An international study that combined satellite information with soil and relief data from the vast Amazon biome has revealed a comprehensive and heterogeneous overview of the varying degrees of vulnerability to drought conditions. The research, described in the journal Nature in May, included work by Brazilians and sought to identify why the vegetation responds so differently in different geographical locations.
The distribution of various characteristics, including water table depth, tree height, and soil fertility, were included in the model, which ultimately showed that the vegetation naturally most sensitive to water shortages is found in the most productive and fertile lands in the southern and southeastern Amazon. The location is concerning because it is also the region most impacted by the so-called arc of deforestation, which stretches from the south of Pará to the north of Mato Grosso.
“There are many studies of the Amazon that use a local scale to determine the impacts of climate on ecosystems, especially drought, but our research looked at the climate situation from a broader perspective,” says US biologist Bruce Nelson, a remote sensing expert from the Brazilian National Institute of Amazonian Research (INPA) and coauthor of the study, describing the holistic approach adopted by the team. “It’s a great example of how data from the ground, satellites, and plants can be used together to consider the system as a whole.”
Southern Amazon: susceptible rainforest in the arc of deforestation suffers severe human pressureEvaristo Sa / AFP via Getty Images
Chinese spatial analysis expert Shuli Chen of the University of Arizona, USA, lead author of the study, spoke with Pesquisa FAPESP via email. She expressed concern that many of the environmental areas identified as less resilient to drought are in the eastern areas of the Amazon biome. The remaining rainforest in the eastern region is also responsible for releasing an immense volume of water vapor into the atmosphere that then heads westward (see Pesquisa FAPESP issue n° 226), providing about 50% of the water needed by trees in the west. Recycled by the forest, this water then continues its journey through the sky before it falls as rain in the rest of Brazil and elsewhere in South America. These systems have been dubbed flying rivers. “Losing vegetation in this part of the Amazon therefore does damage beyond that caused to the rainforest or agriculture in South America: interrupting the water cycle affects the integrity of the entire global system,” explains Chen.
One of the people who helped popularize the concept of flying rivers is geoscientist Antonio Donato Nobre, a retired researcher from the Brazilian National Institute for Space Research (INPE) and coauthor of the recent article in Nature. It was his group’s previous work on detecting and mapping groundwater in the Amazon region using a topographic model known as “HAND” (Height Above Nearest Drainage), that the head of the new study, American climate scientist Scott Saleska of the University of Arizona, decided to look for factors that might explain the observed differences in vegetation death patterns. “Many of the areas that suffer most from drought seemed to do so directly and primarily due to the depth of the water table, but we needed to investigate those that could not be explained by this factor to get a more precise overview of the situation,” said Saleska via email.
With this more detailed overview and more in-depth information about the mosaic of fragility in the Amazon, the scientists argue that conservation plans and public policies can become more focused and efficient.
Nobre emphasizes that this does not mean areas that were not defined as priorities can be deforested, since different parts of the rainforest depend on each other in terms of climate dynamics. The secret to tackling the climate crisis, he says, is the diversity of the system. “The rainforest is complex and needs to be respected in its complexity and integrity.”
Evolution and relief
In general, the northern Amazon has proven better able to cope with intense droughts than the south of the biome. However, there are also resilient regions in the south: floodplains (with vegetation known as igapó) and areas where the water table is close to the surface, giving easy access to water. In igapó forest, dry periods can even be a good thing, offering temporary relief and oxygenation for normally submerged roots—as long as the drought is short-lived.
Where the water table is deep, species with roots that reach the water have survived for hundreds of years. As deep roots usually belong to older and taller trees, plant height was also one of the factors used in the resilience map.
The water table criterion does not apply to the entire biome, however. The vegetation with the greatest resilience of all was found in the Guiana Shield, a plateau in the far north that is home to the biggest trees in the Amazon (see Pesquisa FAPESP issue no. 336). The distance between the surface and groundwater in the Guiana Shield is not a determining factor. This finding could be explained by a particular well-known characteristic of the region: the infertile soil.
Studies from recent years have shown that tree growth is extremely slow in nutrient-poor soils, leading to denser wood and internal vessels, possibly meaning greater resistance to embolisms, a process in which the internal vessels of a plant collapse due to the entry of air where there should only be water when the soil dries out too much.
The size of the trees, the fertility of the soil, and the proximity of the water table all contribute, each with their own advantages and disadvantages that offset one another. The worst-case scenario, which occurs frequently around the arc of deforestation, is a combination of fertile soil, low tree heights, and a deep water table. Trees under these conditions are prone to embolisms, as their roots are unable to reach the water table.
Another parameter used by the research group was that of “greenness.” The Enhanced Vegetation Index (EVI) quantifies the concentration of healthy leaves in the upper canopy. In addition to the EVI, they also used direct measurements of photosynthesis taken via solar-induced chlorophyll fluorescence (SIF). Both indices are based on information collected by NASA satellites over a 20-year period.
