Chaos theory provides hints for controlling the weather:
While weather predictions have reached levels of high accuracy thanks to methods such as supercomputer-based simulations and data assimilation, where observational data is incorporated into simulations, scientists have long hoped to be able to control the weather. Research in this area has intensified due to climate change, which has led to more extreme weather events such as torrential rain and storms.
There are methods at present for weather modification, but they have had limited success. Seeding the atmosphere to induce rain has been demonstrated, but it is only possible when the atmosphere is already in a state where it might rain. Geoengineering projects have been envisioned, but have not been carried out due to concerns about what unpredicted long-term effects they might have.
As a promising approach, researchers from the RIKEN team have looked to chaos theory to create realistic possibilities for mitigating weather events such as torrential rain. Specifically, they have focused on a phenomenon known as a butterfly attractor, proposed by mathematician and meteorologist Edward Lorentz, one of the founders of modern chaos theory. Essentially, this refers to a system that can adopt one of two orbits that look like the wings of a butterfly, but can change the orbits randomly based on small fluctuations in the system.
Journal Reference:
Miyoshi, Takemasa, Sun, Qiwen. Control simulation experiment with Lorenz's butterfly attractor [open], Nonlinear Processes in Geophysics (DOI: https://doi.org/10.5194/npg-29-133-2022)
(Score: 0) by Anonymous Coward on Sunday April 03 2022, @03:50PM (8 children)
You can't predictably modulate non-linear system.
(Score: 3, Informative) by khallow on Sunday April 03 2022, @05:00PM (5 children)
Nonlinear != unpredictable. For a classic example, the Riccati equation [wikipedia.org] is nonlinear, but quite predictable.
And for more complicated systems, if you can find a near periodic solution (with period shorter than the time frame of divergence of the system), then you can modulate so that it becomes periodic (well periodic plus slight noise). Predictability obtained.
Weather systems are hard because they're driven by serious external forcing that has time lengths way longer than the divergence of the weather models (we can predict for about 2 weeks with modest accuracy, seasons are year length forcing and the El Niño\Las Niña phenomena oscillates in time lengths of a few years to few decades).
But still imagine hurricanes that rare see shore by plan (and are predictable well enough that you can arrange sporting events around them - decades in advance), and weather you can predict just by looking 30 years in the past. The question isn't really if you can modulate the weather, but how much energy and surface area would it take? Something relatively simple like forcing it into a decades-long cycle would probably take much less energy (orders of magnitude) than stop a huge hurricane before it hits land.
(Score: 0) by Anonymous Coward on Sunday April 03 2022, @05:57PM (4 children)
The problem is that it's a chaotic system and even a relatively minor change in starting conditions can lead to massive effects further out. We've known about rain shadows and the influence of ocean currents for centuries, at the least, but being able to cope with the minor discrepancies that crop up as you go further out is incredibly challenging. I'm sure we'll eventually get good enough for all practical purposes, but we're nowhere near that. It's only been in recent times that the local weather forecasters could beat the older system of getting calls from ships coming in.
(Score: 1) by khallow on Sunday April 03 2022, @10:16PM (2 children)
The whole point of modulation would be to correct as you're going along so that those relatively minor variations stay relatively minor.
The problem is that there's no serious means by which to control the weather system. The best I can see is albedo control and changing surface friction to wind flow. That's not much of a steering wheel. Nor do we have a scheme for seeing far enough ahead to come up with a controllable scenario like I described above.
(Score: 0) by Anonymous Coward on Monday April 04 2022, @02:44AM
Can't be done with current technology, at least not if you want a useful prediction. By the time you know what to correct, it's already to late. Even small errors propagate into something huge before too long.
(Score: 2) by dalek on Monday April 04 2022, @12:18PM
The summary looks like clickbait science to me. Maybe there's more in the journal paper, but the press release is short on details and doesn't seem particularly novel.
A good example of a system with two states is thunderstorm initiation. On many spring days in the Great Plains, we might describe the atmosphere as strongly unstable. Most of the time, the atmosphere is really metastable. When air near the surface is lifted, it often becomes colder than the surrounding air, particularly if there's a warm layer aloft. In that case, the air sinks back to its original level, because of the inhibition that prevents it from continuing to rise. If the inhibition decreases or the lifting mechanism becomes stronger, the lifted air might become warmer than its surroundings and rapidly accelerate to its equilibrium level, producing a thunderstorm. There are two distinct states, depending on whether or not there's a thunderstorm. The atmosphere is metastable, but given enough forcing, can be nudged into the other state. The effects from the thunderstorm could propagate into larger scales, an example perhaps being the thunderstorm growing into a larger complex, and perhaps altering atmospheric conditions so thunderstorms don't form elsewhere.
