Episode 557 Cirrus Cloud Thinning

David Mitchell thought of the idea that thinning cirrus clouds would allow heat to escape from the world, and that the winter clouds at the poles would be the optimum clouds to thin. Blaz Gasparini is, like Mitchell, continuing the research on the matter, largely by modeling. Stephen Salter is working to design a suitable nozzle for marine cloud brightening, so he shares their interest in the properties of clouds. We discuss whether much can be gained at this stage from actual experimentation with either the winter cirrus clouds or brightening the low-lying summer clouds in the near-Arctic. Probably it is better to wait a year or so to see if current research answers some of the preliminary questions that would enable actual small scale experimentation to be worth the steep. costs. For the video, audio podcast, transcript, and public comments: https://tosavetheworld.ca/episode-557-cirrus-cloud-thinning.

David Mitchell

Blaz Gasparini

Stephen Salter


cirrus clouds, clouds, cirrus, work, arctic, cloud, aerosol, particles, ice, seeding, thinning, satellite, experiment, clean, radiation, atmosphere, ice crystals, heat, long wave radiation, cooling


Metta Spencer, Blaz Gasparini, David Mitchell, Stephen Salter


In a conversation about the role of cirrus clouds in global warming, Metta Spencer hosts a panel including Professor Stephen Salter, Dr. David Mitchell, and postdoctoral researcher Blaz Gasparini. They discuss the impact of these high-altitude clouds and potential strategies for cooling the planet. Cirrus clouds, unlike lower lying clouds, contribute to a warming effect on the planet, as they trap heat. Therefore, thinning these clouds could potentially allow more heat to escape from Earth. This method could be particularly effective in the Far North, where cirrus cloud thinning would be most impactful in releasing trapped heat.

Mitchell explains that sunlight coming into Earth transforms into longwave (thermal, or heat) radiation when absorbed by the Earth and re-emitted. Cirrus clouds at higher altitudes are colder, meaning they emit less radiation into space, effectively acting as a blanket that traps heat and re-emits it back to the Earth’s surface. This trapping effect is stronger during the winter, making it an optimal period for cirrus thinning to release the trapped longwave radiation.

They also discuss ‘Arctic amplification,’ the phenomenon wherein the Arctic warms faster than the rest of the world due to decreasing albedo from diminishing ice and snow, which traditionally reflect sunlight. Amplification is also influenced by dark soot falling on snow, further decreasing its reflective capacity. The discussion offers valuable insight into ongoing research and potential interventions in our warming climate.

Gasparini explains how thunderstorms help cool the planet by mixing surface air to higher atmospheric levels. These processes are generally absent in high latitudes, resulting in less efficient heat emission.

They discuss the different formation processes of cirrus clouds: heterogeneous (dirty) and homogeneous (clean) nucleation. Dirty cirrus clouds form with the help of ice-nucleating particles, often dust from deserts, that enable quicker ice formation. In contrast, clean cirrus clouds form on tiny liquid particles always present in the atmosphere.

Mitchell explains the physical properties of these two types of cirrus clouds and their different climate effects. Clean cirrus clouds have a high concentration of ice crystals, which increases the surface area to trap thermal radiation, making them a thicker atmospheric ‘blanket’. As a result, clean cirrus clouds are more effective at warming the Earth than dirty cirrus clouds.

Gasparini adds that clean cirrus clouds are also more persistent, as they consist of many small ice crystals that do not quickly sediment.           
Dirty Cirrus clouds have fewer ice crystals and allow more heat to escape. The goal of climate intervention would be to convert clean Cirrus to dirty Cirrus, thereby releasing more heat into space.

The conversion process involves introducing more ice nuclei or dust particles into the atmosphere, thus turning clean Cirrus into dirty Cirrus. Larger ice crystals in dirty Cirrus fall faster, reducing cloud lifetime and coverage, which allows more radiation to pass through. Concerns are raised about the impact of dust falling to the ground or on ice surfaces, but the consensus is that the amount would be negligible.

David Mitchell explains a rough calculation of the amount of Bismuth tri-iodide, an efficient ice nuclei at temperatures below -25 degrees Celsius, required for cloud seeding. About 2755 kilograms would need to be introduced into the atmosphere above 45 degrees latitude in both hemispheres every seven days.

Blaz Gasparini cautions that before jumping into engineering solutions, it’s essential to understand and verify whether perturbing these clouds can actually achieve substantial planetary cooling. More research and possible small-scale experiments are needed. Stephen Salter, an engineer, shows interest in the amount and lifetime of the material needed for cloud seeding, hinting at the potential development of a practical application.

