Trees poke through snow at the edge of a boreal forest
Trees poke through snow at the edge of a boreal forest. Such forests cover enough area to rival tropical forests, but until now, they have been historically under surveyed. Credit: Oksana_Gracheva/iStock/Getty Images Plus

Boreal forests run through Canada, Russia, and Scandinavia and represent about a third of Earth’s forested area. Worldwide levels of carbon dioxide fluctuate with the growth and decay of this vast band of trees.

Yet unlike with these forests’ tropical cousins, scientists struggle to collect data on boreal forests. Ground fieldwork is difficult in the remote regions ringing the Arctic Circle, airborne surveys are currently unavailable for large portions of Eurasia’s boreal forests, and survey instruments soon to be operational on the International Space Station will be completely out of range.

But now, scientists can survey boreal forests at higher resolution than ever before. The key to this ability is a technique called spaceborne stereogrammetry.

“Stereogrammetry works similarly to how our eyes see,” said lead research author Paul Montesano, a scientist at NASA Goddard Space Flight Center in Greenbelt, Md. If you look at your hand and close one eye, then the other, you’ll notice that your hand appears to shift from side to side. That’s because each of your eyes sees the world from a slightly different angle, and they rely on your brain to compile both images in postproduction.

Likewise, stereogrammetry takes two flat images of an object from different angles, then combines them into one 3-D rendering. This technology allows researchers to assess the topography of the tree canopy and of the ground below and thus the height of the tree in between.

Researchers can collect detailed images of the sparse perimeter of boreal forests—a biome boundary that is at the leading edge of changing climate.

Although the technique can be used anywhere, it’s a potential boon for studies of boreal forest structure, particularly forest height, Montesano explained. The researchers discovered that by taking certain key steps in data collection, they can collect detailed images of the sparse perimeter of boreal forests—“a biome boundary that is at the leading edge of changing climate,” Montesano said.

The trick, however, is finding the data taken over the same northern location from two different perspectives.

Thanks to imagery collected from Landsat 7, Landsat 8, and DigitalGlobe satellites, Montesano and his team now have these data. Montesano presented the group’s recent and ongoing research Monday at the American Geophysical Union’s 2017 Fall Meeting in New Orleans, La.

Studying the Boreal Region from Space  

Satellites act as scientists’ eyes in the sky as they orbit the globe. For example, Landsat data provides the spatial patterns of tree cover across the entire circumpolar region. But for stereogrammetry, scientists turn to the DigitalGlobe satellites, which orbit Earth 16 times a day. DigitalGlobe takes two photos of the same patch of ground with each pass—one as it heads toward the site and one looking back after it has passed the site.

The researchers then analyze terabytes of satellite data using NASA Goddard’s computing suite called the Advanced Data Analytics Platform (ADAPT). From this, they can calculate tree heights at the edge of the forest.

But how accurate is the technique? Researchers used stereogrammetry to estimate tree heights at the border of a forest in Siberia, and in Tanana Valley, Alaska. They then compared their results against existing ground and lidar surveys of tree heights and found good agreement.

As they analyzed images, scientists learned to take the Sun’s position into account at the time of sampling. If the Sun was high in the sky above the sampled area, the image had less contrast and better captured details near the ground. Samples taken when the Sun was low had more distinct shadows and showed detail in the canopy. “It’s the same reason people prefer to take photos in the evening, when the light doesn’t wash out their features,” said Montesano.

High-resolution estimates of forest height at the edge of a boreal forest in Russia
High-resolution estimates of forest height at the edge of a boreal forest in Russia. Here, in what’s known as the known as the taiga-tundra ecotone, forest structure change can signal the effects of global warming. Credit: Paul Montesano

By combining images from low- and high-angle light, the researchers could capture forest height in great detail. With their technique refined, they then assessed tree height in uncharacterized boreal regions and canopy cover in the Brooks Range, Alaska, and Kheta-Khatanga plain, Siberia.

Their technique allows the researchers to resolve tree heights at the edge of the forest as small as 2 meters. According to their conference poster, these surveys are revealing distinct patterns in forest structure “at unprecedented detail from space.”

Forest Structure and Climate Change

Theory holds that climate change will move tree lines north, Montesano explained. The scientists involved in this study hope to track patterns of shift and resultant forest growth through time. If the biome boundary is indeed moving, this research should help clarify what type of carbon flux coincides with the shift.

Tree cover and overall forest structure influence other factors that affect local climate. For example, snow-covered, treeless surfaces reflect more light than if those surfaces were masked with darker tree cover. These darker surfaces absorb heat and shade the ground, making for complex feedbacks at the edge of the boreal forest that are better understood with detailed maps of forest structure.

“Forest structure impacts climate on a local, regional, and global scale.”

The complex feedbacks highlight a key theme: “Forest structure impacts climate on a local, regional, and global scale,” Montesano said.

Thus, on a warming planet, it is increasingly important to monitor boreal forests and understand their role in controlling greenhouse gases and local temperature, noted Abigail Swann, a professor of atmospheric studies and biology at the University of Washington who was not involved in the study.

“Forests at high latitudes should change quickly—especially as we’re seeing seasonal cycles get larger,” Swann said. Will that expectation play out in the data? We don’t know, she explained.  “Most data have been collected on tropical forests, so it would be useful to know how forests at higher latitudes are changing over time.”

Thanks to refined stereogrammetry, scientists may soon have the answers they need and will be able to see the forest for the trees.

—Nicoletta Lanese (email:; @NicolettaML), Science Communication Program Graduate Student, University of California, Santa Cruz

Correction, 14 December 2017: Text has been updated to better represent Landsat’s role in mapping tree cover, and forest structure’s effect on climate.


Lanese, N. (2017), Eyes in the sky look closer at under-surveyed northern forests, Eos, 98, Published on 14 December 2017.

Text © 2017. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.