Amazon drought experiment finds biomass loss, then stabilisation
The story at a glance
A long-running experiment in Caxiuanã National Forest in Pará, Brazil, has exposed a one-hectare forest plot to reduced rainfall since 2002. A 2025 peer-reviewed study found that above-ground biomass in the treated plot fell substantially before stabilising. The results offer evidence about how this forest responded to prolonged experimental soil-moisture reduction, not a direct forecast for the Amazon as a whole. 12
The biomass loss was associated with high mortality among large trees. Stabilisation followed as more soil water became available per unit of remaining tree biomass, but the plot did not return to its earlier condition: the resulting forest had less biomass and was more open. A separate study at the site found a different response among sampled trees during the 2023 El Niño drought. 135
What you need to know
- Esecaflor diverts about half the rainfall from its one-hectare treatment plot and compares it with a nearby control plot. The treatment began in 2002. This design allows researchers to compare the plots under experimental soil-moisture reduction, but it does not reproduce every feature of a natural drought. 124
- The 2025 study estimated that above-ground biomass in the treatment plot declined by 34% between 2002 and 2016, then remained stable from 2017 to 2023. The decline was linked to disproportionately high mortality among large trees. These figures describe the experimental plot and the periods assessed, not biomass change across the Amazon. 13
- The researchers attribute the later stability partly to increased water availability per unit of remaining tree biomass after biomass had been lost. Stability therefore does not mean the original forest structure recovered: the plot had less biomass and was more open. The result describes a new, lower-biomass state under the experiment, rather than restoration. 13
- A separate 2026 study examined 40 trees at the same site during the 2023 El Niño drought. Trees exposed to long-term experimental drought maintained transpiration better and recovered more fully afterwards than sampled control trees. This is evidence about the sampled trees at this site, not a finding about all trees or forests across the Amazon. 5
- The main study reports measurements through 2023. Researchers removed the rain-diversion panels in November 2024, ending the rainfall-exclusion treatment, and are investigating whether large trees can return. The post-treatment research may help establish whether the plot's forest structure changes after the experimental drought ends. 124
Why this matters
The findings show why a stable biomass reading should not automatically be read as a forest recovering: in this plot, stability followed substantial biomass loss and reflected a more open, lower-biomass forest. They also illustrate the limits of applying experimental results to natural drought across the Amazon. The study manipulated soil moisture over one hectare, whereas natural drought can also involve atmospheric drying and coincide with other stresses. The results are relevant evidence, but they do not establish how other forests would respond. 1234
What happens next
With the rainfall-exclusion panels removed in November 2024, researchers are now investigating whether large trees can return to the plot. The main study's measurements run through 2023, so they do not establish what recovery, if any, follows the end of the treatment. Further findings from the post-treatment phase may clarify how the forest changes after the prolonged experimental drought. 124
What's still unclear
- Whether large trees and the plot's former biomass will return after rainfall exclusion ended remains unresolved. 2
Read the reporting
This WildNews briefing synthesises and summarises reporting from the sources below. WildNews did not produce the original reporting. Learn more about how we use AI
Was this story useful?
Your anonymous feedback helps us make future WildNews briefings more useful.