Satellites spot forest stress two years before bark beetle die-offs become apparent
A faint signal emitted during photosynthesis may reveal physiological stress well before tree mortality appears in aerial surveys, potentially giving forest managers more time to prepare before mortality becomes widespread
image:
Pixels of forested lands that were categorized as a) control (low tree mortality by wildfire and bark beetles from 2011 through 2023), b) wildfire (fire-driven mortality in 2020 and 2021) and c) bark beetle (insect-driven mortality for 2021 and 2022).
view moreCredit: Kunik et. al. Rem Sen Enviro (2026)
Satellite measurements detected declining photosynthetic activity in Western U.S. forests two years before bark-beetle mortality appeared in aerial detection surveys, according to new University of Utah-led research.
The first-of-its-kind study suggests that satellite-observed chlorophyll fluorescence (SIF), a measure of plant photosynthesis, could provide an early warning of forest stress. As drought, wildfire and insect outbreaks occur with increasing frequency across the American West, the technology could also help scientists understand how these disturbances affect forests’ ability to absorb and store carbon from the atmosphere.
“I don't know of any other tool that can detect this type of signal before tree mortality becomes obvious at a scale large enough to assess the health of entire forests,” said lead author Lewis Kunik, who recently completed his doctorate at the U. Kunik’s doctorate was jointly advised by study co-authors and U professors John Lin in the Department of Atmospheric Sciences and David Bowling in the School of Biological Sciences.
“The ultimate goal isn't to predict the exact tree that will die. Rather, the technology could identify areas of concern early enough for land managers to investigate, mobilize crews, allocate funding or otherwise prepare before mortality becomes widespread.”
The study is online ahead of its publication in the October issue of Remote Sensing of Environment.
Fluorescence provides an early warning signal
Many satellites monitor forest health using signals such as greenness and canopy structure because some when some trees become stressed, they might wilt or drop their leaves. But pines, spruces, firs and other evergreen trees present challenges for satellite monitoring: they can keep their needles even while photosynthetically dormant, such as during winter or under other high-stress conditions. This matters for Western U.S. forests, which are largely dominated by evergreens.
Fortunately, several next-generation satellites carry specialized instruments which detect a faint red glow that plants emit during photosynthesis, the process by which plants convert sunlight into energy. This signal is known as solar-induced fluorescence (SIF)—when a leaf’s chlorophyll molecules absorb radiation, some radiation re-emits at longer, red wavelengths known as fluorescence.
When plants get stressed, they absorb more light than they can use, reducing their efficiency and dimming their red glow. By tracking SIF relative to the amount of light absorbed over time, the researchers could identify subtle physiological changes in in evergreen trees that conventional satellite metrics can miss.
The authors used SIF observations from TROPOMI, the instrument on the European Sentinel-5P satellite, to compare changes in fluorescence patterns in forests affected by wildfire- and insect-caused tree mortality with non-affected control areas with similar biogeographic characteristics in forests across the American West.
In forests that would later experience bark-beetle mortality, the researchers detected a significant SIF decline roughly two years before mortality was seen in aerial surveys conducted by the USDA Forest Service. Drought stress alone couldn’t explain the signal. While nearby healthy forests experienced similar levels of drought, the decline in SIF from healthy forests was 10-20% less severe than the decline in the bark beetle-infested forests.
SIF can change for many reasons, including drought, insect infestation, canopy dieback, changes to seasonal timing of growth, reduced sunlight and changes in the mix of plants growing from the forest floor to the top of the canopy. The researchers accounted for these factors, but the complexity of forest ecosystems makes year-to-year changes in SIF difficult to interpret.
In this case, however, their analysis revealed a clear pattern. The findings suggest that SIF signal could provide an early warning of forest stress that precedes widespread tree mortality.
The researchers validated their approach using wildfire mortality as a testbed for detecting SIF changes from a wide range of mortality severities. They saw declines in SIF that scaled proportionally with the amount of vegetation lost due to fire.
