Squeezing secrets from the trees By Sam Bishop

Above: At UAF’s Fairbanks Experiment Farm in March 2020, Jessie Young-Robertson ’20 stands by a stack of firewood that she and a small research team are monitoring for moisture content. UAF photos by JR Ancheta and Sam Bishop.

Metal bands tightly wrapped around birch trees gave Jessie Young-Robertson a problem one freezing cold April day several years ago when she hiked out to her research site north of Fairbanks.

The bands, used to measure trunk circumference, had slipped down the trees. Without the bands in place, she had no data.

Young-Robertson, a UAF research associate professor, suspected what had happened.

Today, her research has not only confirmed her suspicions but also inspired a project that could help clean up Alaska’s wood stove smoke by identifying the best times to cut firewood.

The trees in the research site off the Steese Highway, it turned out, shrank so dramatically that cold April day because they’d pumped much of the water out of their trunks and back into the ground.

The water loss was so great that the sensor bands, which had previously expanded as the tree swelled with spring meltwater, could not keep up with the shrinkage and simply slid down the trunks.

“I had gone out the week before and they were expanding. It would be a couple millimeters at a time,” Young-Robertson recalled. “Then we had a big drop in the temperature. It dropped way below zero. It was annoying, because the tree shrank by like half a centimeter, and the bands just dropped off the tree.”

Young-Robertson has since switched to better sensors. With those and other tools, she has made some interesting discoveries that could help firewood cutters who seek the most efficient way to produce dry wood.

That, in turn, could help improve Fairbanks’ notoriously foul air.

Addressing air troubles

With a sharp whack from an axe, Stefan Milkowski split a birch log at the six-acre plot southwest of Fairbanks that he leases from the state for commercial firewood cutting.

Milkowski had already split the log once late last winter to allow it to dry. The chunk of birch, with its outer bark layer impermeable to water, would otherwise have retained nearly all its moisture over the summer.

After the second splitting, in late August, Milkowski stabbed the log’s freshly exposed surface with a moisture meter. The result: 14.5 percent.

Clockwise from top left: In late August 2021, Stefan Milkowski stands next to Young-Robertson’s solar-powered base station, which collects information from nearby sensors in the trees and soil; Milkowski tosses dry firewood, cut the previous winter, into his trailer for delivery to a client; Milkowski stabs a moisture sensor into a chunk of firewood that has sat for months in a stack covered with a half-sheet of plywood to shed rain. The meter reads 16 percent on the end of this piece after some damp August weather, but it reported moisture as low as 14.5 percent on a freshly split surface that better reflected the wood’s interior water content.

That reading pleased Milkowski, especially since his wood had just sat through one of the wetter Augusts on record. If the moisture content had been higher than 20 percent, Milkowski soon would not have been able to sell his product.

On Oct. 1, the state Department of Environmental Conservation set new rules for the local area considered to be in "nonattainment" of federal air quality rules. Any firewood advertised, marketed or sold in the area must have a moisture content of 20 percent or less.

The idea is to reduce wood stove emissions of particulate matter 2.5 micrometers or smaller. That’s about 1/20th the diameter of a human hair.

When burned, wet wood produces far more of these particles than dry wood does.

The airborne particles can travel deep into lungs and even enter the bloodstream. They cause a variety of health problems, especially among people with heart or lung diseases, according to the U.S. Environmental Protection Agency.

Given that situation, Milkowski said, he is happy to play a small part in Young-Robertson’s project.

Photo caption: At top, a spring-loaded sensor detects expansions and contractions in a birch tree near Milkowski’s woodlot in August 2021. Below that, covered by the small white box labeled “3,” probes in the trunk record the wood’s moisture content.

Young-Robertson and her husband Matt Robertson, a field technician at UAF’s Forest Soils Laboratory, set up a variety of sensors in the woods just off Milkowski’s sale site.

The state-of-the-art tree sensors — not the old drop-prone metal bands — are bolted to birch trees near Milkowski’s firewood stacks. Each sensor has a spring-loaded metal rod that rests against the exterior bark and moves with expansion and contraction. That movement is recorded electronically.

Nearby, a white plastic box, just a few inches long, clings to the bark. Covered by the box, three metal rods fit tightly into holes drilled deep into the trunk. They record water content.

More sensors nearby measure soil moisture content, soil and air temperature, humidity, and rainfall.

The data flows back to a nearby base station powered by a solar panel and car battery.

It’s all helping to paint a continuous, real-time picture of forest conditions, the first step in identifying the best time to harvest firewood.

“We’ve been really hopeful that some of what they learn could help us cut wood efficiently in a way that will dry as quickly as possible,” Milkowski said. “In the big picture, we see firewood as a renewable resource if it’s harvested in a sustainable way and, especially in light of our air quality problem in Fairbanks, if it’s burned when it’s dry.”

Strange tree behavior

Young-Robertson earned a doctorate in ecology and evolutionary biology at the University of Arizona. She’d grown up in New Mexico, but she decided “the desert wasn’t my thing.” So she moved north, becoming a postdoctoral researcher at UAF in 2009.

She later found a research faculty position at what is now the Institute of Agriculture, Natural Resources and Extension. She began tracking tree moisture about a decade ago.

“While I was measuring the tree water content, I was thinking ‘How can this be applicable to other things? How can this be useful?’” she said.

A lifestyle change sparked the firewood angle. In 2016, she married Robertson, the Forest Soils Lab technician. They burn wood to heat their home.

