Flying over the Gifford Pinchot National Forest, you might find yourself wondering if some really REALLY big sasquatch had learned to shoot dice.
If you look to the southeast of the eastern tip of the Swift Reservoir on an aerial map you'll see what looks like a pass line win in a craps game, with a roll of a 5 and 2.
Ken Hattan, a Vancouver motorcyclist, noticed the patterns while riding in the mountains several months ago and called The Columbian to ask about them.
"It's strange," Hattan said. "It made me wonder where they came from."
He's not alone in his curiosity.
Over the years, there's been much speculation on local hiking websites about the patterns, which actually appear in Washington and Oregon in spots from the Olympic Mountains to Crater Lake.
"I'm gonna roll the dice and say that's not random," a user named "arundodonax" wrote, tongue in cheek, on a message board recently at portlandhikers.org.
"The clear cut patches are plainly obvious from aircraft so is this just some kind of logging humor?" a user with the tag "daffish" asked on nwhikers.net.
"I couldn't figure out why anyone would log that way," user "vibramhead" responded on that site.
"I'm thinking it was an experiment in forest crop circles," user "treeswarper" said on nwhikers.net.
The strange patterns certainly raise the question: Do bears play craps in the woods?
The answer to that question, for better or worse, is far less silly.
"That was a study that started in 1992, with the trees cut in 1996 and 1997," said Jon Nakae, a forester at Gifford Pinchot. "The intent was to evaluate ecological aspects of green tree retention -- doing clear cuts but leaving different densities of trees to see the effects on trees, birds, insects, mosses, lichens and other parts of the ecology."
The study looks at forest recovery in clear cuts where different-sized groups of trees were left alone.
And the fact that the patterns look like dice? That's an unfortunate side effect, Nakae said.
"The square design wasn't really the intent," Nakae said. "That pattern had more to do with the rigidity that the scientists wanted for their data. It was more to make things uniform and standard."
Each square is 32 acres, so researchers can compare different types of cuts in a uniform space.
In the dice pattern that looks like a 2, 15 percent of the trees were left alone. In the pattern that looks like a 5, 40 percent of the trees were left. In another pattern that looks like you flipped the colors and rolled a 3, 75 percent of the trees were left. And each part of the study also has a full square where 100 percent of the trees remain.
Researchers from the Pacific Northwest Research Station, the Forest Service, the University of Washington, Oregon State and the University of Oregon all worked or are still working on the project.
The four patterns, along with some others that don't look like dice, were replicated in six spots across the two states.
What researchers have learned so far is that leaving tree patches like the 5 dice -- in five dense 2.5-acre circles with 40 percent retention -- seems to be much more effective at restoring the ecology over time than leaving fewer circles, spaced like that of the 2 dice.
"It's a long-term study," Nakae said. "It's designed to go for 100 years or however long it takes (for the forest to fill back in)."
Every year, the forest comes back a bit more in all of the cuts, he added.
Nakae has been involved in the project from the beginning. He helped pick the six spots for the study, which goes under the acronym DEMO, for "Demonstration of Ecosystem Management Options."
The idea of having six areas was to also look at the role of climate and place in forest restoration, he said.
But he added that he's not exactly thrilled by the attention that the patterns seem to get from folks who aren't familiar with the science.
"People think it's some kind of big crop circle," he said stoically.
Even other folks involved with the study can't deny the similarity.
In the 2007 "Science findings" report for the Pacific Northwest Research Station, researchers Charley Peterson and Keith Aubry admitted the patterns could be seen as odd.
"Looking down on the DEMO harvest design, you'd be forgiven for thinking that managers were playing dice with the forest," they wrote.