Science1 publisher3 min readPublished
Measuring the hottest years moves the Amazon's warming hotspot to the central north
A Lancaster-led study of 1981 to 2023 finds extreme dry-season heat rising at least 0.75 C a decade across a tenth of the Amazon, with the fastest growth in the central north, a region that average-temperature trends had ranked behind the south.
The Scientist · Science desk
What happened
- Researchers at Lancaster University, working with WWF UK and an international team of more than 50 scientists, published a biome-wide assessment of Amazonian climate extremes in Communications Earth & Environment.
- Average Amazon temperatures are rising at 0.21 C a decade, in line with average global warming, and the study says the extreme trends do not appear in those average assessments.
- The central north, an area of high forest cover, natural savannas and vast Indigenous territories that had not been considered the most affected part of the basin, shows the most rapid growth in extremes.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- decision Anyone ranking Amazonian sites for protection, restoration money or fire preparedness now has two conflicting orderings, and the choice of statistic decides which region is first in line.
- constraint Extreme-event planning needs its own trend. A programme that wants to plan for the worst years has to recompute it from the exceptional years, not scale up the average.
- exposure Communities and forests in the central north sit inside the fastest-rising extremes while the average-warming picture placed the pressure elsewhere, so their planning was built from a ranking that put other regions first.
- cost The costs of ranking on the wrong statistic are fire, animal and tree deaths, and heat and smoke exposure, and they fall on the central north, which the average ranking placed lower.
Amazonia straddles the equator, and the dry season north of it does not fall in the same months as the dry season to the south. That mismatch is the main reason the central north stayed out of earlier assessments. Jos Barlow, a professor at Lancaster University, is the study's lead author. "While this has been assessed in the Southern Amazon, which shares a largely similar dry season, it has never been assessed across the whole Amazon," he said [10].
So the team defined the dry season place by place. They divided the biome into 11-kilometre cells and used temperature and rainfall data combining satellite records with local weather stations to identify the dry season for each area [6]. Then they modelled temperature and a water deficit measure across 1981 to 2023. That measure accounts for the effect of temperature on water loss, so a hot year registers as drier than rainfall alone would make it look [7].
They computed the trend twice. Central tendency weights average rates across all years, which is how Amazonian warming has usually been measured; extreme tendency weights the most exceptional years [8]. "We are most interested in things when they are at their hottest and driest, as that is when the most harm can be caused by high temperatures," Barlow said [9].
The two answers are far apart. Across a tenth of the basin, more than 700,000 square kilometres of a biome of roughly 7 million, extreme dry-season temperature has risen by at least 0.75 C a decade and by more than 3.22 C since 1981 [3][4]. The basin's average temperature is rising at 0.21 C a decade, in line with average global warming [5]. That is a factor of about 3.6 [1], and over the 43-year record it works out to roughly 3.2 C against roughly 0.9 C [2][3]. The two figures are not like for like: 0.21 is an average across the whole basin and all years, while 3.22 is the extreme-year trend in the worst-affected tenth [3][5].
Which region comes first depends on which measure you use. On average temperatures, the Southern Amazon, which has experienced significant deforestation and land-use change, is confirmed as the fastest-warming region [11][12]. On extremes, the central north leads [13]. "The rates of change in climate extremes are much higher than the rates of change in average climate in the Amazon," said Nathalia Carvalho, a postdoctoral research associate at Lancaster Environment Centre, "and the most affected regions are also different for extremes and average climate" [16][22].
Read that way, a map that ranks Amazonian places by average warming is sorting them on the wrong statistic when it is used to plan for fire seasons and drought years. The harm arrives in the exceptional years. The study supports that much, and no more. How much of the central north's extreme trend comes from land-use change and how much from global warming is left open, as is the area of Indigenous territory inside the 700,000 square kilometres.
The next test is close. The study's announcement says the El Nino of 2026 is expected to be the largest in living memory, and that the Amazon might suffer extremely high temperatures and drought just two years after the major drought of 2024 [18][19].
What to watch
- Whether the paper's supplementary data give the area of Indigenous territory and protected land inside the 700,000 square kilometre hotspot.
- Whether fire and tree-mortality damage in a 2026 El Nino concentrates in the central north cells the extreme-tendency analysis flags.
- Whether follow-up work separates land-use change from global forcing in the central north's extreme trend.