The participants discuss how these particles, if too concentrated, could produce the opposite of the desired cooling effect by creating thicker cirrus clouds which trap more heat. Hence, achieving the correct concentration across the Arctic is crucial and presents a significant technical challenge.

Several methods of distribution are discussed, such as using aircraft or balloons, although these remain largely hypothetical due to a lack of scientific research on the subject. Another approach is using drones to disperse the particles more gradually, thereby ensuring a more even distribution and mitigating the risk of over-seeding. This process, however, would need to be continuously monitored, ideally by satellites, to ensure optimal results.

The conversation moves towards the feasibility of using similar equipment and methods for both winter and summer climate engineering projects. For example, both projects could potentially use a fine dispersal method, similar to marine cloud brightening techniques. However, creating fine powders for this purpose is considered significantly more challenging than pumping liquids through nozzles, given that fine powders tend to clump together.

The cost of such an experiment, based on estimates from 10 years ago, was roughly $6 million per year. However, it’s emphasized that these calculations were highly idealized, and more research would need to be done to better understand and control the variables in the system before proceeding on a large scale. The speakers suggest that smaller scale, controlled experiments should precede any broader efforts.           
Metta Spencer expresses her concern about the urgency of climate action, advocating for a fast-paced approach despite the traditional caution and procedural rigor of scientific research. She asks whether significant value could be gleaned within five years if the Canadian government funded research in climate change.

Stephen Salter emphasizes the risk of failure due to lack of time and resources, suggesting the importance of building confidence in a solution’s viability before acting.

Spencer later mentions she’ll be presenting a report to the Canadian Pugwash group, contemplating whether to recommend that the Canadian government invest in climate experimentation. While she concedes that current scientific consensus suggests that they’re not ready, she proposes the continuation of discussions.

Blaz Gasparini suggests that funding field campaigns could help develop knowledge on cloud formation and potentially reveal cooling strategies. This initial study would require a budget of around $10 million per year. David Mitchell builds on this by proposing a cheaper alternative involving satellite observation of naturally occurring atmospheric phenomena, including dust storms from Asia affecting cirrus clouds over the Canadian Rockies.

In response to Spencer’s concerns about Canadian involvement, Mitchell suggests the potential for Canadian remote sensing experts to join the project, while Stephen Salter also underscores the importance of test tube chemistry experiments and aerosol size measurements. However, Spencer struggles to identify any Canadian scientists in this area.

The discussion emphasizes the tension between urgency and caution in addressing climate change, the potential of various scientific methods, and the need for collaboration across borders.

Metta Spencer raises a question about the value of exchanging information that might influence the Canadian government to utilize satellites and collect samples for environmental purposes. Stephen Salter hopes that satellite observations could be useful, but doesn’t see other immediate benefits.

Spencer also introduces the topic of the Iron Salt Aerosol project. She suggests a potential future show bringing together experts to discuss their findings under a unified context. Her intent is to guide the Canadian Pugwash group’s decision-making process. She concludes that scientists are not yet prepared to conduct expensive ground experiments or cloud fly-throughs, but recommends continuing dialogues to determine whether there has been sufficient advancement in knowledge to propose a specific project for Canadian funding.

David Mitchell agrees, suggesting that an aircraft could be employed to measure cloud content without actually performing cloud seeding. When asked about using drones, he confirms their potential usefulness.

Blaz Gasparini introduces another proposition about marine cloud brightening and the potential benefit of reducing wintertime Arctic clouds. He suggests introducing ice-nucleating particles into these low-lying clouds to influence their behavior, potentially affecting the climate positively.

Stephen Salter contributes by mentioning a paper he authored about using small aerosols to remove liquid water vapor, thereby preventing other clouds from forming. He suggests using existing technology to create smaller aerosols by adding freshwater to saltwater, thereby reducing the salt content that facilitates nucleation.

Finally, Metta Spencer brings up a paper shared by Michael Diamond about mixed-phase cloud thinning. She questions whether this topic should be included in the conversation, indicating its potential relevance to their discussion.

Gasparini discussed the research on the Wegener–Bergeron–Findeisen process, which involves the growth of ice crystals in supercooled clouds, leading to precipitation and cloud depletion. Mitchell emphasized that a blanket removal of clouds could negate any cooling effects, and discussed the prevalence and influence of ‘clean’ cirrus clouds on the climate.

These discussions indicate that research into cloud manipulation for climate intervention is still in its early stages. Mitchell suggested that small-scale field experiments could provide valuable insights without causing harm to the broader climate. Salter estimated that conducting such experiments in various meteorological conditions over two years could cost a few million dollars.

