Efforts to cool the planet may do more than lower temperatures, they could fundamentally reshape Earth’s climate systems.
As carbon dioxide emissions continue to rise, more policymakers, scientists, and environmental advocates are exploring geoengineering as a potential way to limit climate change.
These large-scale interventions could affect everything from rainfall patterns to global food production, making the risks and consequences significant. In theory, adjusting parts of Earth’s climate system could help offset some warming. However, the planet’s complexity makes predicting the outcomes of these efforts extremely challenging.
Researchers at the University of California, Santa Barbara examined two methods designed to reduce the amount of sunlight reaching Earth’s surface: cloud seeding over the eastern Pacific Ocean and the release of aerosols into the stratosphere. Using models of ocean and atmospheric interactions, they found that the first approach could severely disrupt a major climate system, the El Niño Southern Oscillation, while the second had little measurable effect. Their findings, published in the journal Earth’s Future, highlight the need to carefully assess the broader impacts of geoengineering strategies.
“We need to be careful about implementing geoengineering proposals before we fully understand what’s going to happen,” said first author Chen Xing, a doctoral student at UCSB’s Bren School of Environmental Science & Management.
Xing and fellow graduate student Cali Pfleger set out to explore how these interventions might affect marine ecosystems. To do that, they focused on how geoengineering could alter large-scale ocean cycles, especially the El Niño Southern Oscillation (ENSO).
ENSO is a recurring climate pattern that develops every 2 to 7 years and shifts the distribution of warm water across the tropical Pacific Ocean. These shifts influence weather worldwide. During El Niño periods, warmer waters move toward the western coasts of the Americas near the equator, often bringing wetter winters to California. During La Niña, warmer conditions in the western Pacific strengthen monsoon systems in South and Southeast Asia.
Comparing Geoengineering Strategies
Both strategies analyzed in the study rely on aerosols, but they differ in composition and altitude. Marine cloud brightening (MCB), a form of cloud seeding, involves spraying sea salt particles into the lower atmosphere within about 2 kilometers (about 1.2 miles) of the ocean surface to increase cloud reflectivity. Stratospheric aerosol injection (SAI), by contrast, releases sulfate particles much higher in the atmosphere to reflect incoming sunlight.
Some geoengineering proposals focus on the eastern edges of ocean basins, near the west coasts of continents, because of their strong influence on global temperatures. However, the southeastern Pacific appears to play a key role in shaping ENSO behavior. “Deploying MCB in the subtropical eastern Pacific dramatically reduces ENSO amplitude by approximately 61%,” the authors write.
“It’s hard to get ENSO to change by that much that quickly,” said Associate Professor Samantha Stevenson, a co-author of the study and advisor to Xing and Pfleger.
Marine cloud brightening increases the number of droplets in clouds while reducing their size, making the clouds more reflective and cooling the surface below. At the same time, smaller droplets reduce the likelihood of rainfall, leading to drier conditions.
As this cooler, drier air spreads into the central Pacific, it lowers evaporation and weakens atmospheric convection. This chain reaction intensifies trade winds along the equator, increases ocean upwelling, and further cools sea surface temperatures. Together, these changes disrupt the balance that drives ENSO. In practical terms, the cycle weakens dramatically.
The scale of the effect surprised the researchers. “The authors thought the proposals could have impacts, ‘but we didn’t expect two-thirds of ENSO’s variance to disappear,’ Xing said. The implications seem clear: ‘Don’t do MCB over the eastern Pacific Ocean because it might cause super strong chain reactions from ENSO’s disappearance.’”
Why Stratospheric Aerosols Behave Differently
In contrast, stratospheric aerosol injection produced little to no change in ENSO behavior. This difference likely comes from how and where the particles spread in the atmosphere. MCB operates close to the surface and remains concentrated in a specific region, while SAI occurs high above Earth, where particles disperse more evenly.
Because of this broader distribution, SAI creates a more uniform cooling effect and is less likely to disrupt regional climate systems like ENSO. Still, the findings do not mean that all forms of MCB would have the same consequences.
According to Stevenson, the strong effects seen in the study are tied to the specific conditions of the eastern Pacific. “We’re not saying that all MCB is going to kill ENSO. We’re just saying that this happens if you do it in this specific region,” she said. Other locations might produce different outcomes, although achieving the same cooling effect elsewhere could require larger interventions.
Choosing not to act also carries serious risks. Continued climate change is expected to disrupt ecosystems, weather patterns, and human systems worldwide. Scientists still do not know how ENSO will respond to ongoing warming, which adds another layer of uncertainty. “There’s nothing that compares to the speed with which ENSO would change in these MCB experiments,” Stevenson said. “It just does not naturally drop 60% in 10 years, even under climate change.”