Young-Robertson pauses while talking about her research during a visit to the Fairbanks Experiment Station in March 2020.

“He noticed that if you go out at certain times of year, the wood’s a lot heavier, and it takes a lot more effort to get back to your vehicle,” she said. “So we thought, ‘Well, our data could be useful for understanding when is the best time of year to harvest your wood, based on the tree water content.’”

The incredible shrinking trees in April had already provided some of the first clues. But their 29 sensors, now installed not only at Caribou Creek and Milkowski’s firewood sale but also at their home off Murphy Dome, have since revealed even more intriguing tree behavior.

“They’re a lot more active than people have thought,” she said.

“In the fall, they drop their leaves and you’d think, ‘Oh, OK, they’re dry,’” Young-Robertson said. “But, no, they get really, really wet again in the fall after they drop their leaves.”

And they stay wet until freeze-up.

The most surprising data turned up in late October 2018. Sensors on birches showed that the trees dumped a large amount of their water within just 24 hours after temperatures dropped below freezing and remained there.

“When it gets below freezing, and it’s sustained below freezing, that’s when they start putting their water back into the ground and that’s when they start drying,” she said.

They’re still working on understanding exactly how much water is lost when the temperatures drop below freezing.

The water in the trees probably remains liquid initially, despite the below-freezing temperatures, because it contains solutes, she said. The solutes could lower the freezing point, similar to the chemical effect seen in saltwater.

“These are very cold-adapted trees,” she said.

Young-Robertson continues to investigate the patterns in this process.

“It seems to depend on the winter. You can have an October where suddenly it drops way, way below freezing and stays there, and they’ll dump a lot of water. Or it will dip just a little below freezing, and they’ll dump less water more slowly,” she said.

“We’re still learning how much they dry over the winter,” she added.

In spring, as temperatures rise and fall around the freezing point, the trees again will pick up and dump large amounts of water.

“These are very cold-adapted trees.”

But when spring temperatures remain firmly above freezing, the trees become highly saturated, she said.

Then, as leaves sprout, the water content in the trunks will drop again. “They can put almost 20 percent of their water up into the air,” she said. The moisture injected into the atmosphere can drive spring cloud formation, she said.

Summer tree moisture seems to vary with rainfall. “If you get a lot of rain and you don’t get a lot of hot sunny days, the water just keeps flowing,” she said.

Young-Robertson has created a website that streams the moisture data and more.

“It has a lot of firewood information in general, but people will be able to tune into tree water content to track it over the year to help them make their decisions for themselves,” she said.

Investigating drying times

On the west end of the Fairbanks Experiment Farm, three pallets of split birch rest under a canopy of white roofing. It looks like any other neatly arranged Fairbanks woodpile, except that this one sprouts a variety of wires and electronic devices.

Wires carry moisture data from sensors that Young-Robertson’s research team inserted in covered firewood stacks drying at UAF’s Fairbanks Experiment Farm in late March 2020. The stacks rest on scales, which provide weight data that reflects moisture losses in the wood. A weather station also simultaneously records information such as temperature and humidity, allowing researchers to correlate such factors when looking at how fast the firewood dries.

The stacks form the second part of Young-Robertson’s investigation: a look into how firewood dries.

“We’re trying to cut at different times of year starting at different water contents in the field and figure out how long it takes them to dry under your sort-of-average conditions,” she said.

The stacks were cut in July, September and October 2020, respectively. They sit on scales, and pieces of wood within them have been outfitted with the same moisture sensors used on the live trees.

A hand-written note records the weight of firewood first stacked in September 2020 at the Fairbanks Experiment Farm.

“We’re measuring their dry-down, through weight, and we’re calibrating it to the sensors,” Young-Robertson said.

The goal is to produce an ideal drying schedule based on when a person has cut their wood.

“What we’re doing here is to recreate something that somebody might have under their woodshed at home,” she said. “Matt followed guidelines for how to stack and split and all that.”

At least one other study has looked at drying times for firewood in the Fairbanks area. But Young-Robertson said the design “wasn’t how someone would typically set it up in their yard.”

Her work also will allow people to react in real time. Tracking the data online will let them find the ideal moment to cut wood.

Providing scientific data

Just uphill from Milkowski’s stacks of firewood at his sale site, a plywood tent platform sits between birch trees.

During the past few years, he and his wife have camped for a few months at their sale site while harvesting in late winter.

“It’s sort of an excuse to be out in the woods in the winter living in a wall tent cutting trees down,” he said. The firewood operation is more of a side gig to his carpentry business, Wall Tent Woodworks.

Milkowski said they bought the firewood sale out of a desire to explore sustainable, “climate-smart forestry.”

They only cut in winter when the ground is frozen. They move the firewood on a Siglin-style plastic sled behind a 1990s-vintage 440 Trail Polaris. In summer, if the logging road is dry enough, they can reach the site with a four-cylinder, two-wheel drive pickup towing a trailer.

Their sale terms let them fell only trees of a six-inch diameter or greater, so much of their site still appears well-forested despite a few years of cutting.

Milkowski pulls a trailer full of dry birch firewood away from his woodlot on state forest land southwest of Fairbanks in August 2021.

Milkowski said he sees Young-Robertson’s study as potentially most helpful to private woodcutters or small commercial operators like him who have flexibility to cut at different times of year.

“There’s a lot of anecdotal advice you can get about cutting trees to minimize moisture,” he said, “but it will be really nice to have some scientific data to validate those or suggest which might be most effective.”