Gasparini, a climate modeller, highlighted that current climate models lack the detail necessary to accurately simulate cloud behavior on a global scale. Mitchell agreed, suggesting that further research into cirrus cloud behavior is crucial not only for climate intervention, but also for improving climate models.

Overall, the conversation revealed that while this area of research could be potentially significant for mitigating climate change, it is not yet sufficiently developed for deployment. Metta Spencer concluded the conversation by suggesting the Canadian Pugwash group to support ongoing discussions in this field for at least another year before reassessing the situation.


The following transcript has been machine generated using “otter.ai.” Prior to using information from the transcript, please watch the video to catch any obvious errors.

Metta Spencer  00:00

Hi, I’m Metta Spencer. Today we’re going to have, let’s go up in the clouds shall we, and have a look at these cirrus clouds and what they’re doing to the world and what we can do about them. This is going to be an adventure for me, because it’s a new area of research that I’m becoming interested in, but know nothing about in the past. You know, for the last several months, I’ve been running a series of conversations about how we might be able to cool the planet somewhat by brightening the clouds. And one of our the leading people doing research on Cloud marine cloud brightening is Professor Stephen Salter, of the University of Edinburgh. and he’s with me today. So I’ve, we’ve had a number of conversations and along the way, I’ve found out that in fact, marine cloud brightening is not the only conceivable way of, of cooling the planet or, and specifically of, of cooling the northern the uppermost pole of the planet. And that there are a number of other possibilities, one of which I have not explored before, and that is the possibility of thinning the cirrus clouds, it seems that all clouds have two different effects. One is to, some warming effect and some cooling effect. But the different clouds have different balance. and the low lying clouds are really good for brightening the clouds and keep, and keeping the planet cooler. Whereas the high clouds or the Cirrus, those wispy clouds really high up. and their overall effect is mainly to warm the planet. So if we could thin them out, maybe we’d let some of the heat out of the tougher parts of the world and have some good effects. So that’s what we’re going to explore today. The man who presumably thought that out is Dr. David Mitchell, who is now going to be with us. He’s in the, he works of the Desert Research Institute in Reno, Nevada. Good morning, David. How are you? This is a machine-generated transcript that may contain errors. Do not cite it without checking for yourself by watching the video and catching any obvious errors.  Good Morning.  Good to meet you. and in Vienna, is Blaz Gasparini, who is a postdoctoral researcher at the Department of meteorology and Geophysics at the University of Vienna. Good morning Blaz.

Blaz Gasparini  02:35

Hello. Thanks for inviting me.

Metta Spencer  02:39

Say hi from Edinburgh.

Stephen Salter  02:42

Hi, I’m in Edinborough. I’m doing the engineering for the Twomey  effect and John Latham proposal to use it. I’m a nuts and bolts engineer. So I’d like to know what the problems are for getting stuff up into the Cirrus clouds. I could do something to help.

Metta Spencer  03:01

Wonderful. All right, let’s talk about, I think David sent you that guy who thought the whole idea up in the first place. Why don’t you explain everything to it.

David Mitchell  03:11

It’s probably the most complicated of the three mechanisms you might say that are widely discussed. So Blaz if I leave something out, feel free to chime in and fill in the gap. So the, the general idea is to release more heat so that most well so Stratospheric Aerosol injection is the most studied technique, and marine cloud brightening, I would say is the second most studied, and they both are reflecting sunlight to cool the planet. So the cirrus cloud idea’s a little different. It’s thinning the cirrus clouds and reducing their cloud coverage to release more thermal radiation or heat from the planet. So in that sense, it’s acting more like a greenhouse gas, because the greenhouse gases are trapping heat, and they re-emit it back to the surface where we live, and that’s what’s how the warming works. So the cirrus clouds have a similar role they, they reflect some sunlight, but they trap more long wave radiation or heat radiation, then the other kind of clouds do. So they’re re emitting a lot of the thermal radiation back to the earth and they have a slightly warming effect. So if you can thin them out more, more long wave radiation escapes to outer space and that cools the planet. So that only occurs, or it, that the method would be most applicable in the high latitudes where there’s a lot more long wave radiation around. So in the tropics, that probably wouldn’t have much of an effect. But, say, during the polar night, where there’s no sunlight at all, it would have its strongest effect. Is it’s all long wave radiation in the sunlight.

Blaz Gasparini  05:25

Right. So I guess if I may, it’s more about the lack of sunlight,  and that’s why we’re thinking about the high latitudes. Yet, the heat is there also in the tropics, but solar radiation is also extremely strong. Sorry to have interrupted. Yeah,…

David Mitchell  05:44

Good, yeah, thank you. Yeah.

