In December 2022, a small startup in Mexico called Make Sunsets quietly released two helium balloons into the stratosphere. Each balloon carried a few grams of sulphur dioxide. When the balloons burst, the gas reacted with stratospheric moisture to form tiny reflective aerosols — the same chemistry that cools the planet for a year or two after a major volcanic eruption. The amount of cooling Make Sunsets achieved was infinitesimal. The amount of controversy it caused was not. The Mexican government banned solar geoengineering within its borders within a month. The US National Oceanic and Atmospheric Administration issued a statement. The startup's founder, Luke Iseman, became one of the most-discussed and most-criticised figures in climate technology. The episode crystallised a long-running question: who is allowed to decide whether humans modify the weather?
The Two Very Different Things People Call "Geoengineering"
The first is weather modification: techniques like cloud seeding that nudge atmospheric processes on a scale of hours to days, over areas of kilometres to a country. The second is solar geoengineering: large-scale, intentional intervention in the planet's radiation balance to cool the climate. They use different methods, operate at different scales, raise different governance issues, and are routinely conflated in journalism. Understanding the distinction matters.
Cloud seeding is old. Vincent Schaefer flew over a Massachusetts cloud in 1946 carrying dry-ice pellets, dropped them, and watched the cloud convert to snow. Bernard Vonnegut, in the same year, discovered that silver iodide crystals — chemical formula similar to ice — could nucleate ice crystals in supercooled water. Both techniques have been used continuously since the 1950s. The Soviet Union ran a national programme. The United States used cloud seeding to suppress lightning over the Apollo launch pads. India, China, and most ski resorts in the Alps and Rockies have used variations of the technique at some point. The effect is real but modest: cloud seeding can increase precipitation in a target area by 5-15% on average, and only under specific atmospheric conditions.
What China Is Actually Doing
China runs the largest weather modification programme in the world. The official budget is around 168 million dollars per year as of recent reporting; the actual operational scale, including provincial-level deployments, is probably larger. The programme covers an area roughly the size of India and is used for drought relief, smog dispersal before major events (the 2008 Olympics opening ceremony being the famous example), and hail suppression over farming regions. Chinese authorities have stated the goal of expanding the programme to cover 5.5 million square kilometres by 2025 — over half the country's land area. The international response has been muted because cloud seeding stays within national borders by definition, and because the techniques are decades old and widely used.
Solar Geoengineering Is the Different Conversation
Solar radiation management — the umbrella term for techniques that reflect sunlight away from Earth — is a different category. The leading proposal is stratospheric aerosol injection: releasing reflective sulphate particles into the upper atmosphere, where they remain for 1-2 years and reflect a portion of incoming sunlight. The physics is well established. Major volcanic eruptions — Pinatubo 1991, Tambora 1815 — have done exactly this naturally, and the global cooling they caused was measurable for years afterward. Pinatubo cooled the planet about 0.5°C for two years.
To replicate this deliberately at a climate-relevant scale would require injecting roughly 1-5 million tonnes of sulphur per year, year after year, indefinitely. The cost is plausibly $2-20 billion per year — a rounding error in the global energy budget. The technique would work. It would also have side effects: regional rainfall pattern changes, potential ozone-layer disruption, and the absolute requirement to keep doing it forever, because if you stop, decades of avoided warming resume almost overnight.
The Side Effects That Concern Climate Scientists
Solar geoengineering is sometimes presented as a free option — cheap, effective, available. The scientific consensus is more cautious. Three concerns recur in the technical literature:
Termination shock. If solar geoengineering is deployed for, say, fifty years to mask emissions warming, and then halted for any reason — war, political change, economic collapse — the accumulated underlying warming reasserts itself within months. The temperature jump would be 10-50 times faster than current warming, with catastrophic effects on ecosystems and agriculture that cannot adapt at that pace. The technique creates a dependency that subsequent generations cannot easily exit.