There are techniques that already exist to calculate what modulates the strength of the lift and the inhibition, or other meteorological processes. In effect, we would determine the sensitivity of the lift and inhibition to atmospheric conditions at an earlier time and in different locations. We could target observations in areas of high sensitivity, assimilate those observations into a model, and hopefully improve the forecast. There's already a lot of work to do things just like this to improve weather forecasting. If we wanted to simulate weather modification, we could create synthetic observations in those areas, assimilate them into the model, and see what happens. That's actually useful, too, because it can help to validate the sensitivity calculations.
Describing this as a weather modification technique is clickbait. This doesn't seem especially novel to me, based on the press release. The press release suggests that the researchers ran a control simulation, then ran several other simulations where they perturbed the control run with synthetic observations to test how that might affect the outcome. That's called an ensemble forecast. They're quite common and very useful. Maybe they've devised a better way to create ensembles or do the sensitivity calculations, both of which could be novel and useful. But describing this as a weather modification technique is just clickbait.
(Score: 1) by anubi on Monday April 04 2022, @06:40AM
Now, it's weather satellites sending us pictures!
"Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
(Score: 0) by Anonymous Coward on Sunday April 03 2022, @05:32PM (1 child)
yes, clickbait.
it's pointless to talk about individual trajectories of a chaotic dynamical system, but it does make sense to talk about external forcing schemes that modify the phase space of a given dynamical system.
i.e. you can imagine making a button (and the machine connected to the button), and when you push the button, the probability of individual trajectories "doing X" decreases a lot, or something like that.
but there is a huge difference between controlling a low-dimensional system (which I *think* is what they did) and controlling the weather.
the controlled fusion people have literally been working on this for decades, and they have a lot more control than the weather people --- their system is in the lab, and they can inject relatively large amounts of energy into it. to put it differently, when a fusion plasma becomes chaotic and develops a "tornado", we have the power spin down the tornado (it doesn't work because you can't simply stir the other way in order to stop the chaotic motion --- we don't know WHAT to do with the power we have). for the atmosphere there's no question that we do not have the power to affect on-going phenomena.
things that we can do to affect the weather?
* control tree distribution
* control building distribution
* control color of rooftops/streets
* choose where to dump heat into atmosphere/bodies of water
* choose where to dump aerosols into atmosphere
these are things that could, in principle, lead to long-term control of some weather patterns.
will it work in the context of a changing climate? no idea.
can we do anything else? yes, but I doubt we'd understand the consequences by using anything other than full weather simulations (i.e. no low-dimensional models).
(Score: 2) by hendrikboom on Sunday April 03 2022, @06:49PM
And we can do somewhat approximate full-weather simulations these days. We currently do ones that have a semblance of accuracy a few days to a week out.
If we are just looking at stochastic models and not accurate next-Thursday prediction, that might be good enough to start figuring out how these influences affect things.
(Score: 1, Funny) by Anonymous Coward on Sunday April 03 2022, @06:35PM (3 children)
how to control our rampant consumerism and capitalism?
Maybe then the pollution and weather will be at least manageable.
And as a bonus! turn militarism into a protectorate of the Environment.
Nahhh....who am I kidding.
What a dreamer.
(Score: 1) by khallow on Sunday April 03 2022, @10:19PM (2 children)
For example, egulation, taxation, crushing social mores, and religious opprobrium. The question is not "how", but "why"? Rampant capitalism and such has a huge number of advantages, like creating a society that can adapt to change. The control doesn't have these advantages.
(Score: 0) by Anonymous Coward on Monday April 04 2022, @04:14PM (1 child)
Where do you get your ideas?
(Score: 1) by khallow on Monday April 04 2022, @11:44PM
Regulation: the Raisin Administrative Committee [reason.com].
See also banning plastic straws, large glasses of soda, incandescent light bulbs, and toilets that flush and shower sprayers that spray. Taxation: Taxing luxury boats [boatingindustry.com].
See also sin taxes.
Crushing social mores: sumptuary laws [bustle.com] and other social mores.
It goes on and on. You dressed right for your class, wore an appropriate hat, and didn't show any loose hair as a woman unless you were a prostitute. Some of it was law, but some was excessive social disapproval. They were cracking down (hypocritically) on that materialism - but caring more in the process about one's station than helping anyone.
See also what happened to American Natives. Their social mores were great for the environment they originally were in - rough environment where people had to share to insure the tribe (or city for larger cultures) survived combined with often hostile neighbors. But those social systems were no match for the White Man, the Europeans who brought so much change to the Americas.
Religious opprobrium: The above bit about rules on clothing was partly religion driven. But I think the worst religious shtick is the idea that living requires sacrifice. For example [grist.org]:
This was written in 2009. No tipping points have been found since and nobody is interested in cutting emissions by 80% because they don't want to impoverish their societies.
(Score: 0) by Anonymous Coward on Monday April 04 2022, @12:29PM
lol, guiding chaos isn't sumthing i would go around telling everybody.
it's just shy of giving everyone the right to arm themselfs...