The researchers explained that they took photosynthesis rates into account because EVI and SIF are correlated with plant mortality and growth rates. “Embolisms are also fundamental to understanding the study because we discovered, in previous work, that this was the leading cause of tree deaths during one of the worst droughts in the southern Amazon, in 2015,” says Nelson. “Other scientists have also demonstrated that they were the cause of death in trees exposed to a long artificial drought,” he adds, referring to a study by a group led by British ecologist Lucy Rowland, currently at the University of Exeter, UK (see Pesquisa FAPESP issue n° 238).
More models
Other threats to the water cycle can be added to the models to make them more detailed, accurate, and applicable to the mission of designing ways to preserve the Amazon’s ecosystem and biodiversity. Mathematician Marina Hirota of the Federal University of Santa Catarina (UFSC) and her colleagues focused on the need to address the issue through multifactorial approaches in an article that made the cover of Nature in February of this year.
Hirota also came up with a heterogeneous map of responses and weaknesses in the Amazon. To do so, she combined a series of other environmental disturbances, including the propensity for death by floods, wildfires, and deforestation, in addition to past droughts. The team arrived at an estimate that by 2050, between 10% and 47% of the Amazon rainforest will be at risk of passing the point of no return, when the ecosystem will no longer be able to recover from the impacts it has suffered. In addition to these climate disturbances, the researchers also considered forces that tend to protect vegetation, such as the boundaries of indigenous reserves.
According to Hirota, scientists are discovering that some regions of the Amazon are more resilient to climate change than was known before the 1990s, when people first began to talk about climate inflection points for the entire biome in a homogeneous way. She stresses that the fact that the rainforest has some mechanisms of resistance does not mean that it will not succumb if the changes are too severe. “It just means we have a little more time—but not much,” she emphasizes.
Hirota also coauthored an article published in Nature in 2023 as part of Shuli Chen’s research, which indicated that the rainforest in the southeastern Amazon is evolutionarily more resistant when evaluated by resistance to embolisms. “The more negative the measurements, the more water tension the tree can withstand in its vessels before suffering an embolism,” she explains. But the threat to the region is described by another measure, known as HSM (hydraulic safety margins), which show whether plants are operating within safe ecophysiological limits to avoid death by drought. Based on the HSM, although the southeastern rainforest is more drought-resistant, it is already functioning outside safe limits because it is experiencing a greater water deficit. Although it is less drought-resistant, the western, southwestern, and northwestern rainforest is not so greatly impacted by changes in water availability. The research was led by ecologist Julia Tavares, currently a postdoctoral researcher at Uppsala University in Sweden. Further studies that integrate the indicators and characteristics of each part of the rainforest should allow for a better understanding of the risks and the areas most likely to remain intact.
Antonio Nobre attributes defense strategies to natural systems that he describes as elegant and complex, especially with respect to biological and geophysical diversity. “Such systems, when intact, tend to respond adequately within their capacity to self-regulate—until limits are exceeded, leading to ecological collapse, which is what we are experiencing now.”
One adaptive response could be the emergence of new flora, resistant to modern conditions
American climate scientist Scott Saleska and Brazilian mathematician Marina Hirota both looked at vegetation from a functional perspective, rather than the composition of the flora, choosing to focus on the capabilities and properties of plants without distinguishing between species. Another study, described in the journal Global Change Biology in 2018 by Brazilian ecologist Adriane Esquivel-Muelbert, then at the University of Leeds, UK, was based on 30 years of data on Amazonian species and detected that the composition of the trees was already changing.
The ecologist found that fewer individuals of species that survive better in wet environments are growing in the Amazon, being replaced by plants normally found in drier environments where there is more carbon dioxide in the atmosphere. The gradual change to trees that survive in these conditions, however, is not occurring at the same pace as global climate change. The evidence points in the same direction: climate change has been too rapid for the rainforest to adapt.
The story above was published with the title “Amazonian resilience” in issue 345 of November/2024.
Scientific articles
CHEN, S. et al. Amazon forest biogeography predicts resilience and vulnerability to drought. Nature. Vol. 631, 111–7. June 19, 2024.
FLORES, B. M. et al. Critical transitions in the Amazon forest system. Nature. Vol. 626, 555–64. Feb. 14, 2024.
TAVARES, J. V. et al. Basin-wide variation in tree hydraulic safety margins predicts the carbon balance of Amazon forests. Nature. Vol. 617, 111–7. Apr. 26, 2023.
ESQUIVEL-MUELBERT, A. et al. Compositional response of Amazon forests to climate change. Global Change Biology. Vol. 25, 39–56. Nov. 8, 2018.