“Wildfire mortality has more predictable impacts to forest productivity than bark beetle mortality. There is also a lot more wildfire-affected land to study, and we can use established tools to estimate the severity of those events. Testing our method on wildfires really helped build confidence in our bark beetle assessment,” Kunik said.
The researchers were also able to use SIF to monitor how the ecosystem recovered from wildfire, highlighting the technology’s potential for tracking how disturbances alter forest productivity and carbon cycling over time. Because forests store massive amounts of carbon, tracking these changes will help scientists better understand how disturbances affect the carbon balance of western forests.
“SIF is an emerging tool that Earth scientists can use to show the fingerprint of plant CO2 uptake at regional or global scales,” said Kunik. “Drought, wildfire and bark beetle outbreaks can weaken a forests’ ability to absorb carbon and may release the carbon stored in the trees. Tracking these changes will help us understand whether these disturbances potentially turn forests from carbon absorbers to carbon sources.”
SIF outperformed the other satellite metrics
While other SIF-observing satellites exist, the authors used data collected by TROPOMI due to its coverage and sampling frequency.
The study tested the sensitivity of SIF measurements against several forest-health and vegetation-productivity remote-sensing measures. These included other frequently used satellite signals like land surface temperature, as well as vegetation indices like the Normalized Difference Vegetation Index (NDVI).
SIF showed greater sensitivity to bark beetle mortality than the other canopy remote sensing products tested. It also showed significant stress-related declines earlier than the other products, and roughly two years before aerial surveys first detected mortality.
“The results are exciting because they show the potential of SIF to provide information on forest health over a large spatial region,” said Lin. “Future satellites, such as the European Space Agency’s FLEX mission, will provide SIF at much higher spatial resolution and add to the growing record of SIF that will reveal more patterns in the future.”
The authors are excited for SIF’s potential. The project began through conversations with USDA Forest Service collaborators who have long wanted an early warning system to help them manage forests.
“They want to know as soon as possible when forests may cross a threshold of stress that leave it vulnerable to pests, pathogens or other drought-related impacts,” said Kunik. “We still can’t predict if, or where, mortality will occur just based on SIF observations, but our work shows that SIF could be another powerful tool for identifying areas of concern.”
******
Other co-authors of the study include Brett Raczka, University of Utah and National Center for Atmospheric Research; Jeffrey Hicke, University of Idaho; Christian Frankenberg, California Institute of Technology; Rui Cheng, Claremont McKenna College; and Michèle Slaton, Inyo National Forest, USDA Forest Service.
This research was funded by the NASA Carbon Monitoring Systems program (Award (80NSSC20K0010), the National Science Foundation Graduate Research Fellowship Program (award #2139322) Wilkes Center for Climate Science & Policy at the University of Utah and by the USDA Forest Service Western Wildland Environmental Threat Assessment Center (Agreement #22-JV-11261994–036-M2).
The publication:
Kunik, Lewis et al. (2026). Characterizing effects of tree mortality from wildfire and bark beetles using satellite observations of solar-induced chlorophyll fluorescence. Remote Sensing of Environment, 334(October special issue), 10.1016/j.rse.2026.115550.
Multimedia access
https://www.youtube.com/watch?v=bAhXV6ntsZg
Visualization tracking the Earth’s carbon cycle from 2015 to 2026 using observations from NASA’s Orbiting Carbon Observatory missions. The animation shows the relationship between atmospheric carbon dioxide levels (blue regions contain less CO2 than the baseline, red regions contain more) and photosynthetic activity of terrestrial vegetation. The left map shows atmospheric carbon dioxide; The right map shows solar induced chlorophyll fluorescence (SIF) levels.
Credit: Christian Frankenberg/Caltech
A) the average insect-induced mortality from 2022-2023 USDA aerial surveys and C) 2021 late-summer mean solar induced fluorescence relative to 2019 levels, a low insect-mortality year. Redder hues represent lower levels SIF, a sign of lower photosynthetic activity.
