- A team of researchers looked at changes in tree richness across the lowland and montane forests of the Andes and Amazon over the last four decades.
- While their results didn’t show an overall shift in any one direction, they found that tree richness changed significantly across the six subregions: forests in the central Andes, Guyana Shield and central-eastern Amazon have been losing species, while the northern Andes and western Amazon showed increased tree richness.
- Changes in the seasonality of precipitation, total rainfall, temperature, as well as the degree of forest fragmentation are key drivers for tree richness: forests that warmed more since 1971 lost species faster than those moderately warming; but regionally, precipitation plays a bigger role than temperature in richness changes.
- Forests with a higher number of trees and landscape integrity gain species, so limiting deforestation across the Andes–Amazon ecosystems can protect tree richness, in particular the northern Andes, which could serve as a key refuge for species that can no longer survive the warming Amazon.
The tropical forests of the Amazon and Andes are some of the most biodiverse places on the planet, but across both regions, changes in climate and landscape conditions are driving a shift in the number of tree species, recent research has found.
Although the overall number of tree species across the Andes and Amazon hasn’t changed in recent decades, some subregions are gaining species while others are losing them, according to the study published in Nature Ecology & Evolution. Trees provide vital ecosystem services, and changes in their diversity — what experts call tree richness — could have an impact, including on the forests’ role in temperature regulation or carbon storage.
Using more than 40 years of tree diversity data, the study found that species richness declined in the central Andes, the Guyana Shield, and the central-eastern Amazon subregions. Meanwhile, it increased in the northern Andes and western Amazon, and didn’t change significantly in the southern Amazon. Researchers used data from 406 different forest plots across 10 countries, paired with records of climate indicators.
“Forests are changing and now we have evidence that it’s linked to climate change,” said Belén Fadrique, a research fellow at the University of Liverpool in the U.K. and lead author of the study, which involved more than 160 co-authors.
“We do find that a majority of sites are decreasing in richness,” Fadrique told Mongabay in a video interview. In total, 203 plots declined in tree richness and 146 increased, the research found.
Overall, the richness of tree species didn’t change significantly across the entire region studied, but Fadrique said this is a normal effect of averaging out results from a large number of areas. The southern Amazon, for example, has some of the plots with the highest rates of species decline, but on average, didn’t see a statistically significant change.
A delay in vegetation response to environmental conditions could also influence overall results, the study notes.

Enrique Ortiz, a Peruvian tropical ecologist and senior program director of the Andes Amazon Fund, who wasn’t involved in the research, said the study was interesting because it used data based on fieldwork, rather than modeling, spanning a large region and time interval.
“There are some caveats about it, because trying to find patterns on such a big scale is a major endeavor,” he added, echoing the caution of the researchers from the study.
Species richness drivers
“The patterns are complex,” Flávia Costa, a co-author of the paper, told Mongabay in a video interview. Costa, coordinator of environmental dynamics at Brazil’s National Institute for Amazonian Research (INPA), provided data on the central Amazon and helped Fadrique ensure the reliability of the data used in the analysis.
Fadrique analyzed data on species richness against several climate indicators: maximum temperature, total precipitation, and precipitation seasonality, which is the variation of rainfall throughout the year. By tracking how these changed over time, the researchers could identify which ones drove a change in species richness. The period analyzed varied depending on the monitored plot. For example, the oldest species census spanned four decades, starting in the early 1970s. Some censuses monitored the plots for less than 10 years. Many provided data up to 2020.
“Across the study area, hot, dry and seasonal forests and those becoming warmer and more seasonal are losing species, while forests with higher tree density and higher landscape integrity are gaining them,” the researchers wrote.
Over the past four decades, more than 90% of the plots studied warmed on average by 0.028+/-0.018° Celsius (0.05+/-0.032° Fahrenheit) per year, with the central-eastern and southern Amazon regions becoming hotter faster and losing species at a greater pace than those warming only moderately.
Although 39% of plots saw less rainfall, the researchers found that precipitation seasonality, which increased in 88% of plots, has a bigger say in tree richness. That seasonality can also outweigh the influence of temperature; researchers found that although the central Andes saw cooler temperatures, their forests tended to lose species, possibly due to less rainfall and increased seasonality.
“In some regions, it seems that the strongest driver was temperature,” Costa said. “In others, it appeared to be rain seasonality. And laid onto this, there are changes in the forests’ own dynamics.”
Fadrique noted that the total number of trees in a plot seemed to matter for tree richness. “It’s a very direct relationship,” she told Mongabay. “If you cut trees, you’re likely eliminating a species.”