Metta Spencer  05:47

Now, when you say long wave radiation, I’m I, I have to remind myself, so if I don’t remember, most of our viewers may not either. Can you help us?

David Mitchell  05:58

Yeah, so I’m using a bunch of words, that all mean the same thing. So long wave radiation, heat radiation, and thermal radiation, they all mean the same thing. It’s what what we feel, but we don’t see it. So like, when you have a heater in your house, like electric heater, whatever, you’re feeling that radiation from that, that, and that feels hot, it warms you up. That’s thermal radiation. and sunlight also has about half of the sunlight we receive is thermal radiation, it’s called the near infrared. We don’t actually see it, but it’s still there. and we feel it, it warms us. So the Earth is constantly giving off thermal radiation at much longer wavelengths than the sunlight.

Metta Spencer  06:51

Alright, sorry but where, where do these rays come from and in the Earth is it something like, the core of the Earth is hot, and it just seeps out? Or, is it something that comes from the sun, and it just gets trapped by the greenhouse gas?

David Mitchell  07:08

it’s basically a equilibrium effect between the Sun and the Earth. So what comes in has to go out, so the sunlight is coming in. and that contains a lot of energy that the Earth absorbs. And the earth re emits that energy in the form of long wave radiation, heat or thermal radiation, whichever you want to call it. But it, the Earth takes in the sunlight that we can see visually. and that turns as soon as it hits the surface of the Earth, it heats up the rocks, that heats up everything it hits. and that’s the transformation and energy that occurs naturally. And then the earth is re emitting that energy, but in the form of long wave radiation or heat. So it’s a balancing act basically, what comes in has to go out, the two have to equal each other.

Metta Spencer  08:08

That’s pretty clear, I think.

David Mitchell  08:11

Okay, good. So did you have any questions? Oh,…

Metta Spencer  08:15

Yeah. Yeah, how, how do the clouds, how do these cirrus clouds keep the longwave radiation trapped?

David Mitchell  08:26

They’re at a colder temperature than other clouds are. So if you are like an observer from a spaceship, you would, if there’s a deck of cirrus clouds between you and the surface of the earth, and you’re measuring the temperature of the Earth, you’d measure a cold temperature, because the cirrus clouds are at a high altitude about, well colder than minus 40. Usually, that’s centigrade. So that’s the same in Fahrenheit, it turns out minus 40 Fahrenheit, it’s also the conversion there. So they’re very cold is the point. So because they’re emitting radiation, a lot less radiation into outer space, because of that cold temperature, that implies that they’re trapping radiation in their lower parts of the cloud. So the part, the lower part of the cloud is re-emitting radiation to the Earth’s surface, and that’s like a blanket, you could say it’s what it’s trapping the heat and re-emitting it back to Earth.

Metta Spencer  09:37

Okay, now that that happens all around the Earth, right? That would be true wherever the cirrus clouds are, but the important part is, are the important kinds are the ones in the Far North.

David Mitchell  09:49

Right there, like you said, the cirrus clouds are everywhere. But where this cirrus cloud thinning, cooling method would work best would be the Far North, and especially in the wintertime, not so much in the summertime, but in the wintertime. And that’s because the sun angle is really close to the horizon, or it’s absent altogether. So, when the sun when the sunlight is minimized you, that leaves you with just longwave radiation, because that’s or thermal heat radiation that’s present all the time, where the sunlight is only present part of the time.

Metta Spencer  10:38


David Mitchell  10:38

And a net cooling effect.

Metta Spencer  10:39

If you have more, if you punch a hole, then what you’re going to let out is probably primarily long wave radiation in the winter, where in the summer, it would be a mixture of both.

David Mitchell  10:51

Correct? Yes.

Metta Spencer  10:53

Okay, so you get more bang for the buck, as you put it. Yeah.

David Mitchell  10:57

Correct. and Arctic amplification is strongest in the wintertime. So that’s a beneficial correlation there.

Metta Spencer  11:09

Say it again. Arctic radiation is strongest in the wintertime. Oh, sorry.

David Mitchell  11:13

Oh sorry, amplification.

Metta Spencer  11:15

Amplification, what do you mean?

David Mitchell  11:18

That, it’s a term that climate scientists use a lot. It means that the Arctic is warming up a lot faster than the average temperature of the planet to the global warming. So some people say twice as much, but I think more recent estimates are more like three or even four times.

Metta Spencer  11:41

I’ve heard seven times.

David Mitchell  11:44

Okay. Yeah.

Metta Spencer  11:46

Different people have. I suppose it’s partly a matter of temperament. What, how much?

David Mitchell  11:52

What time period you’re measuring over, as well.