Reducing sunlight at a global scale would also affect photosynthesis, lowering productivity in crops, forests, and marine life. Marine algae, which form the base of the ocean food web, are especially important because they produce about 70% of the oxygen in Earth’s atmosphere. The research team plans to further study how these interventions could affect marine ecosystems.
The study highlights the complexity of geoengineering decisions. Different methods may achieve similar global temperature goals but produce very different regional effects. “Two interventions can get to the same warming target globally and have extremely different regional climate impacts,” Stevenson said. “The most important question is, ‘Are we thinking of all of the potential consequences?’”
Reference: “Subtropical Marine Cloud Brightening Suppresses the El Niño–Southern Oscillation” by C. Xing, S. Stevenson, J. Fasullo, C. Harrison, C. Chen, J. Wan, J. Coupe and C. Pfleger, 4 August 2025, Earth’s Future.
DOI: 10.1029/2025EF006522
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.
9 Comments
There’s a better way – A gigantic net that will be deployed at 5 million miles from the Earth, to block the Sun’s light.
Assuming, for the sake of the argument, that you can produce facts to support your currently unsupported claim that you have a better solution, note that your terse claim also does not address the author’s remark: “Reducing sunlight at a global scale would also affect photosynthesis, lowering productivity in crops, forests, and marine life. Marine algae, which form the base of the ocean food web, are especially important because they produce about 70% of the oxygen in Earth’s atmosphere.” What is the point in reducing a slight warming if it results in less food and oxygen?
What’s the projected cost of such a net in today’s dollars? What is the cost to the environment also of the rocket fuel that must be used to build and maintain such a net? When the net is inevitably damaged by asteroids or meteors, what is the cost estimate for fixing the results of those events?
These are dangerous and reckless experiments and should be stopped immediately.
Unfortunately, there are few actual experiments being done. These predictions are based on imperfect models. At this point in time, what the emphasis should probably be is on gathering empirical data, to make robust models that can be trusted. Making predictions based on guesses and unstated assumptions is not a good way to learn what reality is.
When used properly, models should provide insight on how complicated dynamic systems work. Once they are understood well enough to make accurate predictions about the dependent variables that are constantly changing, THEN one can start using them to play ‘what if?’ games using intentional forcings. However, they still need empirical testing to be sure that some ‘unknown unknown’ hasn’t broken the model.
“…, the planet’s complexity makes predicting the outcomes of these efforts extremely challenging.”
The authors properly remark about the difficulties in climate predictions. However, they barely note that they are relying on models, without indicating how they are validating the models. They claim that aerosol injection into the stratosphere “had little measurable effect.” Yet, it is well demonstrated that there is a correlation between volcanic sulfates episodically injected into the stratosphere and temporary global cooling. That is why aerosol injection is frequently suggested as a potential anthropogenic cooling strategy!
Before we can put much stock in their claim(s), there needs to be a more rigorous validation of the models.
“As this cooler, drier air spreads into the central Pacific, it lowers evaporation and weakens atmospheric convection.”
What cooler, drier air? The transition to this assertion is rather abrupt and probably needs more explanation. I suspect that this explanation is too simplistic. As air cools, it is incapable of holding as much water vapor as the formerly warmer air. (The unstated assumption is that the air is always saturated.) However, the process of evaporation, aided by lateral winds, restricts the evaporation to a very thin surface layer. It is that same surface layer that is warmed by strongly-absorbed IR and UV electromagnetic emissions from the sun, resulting in a strong temperature gradient near the water/air boundary. I think that it is more likely for a cooling lower-troposphere to result in more rapid condensation after evaporation, exhibited as fog. Not unlike the fog banks rolling in off the Pacific Ocean at the Golden Gate, and over the Santa Cruz Mountains. Near the Andes Mountains, I would expect strong onshore breezes as the land heats up and the air rises.
I strongly agree with the remark, “The most important question is, ‘Are we thinking of all of the potential consequences?’” This gets back to my initial complaint that the models are only as good as the guesses and unstated assumptions, which are very difficult to quantitatively assess unless there is a rigorous process for validating the models.
Democracy can not deal with the logical choices that can make a difference. We all try to follow the hipocratic oath. DO NO HARM. So we do nothing the worst possible action. We are also fearful of nuclear power the cleanest safest form of energy. The cheapest forms of geoengineering all have some risk. So we do nothing until too late.
[email protected]
There is a far less risky way to limit climate change than geoengineering: Reduce the burning of fossil fuels.