The following is an edited and revised version of the talk I gave at the Global Catastrophic Risks conference that was held in conjunction with Convergence 08 (and which I reprised for Convergence). I’m posting it here because it seems to me that this is exactly the kind of thing Foresight was founded for: to examine the revolutionary impacts of readily forseeable applications of nanotechnology.

In its present form, the Weather Machine is a work of futurism, not engineering. I have done only back-of-the-envelope calculations, and my assertions about what could be built are based more on instinct and educated guesses than on any major, deep engineering analysis. Even so, as a futurist I am fairly sure that something like the weather machine will be possible within the next few decades.

The Weather Machine, Mark I

Here’s the basic idea for the machine: construct a small aerostat—a hydrogen balloon—at a guess an optimal size is somewhere between a millimeter and a centimeter in diameter. It has a very thin shell of diamond, maybe just a nanometer thick. It’s round, and it has inside it (and possibly extending outside, giving it the shape of the planet Saturn) an equatorial plane that is a mirror. If you squashed it flat, you would have a disc only a few nanometers thick. Although you could build such a balloon out of materials that we build balloons of now, it would not be economical for our purposes. Given that we can build these aerostats so that the total amount of material in one is actually very, very small, we can inflate them with hydrogen in such a way that they will float at an altitude of twenty miles or so—well into the stratosphere and above the weather and the jet streams.

Each aerostat contains a mirror, and also a control unit consisting of a radio receiver, computer, and GPS receiver. It has just barely enough power and fans or other actuators to tilt itself to a preferred orientation. That’s all it does—listens for commands on the radio, and tilts to an angle that is a function of its latitude and longitude. It’s not really a complicated machine.

Now make enough of them to cover the entire globe. For the centimeter size, you’d need about five quintillion of them. This is why nanotechnology makes a big difference. If you tried to cover the earth with something the total thickness of even a current-day party balloon, let’s say about 100 microns, you need on the order of 100 billion tonnes of material, but with the nano-engineered design, just a few nanometers thick, you only need about ten million tonnes. To compare that with the scope of current-day construction, ten million tonnes is roughly the amount of material that is used to make a hundred miles of freeway. This is an amount of material that current-day technology, much less nanotech, can handle straightforwardly.

That’s the weather machine. We have these aerostats which float twenty miles up. They have GPS and controllers and can turn themselves. That’s all there is to it. What could you do with a machine like this? The machine is essentially a programmable greenhouse gas. If you set the mirrors facing the sun, it reflects all the sunlight back. If you set them sideways, it allows the sunlight to come through, and similarly for the longwave radiation coming from the back side of the earth at night.

For comparison, the radiative forcing associated with CO2 as a greenhouse gas, as generally mentioned in the theory of global warming, is on the order of one watt per square meter. The weather machine would allow direct control of a substantial fraction of the total insolation, on the order of a kilowatt per square meter—1000 times as much. It would completely trump any natural or anthropogenic climatic forcing, and allow us to set Earth’s climate to whatever we wanted it to be.

For mere overall climate control, we’d only need to build a few tenths of a percent of a full weather machine, and the controls on the individual aerostats can be very simple. When set to let sunlight through, for example, the mirrors can be several degrees away from edge-on to the sun, and the effect would be a scattering of light (visible perhaps as a slight haziness) but no significant reduction in total insolation.

Kardashev Type I civilization

A Kardashev Type I civilization is one that controls all the energy available on a single planet. A Weather Machine would do that—our total current energy (strictly speaking, power) use, at 15 terawatts, is a completely negligible fraction of the over 100 petawatts the machine would control.

If you read the IPCC numbers on how much global warming is going to cost, the actual estimate is that over the course of the century it is going to be about 3% of the global GDP—a huge sum of money. A Weather Machine could not only prevent that, but probably double the GDP simply by regional climate control. If you could actually control the climate and tailor it, you could make land in lots of places on the earth, such as Northern Canada and Russia, as valuable as California. The economic benefit would be enormous. There is a huge amount of value to being able to control the weather. This is something that people have always wanted to do and therefore, once it becomes possible, they probably will.

The better control we have over our aerostats, the more we can do. Controlling insolation on the scale of tens or hundreds of miles would probably give us the ability to affect daily weather patterns as well as climatic averages. Given really precise control, you could enhance solar power, for example. Build an array of photovoltaics in the desert and and then program about a thousand square kilometers of aerostats above it to focus the sunlight down onto your array. Instead of needing a thousand square kilometers of solar collectors, you need only one. The main reason solar power is expensive is that it’s diffuse, so the capital costs for collecting it are high. But with 1000x concentration it should be quite economical. The concentration is more or less free because you have already built the machine to control the weather. What’s more, you have not changed the energy balance any, because you are shading all the areas that are otherwise under the thousand square kilometers. That gives your concentrated collector, in broad daylight, an energy flux that is approximately the same as a thousand nuclear reactors of a typical size. Of course, you have to cool the collectors fairly vigorously because they are not 100% efficient.