Metta Spencer  11:56

Okay. Now, so. and that’s I don’t think everybody fully understands that. Why, why it’s happening so much more in the north, right. So can, can you explain that a little? Do you think, you know, all of the explanation for why the the Arctic would be warming so much faster than the rest of the world?

David Mitchell  12:20

That’s an area that’s still under research, scientists are still trying to understand it better, but they know that part of it is due to the lack of albedo, like the sea ice is disappearing. The snowpack is less, you know, the seasons, like we’re getting spring earlier in the Arctic and that kind of thing, because there’s less snow. So the snow and the sea ice are reflect they’re like a big mirror that reflects sunlight. So when you take that reflective capacity away, the Arctic warms up.

Metta Spencer  12:57


David Mitchell  12:58


Stephen Salter  12:59

There is also black soot falling on the snow.

David Mitchell  13:04

Yeah, that’s the issue. So one of the problems Stephen just pointed out is there’s air pollution or soot depositing on the snow and making it darker. and that’s going to warm up the planet, it’ll melt the snow faster, and when the snow disappears, and you have more Arctic amplification that way.

Metta Spencer  13:24

Okay. Okay, so I sort of, I knew it was true, I just didn’t. I thought maybe there’s some extra secret reason that I hadn’t heard about.

David Mitchell  13:34

There’s other reasons, too. Yeah, Blaz can talk about that.

Blaz Gasparini  13:38

You know while I’m a climate scientist, I’m not sure if I’m capable to explain it in a very easy to understand way. But I think another reason so even in Arctic where there would be no sea ice would still warm up or response to any change in climate and stronger. and I think if we put it in very simple terms, I think it’s because of the absence of convections in the Arctic atmosphere. and that would, you know, be working that mechanism even even in an Arctic free in an ice free Arctic.

Metta Spencer  14:14

Oh well now explain what is the absence, what is convection and what, why is it absent there?

Blaz Gasparini  14:23

So basically, the thunderstorms mix in air from the surface towards the higher levels of the atmosphere in such way they cool the atmosphere. and these processes are pretty much absent in the high latitude climate. And therefore, if the surface warms, it cannot emit this warmth as effectively as it does in the tropics. So basically, the thunderstorm in the tropics, to some extent help cool the planet.

Metta Spencer  15:04

I, and they don’t have thunderstorms in the in the north.

Blaz Gasparini  15:09

That’s not very common. No. That would be rare.

Metta Spencer  15:14

Okay, fine, thank you. Okay, so here we have this, this blanket. and the blanket is there all the time, but if you poke a hole in the north, you can let some of the heat out.

David Mitchell  15:30


Metta Spencer  15:31

Okay, how are you going to do that?

David Mitchell  15:33

Yeah, that’s the other thing. So the way this technique works is there’s, this is where it gets a little more scientific. There’s two different processes by which the ice forms in these cirrus clouds. and one pathway is called heterogeneous nucleation. Ice nucleation, and the other is called homogeneous, ice nucleation. So, for brevity, I’ll just call him het and hom.

Metta Spencer  16:07


David Mitchell  16:08

So there’s a lot of research going on, to try to understand these two mechanisms in cirrus clouds. And that’s where my work is focused, actually is trying to understand which mechanism or what is the different percentages of contribution from each process.

Blaz Gasparini  16:30

David, can I suggest you instead of het and hom, I believe we can call them in a simpler way as clean cirrus, and sort of dirty cirrus. I mean, of course, that’s an approximation, but it might might help to.

David Mitchell  16:43

Okay, yeah. to conceptualize.

Blaz Gasparini  16:45

Yeah,I feel like because hat and home that gets lost.

David Mitchell  16:49

Yeah, yeah. So I was just looking for a straignt way of saying it.

Metta Spencer  16:55

Explain your analogy Blaz.

Blaz Gasparini  16:56

So yeah, I mean, that’s where David was anyway, going to so this hom cirrus, or as I call them, clean cirrus, they form,on just some small tiny solution droplets, tiny particles, liquid particles, which are present always in the atmosphere. That’s why they call them clean. But the dirty one, the heterogeneous one, het one, they form with the help of some ice nucleating particles. And those are, as the name says, some particles that help form ice crystals. Those are particles that are floating in the atmosphere, but they’re not available all the time everywhere. You can think of those particles as dust from some of the world’s deserts, from Sahara or from some other deserts. And these particles will start the ice formation quicker compared to the situation when these particles wouldn’t be present and we would get to clean Cirrus, so dirty Cirrus with dust, aerosols or some other particles that are floating in the atmosphere. Those would form easier compared to those clean cirrus. And I think David, you can you can now follow up on that. Okay.

