How light reflects on leaves may help researchers identify dying forests

Trees at UNDERC
Photo Credit: Barbara Johnston/University of Notre Dame

Early detection of declining forest health is critical for the timely intervention and treatment of droughted and diseased flora, especially in areas prone to wildfires. Obtaining a reliable measure of whole-ecosystem health before it is too late, however, is an ongoing challenge for forest ecologists.

Traditional sampling is too labor-intensive for whole-forest surveys, while modern genomics—though capable of pinpointing active genes—is still too expensive for large-scale application. Remote sensing offers a high-resolution solution from the skies, but currently limited paradigms for data analysis mean the images obtained do not say enough, early enough.

A new study from researchers at the University of Notre Dame, published in Nature: Communications Earth & Environment, uncovers a more comprehensive picture of forest health. Funded by NASA, the research shows that spectral reflectance—a measurement obtained from satellite images—corresponds with the expression of specific genes.

Reflectance is how much light reflects off of leaf material, and at which specific wavelengths, in the visible and near-infrared range. Calculated as the ratio of reflected light to incoming light and measured using special sensors, reflectance data reveals a unique signature specific to the leaf’s composition and condition.

The Gillen Director of UNDERC and lead author of the study Nathan Swenson
Photo Credit: Barbara Johnston/University of Notre Dame

“This has the potential to revolutionize forest health monitoring,” said Nathan Swenson, the Gillen Director of the University of Notre Dame Environmental Research Center (UNDERC) who led the study. “By connecting reflectance with gene expression, we can get a real-time measure of forest health at the genomic level that picks up the early indicators of declining forest health and connects them back to real changes happening on the cellular level.”

While reflectance is a strong indicator of both physical and chemical leaf properties, the utility of knowing these features is limited without the ability to determine their molecular origin.

“We now have the ability to fly an airplane over a whole forest and rapidly document the traits of every tree’s canopy, but what we can actually say about a certain tree’s condition is still quite simple,” said Swenson, professor in the Department of Biological Sciences. “So, we wanted to go beyond that, asking: Is there a significant relationship between the reflectance of a leaf and its gene expression?”

In short, the answer is yes.

Swenson, with the help of graduate students and postdoctoral scholars, collected leaf samples from two common tree species—sugar maple and red maple—at the University’s UNDERC field site in northern Wisconsin and the Upper Peninsula of Michigan.

At the point of collection, reflectance data for the surface of each leaf was measured and recorded, before the sample was preserved and processed for gene expression analysis. This analysis focused on genes related to water response, drought, photosynthesis and plant-pest or plant-pathogen interactions. The reflectance data was also processed to determine the wavelengths of light reflected or absorbed by a particular leaf.

For more than half of the genes analyzed, the researchers found a strong correlation with specific reflectance wavelengths. This means that across most of the trees surveyed, those whose leaves expressed a certain gene reflected or absorbed the same “signature” wavelengths of light as other leaves that expressed the same gene.

“We’ve done it here on just a small scale, but the potential for predicting the expression of hundreds to thousands of ecologically important genes from reflectance is immense,” Swenson said. “We could monitor whole forests on the genomic scale, via sensors on the international space station.”

To apply this newly-defined correlation to whole forests, Swenson is looking to scale previous research. A 2024 study published in PLOS Biology combined satellite images with artificial intelligence-enabled computational networks to create tree species maps for the National Ecological Observatory Network.

The AI model, developed by a multi-institutional team including Swenson, can be trained to identify particular trees by species using images of the whole forest’s canopy collected by sensors. When layered together with reflectance and gene expression data, the model has the potential to generate a complete profile for a single tree based on its species, reflectance signature and the gene expression map for that species. Doing so would allow researchers to single out struggling individuals or clusters more efficiently for intervention.

"You can take these models that we're generating at the leaf level and apply them to those new data sets of reflectance whether that's from an airplane or from a satellite. And then you can build a map of gene expression on the scale of a national forest,” Swenson said. “The end goal here is using the right data to rapidly assess how trees are responding to stressors, so that we can intervene before the forest hits a crisis point.”

Such an undertaking requires the input of experts in remote sensing, genomics and ecology, all of which are members of Swenson’s research team within the University’s Department of Biological Sciences. Co-authors of the Nature Communications study include postdoctoral scholar Yanni Chen, graduate student Alexander Cox, and former graduate students Logan Monks and Vanessa Rubio.

“This work doesn’t happen without scientists from vastly different fields, ecologists alongside genomicists alongside data scientists, sitting down at a table together and engaging with the same question from different angles,” Swenson said. “We need all of our individual strengths pulling together to meet these challenges.”

Published in journal: Communications Earth & Environment

Title: Linking leaf hyperspectral reflectance to gene expression

Authors: Yanni Chen, Logan Monks, Vanessa E. Rubio, Alexander J. Cox, and Nathan G. Swenson

Previous Study: Individual canopy tree species maps for the National Ecological Observatory Network

Source/Credit: University of Notre Dame | Erin Fennessy

Reference Number: env010826_02

Privacy Policy | Terms of Service | Contact Us