In 2029 the asteroid Apophis is going to come within the orbit of the moon, passing the earth, in what will be something of a butterfly effect incident where it is fairly difficult to predict exactly what is going to happen to it after that close encounter. Apophis will return in 2036 and we can’t say for sure whether it will strike the Earth or not. In 2029, if a highly controllable Weather Machine had been built by then, as the asteroid comes tumbling past we could focus a few petawatts of sunlight on it and give it a kick. This is probably a lot more appropriate in the case of Apophis than many other asteroids because it is going to be so close. A small kick in 2029 will have a huge result in 2036, almost certainly enough to prevent a strike if that should actually be the trajectory it’s on.

The Weather Machine, Mark II

The Mark I Weather Machine is something like nanomechanical rod logic—an gedanken experiment existence proof that a given level of technology will have a given capability. We can go a bit farther and talk about what the capability might be like given closer control of light and matter, bearing in mind this is somewhat more speculative.

Take the same aerostat, but inside put an aerogel composed of electronically switchable optical-frequency antennas—these are beginning to be looked at in the labs now under the name of nantennas. We can now tune the aerostat to be an absorber or transmitter of radiation in any desired frequency, in any desired direction (and if we’re really good, with any desired phase). It’s all solid state, with no need to control the aerostat’s physical attitude. Once we have that, the Weather Machine essentially becomes an enormous directional video screen, or with phase control, hologram.

Astronomers hated Weather Machine Mark I, but they love Mark II because it turns the entire earth into a telescope with an aperture of 8,000 miles. Mark I could zap Apophis as it flew by inside the Moon’s orbit; Mark II could zap asteroids at much greater distances, or power laser-propulsion spacecraft.

Mark II, with the ability to shift frequencies and directions independently, is powered at night. Mark I could cool the Earth by shading the sunlight on the dayside, or warm it by reflecting back the infrared that pours into the night sky. The total power going in and out is roughly the same (although more goes out from the dayside for a variety of reasons). Thus there’s plenty of power available for the nightside to do street-lighting, or show ads in the sky, or whatever you’d like. Remember that because it’s a hologram, it can have a completely different effect for each spot on the surface: my night sky can be a giant telescope, and my neighbor’s can be a giant video game.

Coming soon to a planet near you

I’m fairly certain that a Weather machine will be built sometime this century. It seems straightforwardly within the capabilities of molecular manufacturing, particularly the Mark I form. There are plenty of people worried about things like climate change or asteroid impact that it could prevent or ameliorate. It could have an enormous economic value. All of these indicate that it should be built, but the most pressing and cogent reason that it will be is likely military.

Rremember the solar power plant. What if instead of a power plant beneath this thousand square kilometers of concentrated sunlight, there were an army, fleet, or indeed a city? Another way of specifying the amount of energy that would get pumped into the area is that it would be the equivalent of exploding a one-kiloton bomb every second for as long as you wanted. A Weather Machine would be a very potent weapon, even the Mark I version. Mark II could shoot down the moons of Mars.

Even without direct attacks, whoever can control the weather on Earth is pretty much in charge here. Anyone who objects and starts rattling their sabers gets twenty years of no summer and no growing season. For that reason alone, given the technological capability of doing it, it will be done. I cannot see the US government understanding that this is possible and not doing it. In fact, there are several other governments I cannot see understanding it can be done and not doing it.

If you are a smaller government without enough conventional forces to defend yourself well while your Weather Machine is being built, I would guess that approximately 5% of one is all you would need to have a setting that was a dead man switch: if someone came and blew you up and you quit sending out the control signals, all of the aerostats would default into snowball earth mode. It would be a doomsday device. This is troubling.

Once somebody gets 5% of one built, you’re stuck listening to them. You had better start building your own first, or at least simultaneously. In fact, it seems reasonable to imagine that by later in the century there are going to be several competing clouds of these things around. Hopefully they won’t end up physically competing with each other, but that the people in charge of them will come to some negotiation. That’s going to be all the more reason for someone wanting to be in the game. You have three quintillion balloons up, and I have one quintillion, and this guy over there has two quintillion, which means we get that many votes in the weather control world government.

The ultimate implications of a Weather Machine are mind-boggling. I can’t even come close to seeing all of the implications that it will have, but I’m fairly sure that it’s possible and that it will happen. It’s worth thinking about.