David Mitchell  18:28

Okay, so there’s a difference in the physical properties of the cirrus clouds between dirty Cirrus and clean Cirrus, so that the clean Cirrus form a higher concentration of ice crystals, we call it the number concentration. It’s a lot higher for the clean Cirrus, the dirty Cirrus, well, you could say that the ice nuclei population is a subset of the action of the overall aerosol concentration. So just all the particles in the atmosphere there we call them aerosol particles. That’s a lot of particles, but the ice nuclei is a very small percentage of those. So that mineral dust and a few other types of ice nuclei are forming relatively small number of ice crystals compared to homogeneous or clean Cirrus nucleation.

Metta Spencer  19:29

So clean is more common?

David Mitchell  19:32

It’s more common without the ice nuclei that well yeah, it’s a it’s sort of like a backdrop or a back nature’s backup plan. If there’s no ice nuclei around, then that homogeneous or clean Cirrus process takes place. There’s always enough aerosol for that.

Metta Spencer  19:52

Oh, I thought the aerosol was the dirty?

Blaz Gasparini  19:57

In both situations, we’re talking about some sort of aerosols are some sort of particles that are floating around. Those for the clean cirrus, they are liquid, and they’re very small and they are present all the time, all around. While the ones for the dirty cirrus they are, sometimes they’re sometimes not. And we’re not quite sure how many of those that are around, and that’s why we’re also not quite sure how frequently those dirty cirrus form compared to the clean cirrus.

Metta Spencer  20:28

But now when you say clean, and ice, or liquid, is it a water? Or is it some other liquid?

David Mitchell  20:37

It’s liquid water. I mean it’s yeah.

Metta Spencer  20:42

So it’s not aerosol. Right?  Aerosol was explained to me as anything that’s not water. Is that is that fair?

Blaz Gasparini  20:52

I, I think if you I would think of that as some sort of very small droplet, which is liquid, liquid phase, and this is in liquid phase, even at very cold temperatures at minus 40 degrees Celsius, these droplets would still be in liquid phase. That’s very weird. Sounds very weird that 40 minus 40. Normally, everything is frozen. But those particles, they’re not quite water. So that’s why they can still be in liquid phase in those conditions.

David Mitchell  21:27

Yeah, you could call them haze droplets.

Blaz Gasparini  21:29

Right? Yeah.

Metta Spencer  21:32

Okay, so we got these two kinds of particles, and what difference does it make in in the effects? I presume they have different consequences.

David Mitchell  21:44

Okay, so the, the clean cirrus that have a high concentration of ice crystals, they have more ice surface area that can trap the long wave radiation or the thermal radiation. So they’re like a thicker blanket, you could say, you know, we can think of the cirrus as have a having a blanketing warming effect, trapping or re-emitting the thermal radiation back to the surface. So the clean cirrus clouds do that better than the dirty cirrus clouds.

Blaz Gasparini  22:19

They’re also, because they’re a form of many small ice crystals, they will also stick around for longer. So those ice crystals don’t sediment very fast, because they’re small compared to those dirty cirrus, which are rather composed of smaller number of larger ice crystals.

David Mitchell  22:38

Right, and that’s an really important point if this, it is a rather an involved process, but that’s…

Metta Spencer  22:47

I need more help. Because I thought I understood that clean, clean cirrus clouds are mostly made of water, particles, liquid things, but now I’m hearing you call them ice.

Blaz Gasparini  23:02

Yeah, sorry, I think we have to make clear all of those clouds are form are ice crystals. However, the clean ones form on haze particles and the some sort of liquid party particles, which is…

Metta Spencer  23:21

But the overall particle in the long term is, is ice.

Blaz Gasparini  23:27


Metta Spencer  23:28

And they are the kind that you, that will make a difference. If you can knock them out. You let the the heat out of the world,

Blaz Gasparini  23:37

Right the haze particle before you find you form and ice crystal doesn’t really have much of an influence on climate. But that will change indeed once the ice crystal is formed.

David Mitchell  23:51

Another way to think of it is the clean cirrus start with super small, liquid particle, super small, and that freezes and immediately it turns into a nice crystal at that point when it freezes. The dirty cirrus they form on particles so we don’t really consider the clean cirrus forming on particles. We consider it as a solution droplet. It’s a small distinction, but there’s no insoluble particle involved. It’s just liquids substrate. The other dirty cirrus form on something you could look at as a particle, you could look under a microscope and go Ah yes, there’s a particle there, and that’s where the ice grows from.

Metta Spencer  24:41

Okay, now I’ve got the impression so far, that if you only had to think about the clean Cirrus clouds, you’d be easier, but there’s dirty serious clouds confuse you.

