Cloud seeding: what the rain-makers can and cannot buy

Rain enhancement is the region’s longest-running water experiment. Morocco and the United States seeded clouds on 144 days between 1984 and 1989, in a project that gave Morocco its first weather radar.¹ The UAE flew 311 seeding missions in 2022, and Saudi Arabia’s programme had flown about 752 seeding flights by August 2025.⁵¹⁰ It is also the water technology with the widest gap between what is spent and what is proven.

What a flight does

An aircraft or ground generator releases particles into a cloud. In a cold cloud, silver iodide gives supercooled droplets a surface to freeze onto. In a warm cloud, hygroscopic salt flares grow droplets until they fall, and most UAE operations have used that method.¹³ The physics is real. The hard question is how much extra rain reaches the ground, because a seeded storm and an unseeded one cannot be told apart afterwards.

The one number science has produced

The cleanest measurement came from Idaho in 2017. Aircraft seeded winter clouds over the mountains while radar tracked the extra snowfall downwind of the flight lines. Friedrich and colleagues put the water generated on three occasions at 120,000, 240,000 and 340,000 cubic metres, and gauges under the seeded bands recorded 0.05 to 0.3 millimetres.¹¹ Together that is about 700,000 cubic metres, less than one day’s contracted output of Taweelah.¹⁹

It was also the easy case. The WMO’s 2025 statement finds an evidence-based causal relationship for wintertime seeding of mountain clouds. For warm-cloud and hygroscopic seeding it describes a weak signal against large natural variability, and warns that results from one trial cannot be applied to a different environment. The statement gives no effectiveness percentage at all.¹²

What the programmes claim

National figures run ahead of that standard. UAE meteorologists have said seeding could raise rainfall by about 30% in a clean atmosphere or 15% in a dusty one, while adding that it is complex and needs more research. The centre’s deputy director general said at least 20 years of data were needed for a realistic number.⁵ A 2021 evaluation co-authored by the centre’s staff reported a 23% average increase in annual rainfall over seeded areas. It compared storms after the fact rather than randomising which were seeded.⁸

Morocco’s meteorology directorate reports gains of 14 to 17% in rain and snow over targeted basins, also its own assessment.⁴ Independent reviews are more cautious. The US Government Accountability Office found published estimates ranging from 0 to 20%, and noted that seeding works only when suitable clouds are already present.¹⁴ A randomised experiment in Australia’s Snowy Mountains, run from 2005 to 2009, found a positive but not statistically significant effect on snowfall.¹³

The research bet

The UAE has put money into closing the gap. Its Research Program for Rain Enhancement Science, run by the National Center of Meteorology since 2015, has paid up to US$1.5 million per winning proposal over three years.⁶ The seventh cycle has a total budget of US$3 million across up to five awards, with the solicitation due later in 2026.⁷ That is serious research money and a small line in a national water budget. The WMO’s bar for proof is a randomised design, in which only some suitable clouds are treated.¹²

The April 2024 lesson

When extreme rain flooded Dubai on 16 April 2024, attention turned at once to cloud seeding.¹⁶ The National Center of Meteorology stated that it did not conduct any seeding operations during the event, and that it does not seed during extreme weather.¹⁵ World Weather Attribution concluded that seeding had no significant influence, because the storm was driven by atmospheric moisture that seeding cannot add.¹⁷ In October 2025 Bloomberg reported flight-tracking data showing a seeding aircraft flying patterns typical of seeding on 15 April. The data cannot show whether flares were fired.¹⁸

The science on the storm is settled; the public argument is not. A programme that flies over a country owns the weather in the public mind. Published flight logs, independent evaluation and a clear operating rule for severe weather are the only defence, and they have to exist before the storm.

Where the market is

Aircraft operations, flare chemistry and weather radar are the visible procurement. The growing line is verification: randomised trial design, radar-based quantification and public reporting. Seeding is defensible as insurance over a dam catchment, as long as it is described as a bet and kept out of the supply column of a national water balance.

In the region

The three national programmes are organised differently. Morocco’s Al Ghait became operational in 1996 and moved to year-round operations in 2024, with 20 fixed and three mobile ground generators.³ A national committee coordinated by the Gendarmerie royale steers it; the Royal Air Forces fly, and the meteorology directorate gives technical support.² Saudi Arabia’s programme began in 2022 and flew 415 seeding flights in 2023 across six regions.⁹ The UAE’s is the most research-intensive, anchored by a grant programme open to scientists worldwide.⁶

What to watch

• 2026The seventh-cycle solicitation of the UAE’s rain-enhancement research programme: US$3 million across up to five awards, under a new flexible funding structure.⁷

Further reading

• World Meteorological Organization, statement on weather modification (2025). The current scientific consensus, and the evidence bar a programme should meet. wmo.int
• Friedrich, K. et al. (2020), “Quantifying snowfall from orographic cloud seeding”, PNAS. The one clean measurement of seeded water. ncbi.nlm.nih.gov
• US Government Accountability Office (2024), Cloud seeding technology: assessing effectiveness and other challenges. A plain review of what is and is not known. gao.gov
• World Weather Attribution (2024), analysis of the April 2024 rainfall in the UAE and Oman. How the storm was built, and why seeding did not build it. worldweatherattribution.org