Global satellite forest loss mapping conflates transient canopy disturbance with forest destruction
This study demonstrates that the widely used Hansen Global Forest Change product often misidentifies severe, temporary canopy disturbances in fire-adapted eucalypt forests as permanent ecological forest loss, as these areas typically recover structurally and biologically within five years despite initial satellite signals of deforestation.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Forests are the lungs of the planet, storing vast amounts of carbon and sheltering countless species. When trees disappear, the consequences ripple through the climate system and the web of life. For decades, scientists have relied on satellite images to track where forests are vanishing, using these maps to understand global environmental change. These satellite systems work by watching for sudden changes in the color of the land. When a forest is cut down for farming or a fire sweeps through, the green canopy turns brown or bare, and the satellite sensors register this as a loss. However, not all green turning brown means the forest is gone forever. In many parts of the world, particularly in forests adapted to fire, trees can lose their leaves or have their branches scorched but survive to grow back. The challenge for scientists is distinguishing between a forest that has been permanently destroyed and one that has merely been injured and is in the process of healing. If we mistake a recovering forest for a lost one, we risk drawing the wrong conclusions about the health of our planet and the effectiveness of our conservation efforts.
A team of researchers set out to test whether the most widely used global map of forest loss is making this exact mistake. They focused on the aftermath of the 2019–2020 Black Summer fires in Australia, a catastrophic event that burned millions of hectares of eucalypt forest. These forests are famous for their ability to bounce back after fire, often regrowing vigorously even after severe damage. The researchers examined the data from the Hansen Global Forest Change product, a standard tool used by governments and scientists worldwide to track deforestation. This tool flagged large areas of the burned landscape as "forest loss." The team wanted to know if this label was accurate. Did the satellites see a forest that was truly dead, or were they simply seeing the temporary scar of a fire that the trees were already healing from?
To find the answer, the researchers looked closely at the Blue Mountains region in New South Wales, an area dominated by fire-adapted eucalypt trees and largely untouched by logging or land clearing. They compared the global forest loss map with their own detailed observations of fire severity, satellite images of the forest canopy, and measurements of tree structure taken from the ground. They found that the global map was indeed picking up the fires, but it was conflating the intensity of the damage with the permanence of the loss. The map labeled areas as lost only where the fire had been so intense that it scorched or consumed the canopy completely. In these extreme cases, the trees looked dead from space. However, the researchers discovered that this visual signal did not mean the forest was gone.
Five years after the fires, the researchers returned to check on these "lost" forests. They found that the trees had recovered remarkably well. The satellite images showed that the canopy had regrown, with most of the areas classified as lost having regained at least 80 percent of their pre-fire greenness. When they measured the physical structure of the forest, including the density of the leaves and the amount of wood in the trees, the data told the same story. The forests were not empty; they were rebuilding. The researchers also walked through the burned areas to count the trees directly. Even in the most severely burned spots, where the fire had been at its worst, they found that only about 17 percent of the tree trunks had actually died. The majority of the trees, though stripped of their leaves and branches, were still alive and had begun to sprout new growth from their trunks or roots.
The study concludes that the global satellite map is not measuring the death of the forest, but rather the severity of the injury. In fire-adapted eucalypt forests, a massive fire can strip the canopy bare, causing the satellite to record a "loss," but the forest itself is often just resting before it regrows. The map is effectively a detector of disturbance, not a detector of permanent destruction. This distinction is crucial. If scientists and policymakers treat these recovering forests as if they are gone forever, they may overestimate the amount of carbon released into the atmosphere and misunderstand the resilience of these ecosystems. The researchers suggest that while the satellite data is useful for spotting where fires have burned, it must be interpreted with caution. In landscapes where trees are built to survive fire, a scar on the satellite image does not necessarily mean the forest has been erased from the earth.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.