Blaz Gasparini  24:54

That’s my summary. Yes. I think that’s one of the big questions in our field, is, as I said already before, what fraction of those clouds that are out there today are formed by one versus by the other mechanism, because this will again really limit our, you know, impact that we may have by by but by trying to perturb to change this clouds and get some cooling effect.

David Mitchell  25:28

So you, you might be wondering, why do we want to change the clouds? Is that coming up for you? Alright.

Metta Spencer  25:37

Well, I know the answer, if I’m not mistaken, and that is that, if you can get if you can poke a hole in the in the clouds, some of the heat from the world is going to go out into outer space and we’ll all be better off.

David Mitchell  25:50


Metta Spencer  25:51

Is that answer?

David Mitchell  25:53

Yes, but the more detailed answer would be when you say poke a hole, that means converting the clean cirrus to dirty cirrus. That’s how you poke a hole.

Metta Spencer  26:08

Okay, alright

David Mitchell  26:10

Yeah, so earlier, I was saying that the clean cirrus are a better blanket for trapping the heat than the dirty cirrus. So if you convert the clean cirrus to dirty cirrus, then more heat can escape, because their physical properties change. So there’s fewer ice crystals in the dirty Cirrus. The concentration, number concentration of crystals is much lower in the dirty Cirrus compared to the clean Cirrus. and you could also say that that ice mass content is much higher in the clean Cirrus versus the dirty cirrus, and sometimes the crystals are a lot smaller as well in the clean cirrus.

Metta Spencer  26:55

So you’ve got these two kinds of clouds up there bumping into each other. and when when a dirty one bumps into a clean one, it knocks out the clean one and, and that’s how you get [inaudible]

David Mitchell  27:10

No it’s not quite like that, it’s that you want to introduce more ice nuclei into the atmosphere. So the way you convert the clean to dirty is to just inject more dirt, you might say, or ice nuclei or dust particles into the atmosphere. and that’ll automatically shift the clean cirrus into dirty cirrus.

Metta Spencer  27:38

But what’s the advantage of having dirty cirrus if it unless it knocks out something else? Because it does a dirty cirrus happen to just fall quicker? Or?

David Mitchell  27:50


Metta Spencer  27:51

Why do you want to make things dirty?

David Mitchell  27:53

Because like Blaz pointed out earlier, when you have fewer ice crystals, they’re bigger ice crystals, because he have the same amount of condensation occurring in the cloud. So that same amount of condensed, condensed water mass goes into fewer but larger ice crystals. and the larger ice crystals fall faster, the faster they fall that the shorter the lifetime of the cloud is. So the clouds don’t linger in the atmosphere as long, their lifetime is shorter. So there’s less cloud coverage that way, because the crystals are bigger, and they fall out faster. So that’s the key ideas, they fall flat, the crystals are falling faster. So that reduces the cloud lifetime. and it also makes the optical thickness less. So the clouds in terms of the radiation penetration. More radiation can pass through the cloud now because they’re thinner.

Metta Spencer  28:55

Okay, that sounds good. Except then we’ll let’s go back to this albedo effect. Where we a few minutes ago talked about the fact that if you get your ice on the ground and the snow on the ground dirty, this reflects, this cancels out the reflectivity of the albedo. Therefore, if you dirty up the clouds, those and those, those clouds are going to fall someplace, it’s going to dirty up the ground of the ice, isn’t it?

Blaz Gasparini  29:24

This effect is extremely likely to be negligible. It’s really just a tiny amount of those particles, dust, that you would need to perturb the clouds compared to what is anyway deposited today on ice surface from sources which are near the surface.

Metta Spencer  29:46

So it’s just too small an amount to worry about.

Blaz Gasparini  29:49

I think your question is very well posed. I think we should study that, but my strong feeling is that this will not be a problem, but someone should to go and do this calculation, I think.

David Mitchell  30:02

But we do have some information from conventional cloud seeding. There’s operational cloud seeding going on all the time, and the Sierra Nevada mountains of California, Nevada, and also the Rocky Mountains. And they look at the snow after they do the cloud seeding, because they’re seeding regular winter storms to try to make, increase the snowpack in the mountains. So people go out and measure the amount of silver iodide in the snowpack, and, you know, the silver iodide, just what they used to seed the clouds that it mimics and ice crystal. So it’s a seeding agent or seeding aerosol. So they go into measure the silver content of the snowpack, and it’s around a maximum of 10 parts per trillion, usually much less.

Metta Spencer  30:56

I am impressed, except I have no idea how much that is really.