Regional asymmetry. Sulphate aerosol distribution affects rainfall patterns. Modelling studies consistently show that solar geoengineering would shift monsoons, reduce Sahel and Indian subcontinent rainfall, and create localised effects that vary by latitude and season. Some regions would benefit; others would suffer. There is no globally neutral implementation, and the regions most affected by warming today (low-latitude, low-income) are often the same regions most affected by intervention.
Ozone interaction. Stratospheric sulphate provides reactive surface area for chlorine and bromine chemistry — the same chemistry responsible for the Antarctic ozone hole. The interaction is well documented from volcanic eruptions and modelled extensively. Solar geoengineering at climate-relevant scale would slow ozone-layer recovery by 30-70 years, which has its own health and ecological consequences.
The Governance Vacuum
No binding international treaty currently governs solar geoengineering at scale. The 1977 ENMOD convention prohibits hostile use of environmental modification techniques in warfare but says nothing about peaceful unilateral deployment. The Convention on Biological Diversity adopted a non-binding moratorium on geoengineering field tests in 2010; it has limited enforcement. The Montreal Protocol governs ozone-depleting substances but doesn't cover sulphate aerosols.
In practice, a single nation — or even a single wealthy individual — could deploy stratospheric aerosol injection at climate-relevant scale, and the international community has no formal mechanism to stop them. The Make Sunsets episode involved a few grams. A serious deployment would involve thousands of tonnes per year. Nobody in the climate-policy community is comfortable with this gap.
Who Is Actually Researching It
| Programme | Country | Status |
|---|---|---|
| SCoPEx (Harvard) | USA | Stratospheric balloon experiment; cancelled 2024 after community opposition |
| GeoMIP | International model comparison | Ongoing climate-model intercomparison for geoengineering scenarios |
| SilverLining (NGO) | USA | Funding solar geoengineering research and governance work |
| Make Sunsets (commercial) | Mexico-based | Rogue commercial deployment; banned by Mexican government |
| UK ARIA programme | United Kingdom | £50M research budget announced 2024 for climate intervention research |
| China research network | China | Active scientific publication; no announced field deployment |
The Reasonable Position
The reasonable position on solar geoengineering is held by a wide cross-section of climate scientists and policy experts and amounts to this: the physics works, the technique should be researched seriously, the governance framework is woefully inadequate, and unilateral deployment is genuinely dangerous. Research can and should proceed in transparent international consortia. Deployment should require explicit multilateral consent, comparable to nuclear weapons agreements. The technology is more likely to be developed than not; the question is whether it is developed responsibly or not.
What This Means for the Climate Story
Geoengineering is increasingly likely to feature in 21st-century climate policy, but not as the primary tool. Decarbonisation remains the only intervention that addresses the underlying problem rather than masking it. Solar geoengineering, if used at all, will function as a temporary buffer to reduce extreme tail risks while emissions cuts take effect. Weather modification at the cloud-seeding scale will continue and probably expand, but its impact remains regional and modest. The conversation that needs to happen — about authority, consent, and the rights of the global majority that doesn't get a vote on whether to dim the sun — has barely started. The Make Sunsets episode forced it onto the agenda, and the next iteration is unlikely to be so easily contained.
Sources and Further Reading
National Academies of Sciences, Engineering, and Medicine (2021). "Reflecting Sunlight: Recommendations for Solar Geoengineering Research and Research Governance." National Academies Press.
Robock, A. (2008). "20 reasons why geoengineering may be a bad idea." Bulletin of the Atomic Scientists, 64(2), 14-18.
Keith, D. W. (2013). A Case for Climate Engineering. MIT Press.
Kravitz, B., MacMartin, D. G. (2020). "Uncertainty and the basis for confidence in solar geoengineering research." Nature Reviews Earth & Environment, 1, 64-75.
IPCC AR6 Working Group III (2022). Chapter 14 on cross-sectoral mitigation and solar radiation modification.
Make Sunsets (2024). Company website and public statements; Mexican government regulatory actions on geoengineering.