Considering that fire and fragmentation tend to make forests more vulnerable to species loss, the research also looked at landscape integrity, or the extent of forest cover on the plots. A single data point, from 2015, was used for this.
“It seemed that the more intact the forest was around [the studied area], the more there was an increasing trend of species locally. Or the opposite: the more degraded the surrounding forest was, the more the monitoring sites were showing a decline in species,” Fadrique said.
“This confirms what other papers mention: that fragmentation negatively affects the diversity of species,” said Bonifacio Mostacedo, a professor at the Gabriel René Moreno Autonomous University in Bolivia, who wasn’t involved in the research.
The study nonetheless found that the impact of environmental conditions on tree richness was uneven across the areas studied.
Mostacedo noted that species in tropical dry forests, found in parts of the Andes, have different capacities for adaptation from those in tropical wet forests such as the Amazon.
“While fires are severe in dry ecosystems, trees have some adaptive strategies, such as thick bark or a high capacity for regrowth, which makes them more tolerant,” he said. “In contrast, the Amazon Rainforest is more susceptible to fire, and very few species possess these adaptive strategies.”
Impact of diversity loss
The paper didn’t look at which species are migrating or being lost, or at the impact of disappearing species on ecosystems.
Future research should go deeper and look at taxonomy and impacts, Fadrique said. “We know that … the stability of the forest may depend on diversity,” she said.
The changes identified in the paper are subtle — between -1.95% and +3.3% variation in species richness per year depending on the subregion — and aren’t visible to the naked eye, Costa said. “What does a 1% change mean when you have around 200 plants in a monitoring plot? It means that two plants changed — it’s not an impressive change,” she said.
But, she added, “an accumulation of subtle patterns can end up generating serious results.”
There’s still little evidence-based research about the impacts on ecosystems of the decline of species diversity, she said. But forests provide ecosystem services like carbon capture and the regulation of water cycles, and changes like the loss of tree species likely disrupt these services.
“In practical terms, the loss of species certainly means extinctions,” Ortiz said. “This also implies that ecosystem services diminish.”
According to Costa, other elements that could be affected include trophic networks — the various food chains interacting within an ecosystem — and microorganisms living on leaves.
Tackling deforestation
One of the study’s conclusions was that tree abundance helps preserve diversity, underscoring the importance of avoiding deforestation and protecting forests.
“Controlling climate change depends on a global effort; no government will do that alone,” Costa said. “Controlling deforestation is more accessible and has an important effect.”
The Amazon lost 1.7 million hectares (4.2 million acres) to deforestation in 2024, and an estimated 75% of the tropical Andes’ native forests have been converted for urbanization and agriculture. Both biomes continue to be threatened by the expansion of agricultural activities and industries like mining. Protected areas, like national parks, help conserve forests, but their presence varies greatly between countries, Costa said.

“In terms of prioritizing conservation, there are a number of lessons in this paper,” Ortiz said, noting that this should be done through having protected areas that account for the role of local communities, and connecting fragmented forests.
“If you increase connectivity between sites, this will increase the chance of all species to keep on recruiting,” Fadrique said.
According to the study, the northern Andes could serve as refuge for tree species that are displaced as the Amazon becomes hotter.
“The Andes are this escape route for all Amazonian species,” Fadrique said. This is because temperatures at higher altitudes remain cooler, albeit while warming, making the region more attractive for warm-adapted species from the Amazon. This migration of plants to a higher altitude is known as thermophilization.
The northern Andes is also a wet region, like the western Amazon, which favors species migration. The central Andes doesn’t have the same potential because it’s drier and more seasonal than the northern Andes. It also appears to be getting cooler, the study found.
The Colombian Andes, part of the northern Andes, lost 3,671 square kilometers (1,417 square miles) of forest between 2013 and 2018, but it concentrates the majority of the country’s national parks, and these could help keep deforestation below 1%, according to a different study.
Ortiz said the role of the northern Andes is “an important detail” that shows the importance of maintaining continuous forests between lower and higher elevations, functioning as climate corridors that allow species to migrate.
“We need to look at the Amazon and the Andes as a whole,” he said.
Banner image: Emergent tree in the Amazon Rainforest, Ecuador. Image by Rhett A. Butler/Mongabay.
Citations:
Fadrique, B., Costa, F., Cuesta, F., Arellano, G., Cayuela, L., Baker, T. R., … Phillips, O. L. (2026). Tree diversity is changing across tropical Andean and Amazonian forests in response to global change. Nature Ecology & Evolution, 10, 267-280. doi:10.1038/s41559-025-02956-5
Schnabel, F., Beugnon, R., Yang, B., Richter, R., Eisenhauer, N., Huang, Y., … Bruelheide, H. (2025). Tree diversity increases forest temperature buffering via enhancing canopy density and structural diversity. Ecology Letters, 28(3), e70096. doi:10.1111/ele.70096
Hethcoat, M. G., King, B. J., Fernandez Castiblanco, F., Ortiz-Sepúlveda, C. M., Prada Achiardi, F. C., Edwards, F. A., … Edwards, D. P. (2019). The impact of secondary forest regeneration on ground-dwelling ant communities in the tropical Andes. Oecologia, 191(2), 475-482. doi:10.1007/s00442-019-04497-8
González-González, A., Villegas, J. C., Clerici, N., & Salazar, J. F. (2021). Spatial-temporal dynamics of deforestation and its drivers indicate need for locally-adapted environmental governance in Colombia. Ecological Indicators, 126, 107695. doi:10.1016/j.ecolind.2021.107695
Rubiano, K., Figueroa, D. C., Guatibonza, N. B., & Clerici, N. (2026). The future of Colombian Andean forests under different deforestation scenarios. Ecological Indicators, 183, 114605. doi:10.1016/j.ecolind.2026.114605
FEEDBACK: Use this form to send a message to the author of this post. If you want to post a public comment, you can do that at the bottom of the page.