Clouds cost solar farms money, and the industry treats that the way it treats sunset. The forecast says overcast, the plant makes a fraction of its clear-sky output, and the financing model already assumed it would.<\/p>\n
Meteoric Technologies showed up on August 21 with a different plan. The company wants to send fleets of autonomous drones up into the cloud deck above a solar farm, mechanically change the water droplets that make that deck reflective, and hand some of the lost sunlight back to the panels underneath.<\/p>\n
No silver iodide. No chemicals at all.<\/p>\n
Two Cambridge-trained engineers run it out of San Francisco, Y Combinator took it into the Summer 2026 batch, and the entire company is those two people. The pitch is enormous and the published evidence fits in a paragraph, so let’s go through both.<\/p>\n
In its Y Combinator launch post<\/a>, the company says low and mid-level overcast cuts incoming sunlight by 73 to 82 percent while it sits overhead, citing a 1980 paper in the journal Solar Energy<\/em>. CEO Mete Karslioglu says the drones win part of that back. Meteoric’s own cloud-loss model puts the recovery at 10 to 30 percent of annual generation across major US grid regions, which the company values at $5,000 to $28,000 per megawatt.<\/p>\n A prototype dissipated an artificial cloud by 13 percent in cloud-chamber tests. That is the whole traction section.<\/p>\n TechSpot reported<\/a> that the company has not published the chamber size or the measurement method, and Meteoric has not described how the drones alter the droplets in the first place. No aircraft has flown over an operating solar plant. Meteoric itself calls the 10 to 30 percent range a modeled estimate rather than a demonstrated result, which is the honest way to label it, and it is still the number in every headline the launch produced.<\/p>\n Cost is the one piece of the pitch resting on somebody else’s arithmetic. Meteoric puts electric drone flight at $30 to $60 an hour against the $2,000-plus an hour it cites for crewed cloud-seeding aircraft, and crewed seeding really is that expensive, because you are paying for a turbine engine, a pilot, and an insurance policy that covers flying a human being into weather on purpose. Drones remove all three line items. That part of the argument holds up before anybody measures a single droplet.<\/p>\n Karslioglu’s research at the Cambridge Centre for Climate Repair, funded by ARIA, the UK’s Advanced Research and Invention Agency, was on making clouds more<\/em> reflective using seawater spray nozzles. That work is marine cloud brightening, a solar geoengineering technique built to bounce sunlight back into space and cool the planet down.<\/p>\n He is now selling the same physics pointed the other way, at a smaller scale, to power plants.<\/p>\n Co-founder and CTO Eric Nilsson came out of microfluidic chips that control water and droplet-removal hardware for LiDAR lenses. Between the two of them, that is a decade of hands-on work on the exact question of how to make a water droplet do what you want.<\/p>\n Which cuts both ways. Droplet control in a lab is the credential that makes the mechanism plausible, and it is also the credential that makes the leap obvious, because a nozzle on a bench and a stratus deck over 2,000 acres of west Texas are not the same object.<\/p>\n That 73 to 82 percent figure is real, and it is old. It measures how much sunlight a site gives up when a low overcast deck moves over it, which is a settled piece of solar engineering. It says nothing about how much of that loss anybody can claw back, and clawing it back is the entire business.<\/p>\n The World Meteorological Organization’s standing statement<\/a> on weather modification gets close to Meteoric’s approach in one place. Fog, the WMO says, can in principle be dispersed by enough heating or mechanical mixing, though those methods are usually impractical and expensive. Mechanical dispersal of a cloud is a recognized idea. It has just never been the cheap one.<\/p>\n Fog is a cloud sitting on the ground. It stays put. A stratus deck at 5,000 feet moves with the wind, regenerates from the moisture underneath it, and covers hundreds of square miles at a time, so a drone fleet that opens a hole in it has to keep opening that hole for as long as the sun is up. The WMO also notes that dispersing supercooled fog with glaciogenic material is well established, while warm cloud stays the hard case. Meteoric’s target is exactly the warm case.<\/p>\n Meteoric wants to work between 3,300 and 16,400 feet above ground, and FAA Part 107 caps small drones at 400 feet. Flying beyond visual line of sight still takes a visual observer or a case-by-case waiver, six years into the data-gathering meant to replace that system.