Credit
Adapted from Kunik et. al. Rem Sen Enviro (2026)
Adapted from Kunik et. al. Rem Sen Enviro (2026)
Journal
Remote Sensing of Environment
Method of Research
Observational study
Subject of Research
Not applicable
Article Title
Characterizing effects of tree mortality from wildfire and bark beetles using satellite observations of solar-induced chlorophyll fluorescence
Article Publication Date
1-Oct-2026
New study demonstrates TanSat-2's potential to distinguish human carbon emissions from ecosystem carbon sinks
Institute of Atmospheric Physics, Chinese Academy of Sciences
As the world works toward a clearer picture of where carbon is emitted and absorbed, satellites are becoming increasingly important tools for tracking the global carbon cycle. China is preparing to launch its next-generation carbon monitoring satellite, TanSat-2, which is designed to provide global and regional observations of column-averaged carbon dioxide (CO2) and methane (CH4).
One of the major challenges in satellite-based carbon monitoring is distinguishing CO2 released by human activities from the carbon exchanged between the atmosphere and terrestrial ecosystems. Atmospheric CO2 observations contain signals from both processes, making it difficult to determine how much of the observed CO2 originates from fossil fuel emissions and how much is associated with ecosystem carbon uptake and release.
A recent collaborative study by Chinese and UK researchers published in Advances in Atmospheric Sciences explores how TanSat-2 could help address this challenge.
This study introduces a newly developed carbon flux inversion approach that combines atmospheric CO2 measurements with solar-induced chlorophyll fluorescence (SIF), an observational proxy closely related to vegetation photosynthetic activity. By bringing these two complementary measurements together, the approach can simultaneously constrain net primary productivity (NPP) and fossil fuel combustion emissions (FF).
To further improve the separation of natural ecosystem carbon fluxes and fossil fuel emissions, the researchers applied empirical orthogonal function (EOF) analysis to prior NPP and FF inventories. This identifies their dominant spatial and temporal patterns while reducing the number of variables that need to be optimized in the inversion.
The researchers then put the proposed TanSat-2 observing strategy through a virtual test using Observing System Simulation Experiments (OSSEs). Under idealized conditions in which observational biases are effectively controlled, the results show that TanSat-2 CO₂ and SIF observations could lead to an NPP error reduction of up to 95% over Siberia and the Amazon, and to error reductions of about 80% for FF emissions in regions including Siberia, northern Asia, the United States, and South Africa.
The study, however, also highlights an important challenge. Even small systematic biases in satellite-derived column-averaged dry-air mole fractions of CO2 (XCO2) can substantially distort inferred CO2 sources and sinks—or even lead to their misattribution. Such biases could ultimately compromise the reliability of carbon flux estimates, underscoring the importance of carefully identifying and correcting systematic errors in satellite retrievals and subsequent data applications.
The researchers also investigated whether increasing TanSat-2’s cross-track swath width could strengthen its carbon-monitoring capability. Their results indicate that a wider observational swath can improve the robustness of CO2 flux estimates. The study further develops an error-matrix analysis framework that could help evaluate observing strategies and support the design and optimization of future carbon-monitoring satellite missions.
“Reliable monitoring of carbon fluxes at both global and regional scales cannot rely on satellites alone.” The lead author, Dr. YANG Dongxu from the Institute of Atmospheric Physics at the Chinese Academy of Sciences concluded their study and emphasized, “Instead, it requires the integration of satellite, ground-based, airborne, and in situ observations. Together, these complementary observing systems will be essential for building a more accurate and comprehensive picture of Earth’s carbon sources and sinks.”
Article Title
A step forward to global segment CO2 flux estimation benefiting from large swath of coordinated CO2 and SIF measurement from TanSat‑2 mission
Article Publication Date
2-Sep-2026
