Climate change is causing rivers worldwide to lose oxygen, and tropical waterways are being hit hardest.
Climate change is causing rivers around the world to lose oxygen at a troubling rate, according to a study published May 15 in Science Advances. Researchers found that this widespread and persistent deoxygenation is affecting river ecosystems globally, with tropical rivers emerging as the most at risk. The findings highlight an urgent need for strategies to slow oxygen loss and protect freshwater ecosystems.
The study was led by Prof. Kun Shi of the Nanjing Institute of Geography and Limnology (NIGLAS) at the Chinese Academy of Sciences. Dr. Qi Guan was the study’s first author, and the work also involved collaboration with a researcher from Tongji University.
Global Rivers Are Losing Oxygen
Oxygen is essential for healthy rivers. It supports fish and other aquatic organisms, helps maintain biodiversity, and plays a major role in regulating biogeochemical cycles within freshwater ecosystems. When oxygen levels decline, river health and ecosystem stability can suffer.
To examine long-term changes in dissolved oxygen, the researchers used a machine-learning stacking algorithm to study data from 21,439 river reaches worldwide collected over nearly four decades (1985–2023).
Their analysis showed that river oxygen levels are declining at a rate of -0.045 mg L-1 decade-1. Overall, 78.8% of the rivers included in the study experienced deoxygenation.
Tropical Rivers Face the Greatest Risk
The strongest oxygen declines were observed in tropical rivers located between 20°S and 20°N, including rivers in India. Scientists had previously expected rivers in higher latitudes, where warming is often more intense, to be the main hotspots for deoxygenation.
Instead, the study found that tropical rivers are especially vulnerable because they already tend to have lower oxygen concentrations. Faster oxygen decline in these waterways increases the risk of hypoxia events, when oxygen levels become dangerously low for aquatic life.
Flow Conditions and Dams Affect Oxygen Levels
The researchers also investigated how river flow patterns and dam impoundment influence oxygen loss.
Both low-flow and high-flow conditions appeared to partially reduce deoxygenation compared with normal river flow. Low-flow conditions were associated with an 18.6% lower deoxygenation rate, while high-flow conditions resulted in a 7.0% lower rate compared with normal-flow conditions.
Dam impoundment produced different outcomes depending on reservoir depth. In shallow reservoirs, impoundment accelerated deoxygenation. In deeper reservoirs, however, impoundment helped reduce oxygen loss within the reservoir area.
Heatwaves Intensify River Deoxygenation
Further analysis showed that climate-driven reductions in oxygen solubility were the primary cause of river deoxygenation, accounting for 62.7% of the observed decline in oxygen.
Ecosystem metabolism, influenced by factors such as temperature, light, and water flow, contributed 12% of the deoxygenation.
The team also analyzed the impact of heatwaves on river oxygen levels. The researchers found that heatwaves accounted for 22.7% of global river deoxygenation and increased the deoxygenation rate by 0.01 mg L-1 decade-1 compared with average climatological conditions.
The findings highlight the growing effects of climate warming on lotic ecosystems, which are ecosystems associated with flowing freshwater. The researchers identified tropical rivers as the ecosystems most urgently in need of mitigation efforts to address worsening oxygen loss. They also said the study provides an important scientific foundation for policymakers developing strategies to combat river deoxygenation worldwide.
Reference: “Sustained deoxygenation in global flowing waters under climate warming” by Qi Guan, Kun Shi and Xuehui Pi, 15 May 2026, Science Advances.
DOI: 10.1126/sciadv.aef3132
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.
3 Comments
“Overall, 78.8% of the rivers included in the study experienced deoxygenation.”
The question that should be asked, but isn’t, is “If the world is warming, why aren’t
ALL the rivers showing de-oxygenation?” It amazes me how quick most researchers are to include ‘climate change’ as being responsible for problems without asking if it makes sense. It is commonly accepted that the global average change in terrestrial air temperature is less than 0.2 deg/decade, with most of that in the Arctic, at night, and in the Winter; that is, less than 1 deg C in the last 50-years. It is about half that rate for the oceans. Even before Trump started his second term, the EPA’s website posted a graph making the case that heat waves were stronger and more frequent during the 1930s than in recent years. My personal analysis suggests that there is no evidence that the USA is experiencing more or greater heatwaves: https://wattsupwiththat.com/2019/09/06/the-gestalt-of-heat-waves/ .
While Henry’s Law predicts lower solubility of gases in warm water than in cool water, I would suggest that reservoirs contribute to de-oxygenation by drowning pre-existing vegetation and subsequent biogenic oxidation of the organic material, reducing the rate of oxygenation by decreasing the ratio of surface area to volume, and reducing turbulence/mixing found in streams (particularly in mountains) compared with with essentially stagnant water bodies. If one were to run the numbers, I would expect that it wouldn’t be necessary to resort to implicating ‘climate change.’
I remember once as a teenager, going on vacation with my parents to Shasta Lake (Calif.) to go water skiing in July. One day I decided to dive below the surface, which was like bath water, and surprised to discover just how cold the water was about 2 meters below the surface. I immediately returned to the surface! Oxygen can’t dissolve significantly in that very warm water. Therefore, it can’t get past the thermocline barrier to re-oxygenate the deep water.
Any thoughts on ecological degradation outside of “climate change”?
Climate change? No, it’s not. Just shut up. This is another grant abuse article.