<\/p>\n The FAA proposed Part 108, the rule that would set common BVLOS standards, on August 5, 2025. The agency reopened the comment period in January 2026, then missed its own June 2026 deadline for a final rule, and the Department of Transportation said this month that the delay is deliberate.<\/p>\n On August 27 the DOT announced a Phase 2 expansion of the FAA’s BEYOND program<\/a>, adding up to eight new lead participants specifically to test larger drones flying above 400 feet.<\/p>\n Rainmaker Technology is the closest thing to a precedent, and it is not an encouraging one. The El Segundo company flies a 50-pound quadcopter with a 15,000-foot ceiling, carrying silver iodide into supercooled clouds over the mountain west. It petitioned the FAA in July 2025 for relief from the hazardous-materials rule, and DroneXL reported on August 24 that no grant or denial had appeared in the public record 13 months later. The Air Line Pilots Association filed against that petition on safety grounds.<\/p>\n Meteoric carries nothing, which deletes the hazmat fight. The airspace fight is untouched.<\/p>\n Tennessee banned weather modification in 2024. Florida followed with SB 56<\/a>, effective July 1, 2025, which turns unauthorized weather modification into a third-degree felony carrying up to five years in prison and a fine of up to $100,000. The Florida statute reaches acts intended to affect the temperature, the weather, or the intensity of sunlight.<\/p>\n Meteoric’s product description is a system intended to affect the intensity of sunlight reaching the ground.<\/p>\n No party has alleged that Meteoric is in violation of any law, and the company has not said it plans to fly in Florida. Its launch post asks to hear from solar operators in Texas, the northern US, the UK and northern Europe, and from hail-suppression programs in North Dakota, Texas and Alberta.<\/p>\n Montana went a third way with SB 473, banning geoengineering while explicitly carving out cloud seeding, which is the kind of drafting distinction that decides whether a company like this has a market in a given state.<\/p>\n Solar developers are entertaining ideas like this because the alternative is the interconnection queue. Berkeley Lab’s Queued Up 2026 report<\/a> counted 773 gigawatts of solar sitting in US interconnection queues at the end of 2025, down 19 percent year over year, with the median wait from interconnection request to commercial operation running past five years for projects finished in 2025.<\/p>\n Squeezing more out of a plant that already exists skips that line entirely, which is the same logic behind the perovskite front glass three manufacturers just signed 13 gigawatts of<\/a>.<\/p>\n Karslioglu wrote in the launch post that the company’s “ultimate goal is to reduce the intensity of severe storms and hurricanes.”<\/em> Meteoric targets its first storm operation for late 2028, one year after the first large-scale cloud-clearing flight.<\/p>\n The WMO’s answer to that is one sentence long: no generally accepted evidence suggests that tropical cyclones can be modified at all. The organization goes further on the general case, saying weather systems carry so much energy that eliminating severe weather is impossible, and that technologies claiming large-scale or dramatic effects should be treated with suspicion.<\/p>\n That is a UN agency writing about the category, not about Meteoric, and the company has not claimed to have solved it. It has claimed it intends to try.<\/p>\n Between those two positions sits the thing solar operators actually have to price, which is weather they cannot control. Two Minnesota arrays swapped gravel for prairie and watched native bee populations climb twentyfold in five years<\/a>.<\/p>\n Texas farms now tilt their panels to 77 degrees when hail crosses a 30-mile line, and insurers price the ones that duck differently from the ones that don’t<\/a>. German scientists spent this spring hauling LiDAR into the UAE to test whether a big enough solar farm can brew its own rain clouds<\/a>. Meteoric is the first outfit proposing to go the other direction and take the cloud away.<\/p>\n Meteoric has put a date on the only test that settles any of this. The company says its first large-scale cloud-clearing flight happens in 2027, at a solar farm it has not named, with a drone it has not shown. Until somebody outside the company measures what comes out of a real overcast deck, the number on the record is 13 percent, in a chamber, posted August 21, 2026.<\/p>\n","protected":false},"excerpt":{"rendered":" Clouds cost solar farms money, and the industry treats that the way it treats sunset. The forecast says overcast, the … <\/p>\nThe CEO spent his master’s degree making clouds brighter<\/h2>\n
Cloud physics knows the loss number and not the recovery number<\/h2>\n
The FAA caps small drones at 400 feet and the rule that would change that is late<\/h2>\n
Two states have already made this kind of thing a felony<\/h2>\n
The hurricane goal has a direct answer from the WMO<\/h2>\n