."

Would Weidensaul care to predict vole populations and owl invasions for Minnesota?

"Vole cycles tend to be pretty localized, and usually run in three- to five-year periods," he said.

So, no predictions for a repeat of 2013-14's exceptional owl winter.

(You can find project details at projectsnowstorm.org/maps/ramsey)

Read Jim Williams' birding blog at startribune.com/wingnut.