• Matthew Watson studies Natural Hazards at the University of Bristol.
Wimbledon, 2026. Bright blue skies and a wonderful late afternoon sun lights up the lush green grass of centre court. Out strides the British No 1 and four-time winner, Andy Henman, to the cheers of the excitable, partisan crowd.
Somewhere nearby, at the headquarters of WeatherMod Inc, technicians are busily checking data, confident their efforts have worked.
They have been in contact with two pilots who have just completed their spray sorties and are returning to land at Heathrow's new third runway. Thanks to the delivery of 4kg of, in its pure form, a yellowish powder known as Silver Iodide (AgI) into clouds upwind of London, it is now raining over the Salisbury Plain, 160km away, and the rain predicted for later in southwest London suburbs is now 92 per cent less likely.
This scenario probably sounds a little far-fetched - not least the bit about the repeatedly successful British tennis player.
However, weather modification occurs more often than most people are aware. For example, as I wrote that first paragraph I genuinely didn't realise that a Weather Modification Incorporated actually already exists in Fargo, North Dakota. They, and other companies like them, have sprung up over the past few years promising to manage water for crops, clear fog and even protect wedding days from ill-timed hail.
But two questions need further investigation to consider the likelihood of the above scenario at Wimbledon: can we do it (that is, does it work) and should we do it?
Neither, it turns out, is particularly easy to answer.
To make rain several processes need to occur. First, small particles known as cloud condensation nuclei (CCN) are required on to which water can condense. Then these droplets need to grow to a size where they precipitate out of the cloud, finally falling where and when required.
In our hypothetical scenario we would therefore need to be able to either control or at least predict accurately the concentration of CCN, the rate at which droplets form, and the evaporation rates within the clouds. We'd also need to have some handle on the rate and direction in which rain would fall.
In reality, cloud seeding with AgI - the current default option - really tackles only the first of these processes, forming the condensation nuclei. Even if clouds are seeded, it is still a matter of debate as to whether they actually create much additional rain. While companies claim success, some scientists are more wary.
Although other seeding agents (and methodologies) exist, it is worth noting that, in the case of AgI, the type of clouds the particles are injected into will govern the outcome.
Seeding works best in clouds which have a pre-existing mixture of water droplets and ice, as this type of nucleation requires ice-crystals to form. Following the production of CCN we'd then need to be able to predict, through computer modelling, how small droplets will form into rain and eventually fall.
One of the major drawbacks of cloud seeding is a lack of proof that it works: given weather forecasting remains imperfect, how would you know what would have happened without intervention? The second part of the question is arguably even harder to approach. What are the ethics of removing water from one part of the world, even on a small scale, and moving it somewhere else?
Is this "messing with nature" or "playing God"? Water is, after all, the most precious commodity on Earth.
Let's assume for now that it is possible to alter local weather patterns and to prevent or cause rain. This could be used for both good and evil, and the potential for abuse is worth considering.
While manipulating the weather as a weapon is now explicitly outlawed by United Nations treaty, there have been efforts to alter the outcome of conflict using cloud seeding.
Deliberate and accidental effects from commercial activity also seem possible. That dreamy, rain-free wedding ordered up by an anxious billionaire could easily ruin a school sports day in a nearby town.
The question of attribution is possibly the most challenging. Without any alternative outcomes to analyse, how can you really know what are the impacts from your actions. Some even say, quite incorrectly, that cloud seeding experiments caused floods, such as those that killed 35 people in the English village of Lynmouth in 1952.
Expert opinion leans strongly against that idea being correct. Nonetheless, conspiracy theories persist. If, in our hypothetical Wimbledon scenario, bits of Wiltshire flooded, who would foot the bill?
It's certainly possible in theory to prevent rain in one place by using cloud seeding to induce it in another, upwind. But there are huge challenges and the jury is still out about whether such efforts really work.