David Mitchell  31:00

Yeah, it’s really low. It’s like 1000 times less than parts per million.

Metta Spencer  31:05

Okay. Okay. So, so that’s not enough to worry about. Okay. But you know, I’ve been talking at times with Franz Oeste who’s talking about using for the iron salt aerosol project in the north. He would use titanium instead of iron, because it’s white, and so when it falls, it’s it doesn’t dirty up the soil. So I was just wondering if you use titanium for seeding the cirrus clouds? Would that work?

David Mitchell  31:38

Not sure. I haven’t heard any results on that. Usually, it wouldn’t work because they have to have a hexagonal geometry to mimic an ice crystal and activate ice crystal growth.

Blaz Gasparini  31:56

I think there might be some…

David Mitchell  31:57

Not always like that. But a lot of times, go ahead.

Blaz Gasparini  32:00

I think Titanium Dioxide might have the right structure to nuclear ice. But I think nobody really wants to put something that is not naturally present, and that we don’t know very well, in such a pristine environment like the Arctic. So yeah, that’s why I’d be a bit hesitant about this ideas at this point.

Metta Spencer  32:28

Okay, okay, well, that I’m gonna, I’m gonna shut up. and, Stephen, I’ve been quiet your turn, you interrogate these guys.

Stephen Salter  32:36

Well, I’d like to know the sorts of material that you might want to put up with the weights and the lifetimes and how fast it falls. And I think the marine cloud brightening suffers the same problem, in that it doesn’t work in the winter, because it acts as a blanket. And so something could get rid of the blanket is, is, is a useful counter to that we would have to stop our spraying in September in the Arctic, we want to do it when we got a very high solar input in sort of May, June, July, that we know that it would be working in the wrong direction, if we don’t stop. Unfortunately, it has a much, much shorter life than the sulfur aerosols, they want to use. For stratospheric treatment, we have a very short life, and this means that we have more control about where and when. Because if we stopped doing it, it’ll go away way quicker. We can choose the regions and the times when we want to treat. But we do want to do it in the Arctic for the winter. Sorry, for the summer.

David Mitchell  33:56


Stephen Salter  33:57

But please let me know what, how much material you need, and how you want to put it and how long it’s going to stay there. Maybe I can think of ways of solving that problem.

Metta Spencer  34:08

Stephen is a nozzle designer. He’s working on a nozzle.

Stephen Salter  34:13

Yes. Well we have 200 million nozzles and an eight inch diameter silicon wafer. Yeah. Got to keep them clean as well.

Metta Spencer  34:23

Have you guys got to the point of having to think about what kind of nozzle you want? Or is your notion just still pretty theoretical?

Blaz Gasparini  34:31

Yeah, we I mean, so I mean, Stephen, of course, as an engineer, he’s thinking about problems of engineering. and I think it’ll be very interesting to make a study on that. However, I have to say first that unlike stratospheric sulfur injections, or marine cloud brightening for cirrus cloud thinning, we are not yet sure whether we could really achieve some substantial cooling of the planet or not. And that’s why at this point, I would say going In the engineering details is to me a bit in the second plan to really be sure that if we perturb these clouds, we really get a cooling effect and not just the opposite. So here we’re really playing at the narrow, in some conditions where if you modify clouds, which are already, you know, formed by aerosols, dirty, cirrus. If you add more dirt more dust up there, then actually what you get is just the opposite of what you want to, you get more clouds so we’ve got more heating. That really points out our lack of knowledge about this nucleation mechanism, and that mean, many people are working on that, including David. But ultimately, we may want to go up there and first make very detailed measurements from aircrafts to be really sure how those clouds are formed before trying to perturb them or trying to perturb them in large scale. I think it might be also work to make some very small scale experiment that would let us know what happens if you perturb the cloud. Another point, I mean, of course, you may ask. So what about the laboratory studies? Right? I mean, stratospheric folks, they did some experiments in the controlled laboratory conditions. So what about Cirrus and cirrus cloud thinning? And here again, we do have only one single study looking at that, which is not yet even published. So we are at a very, very early stage of research of those Cirrus modifications. Yeah, I don’t know, David, if you want to add up more on that.

David Mitchell  36:57

Yeah, I have a long time ago. Back in, maybe 10 years ago or so I did these rough calculations of if you seeded all latitudes poll, [pollwards] of 45 degrees latitude, how much seeding material you would need, and it takes about seven days, but we assume Bismuth tri-iodide because that if instead of silver iodide, which is conventional cloud seeding, but bismuth tri-iodide activates, it becomes a efficient ice nuclei at temperatures colder than minus 25. And after when you go colder than that, it’s as effective as silver iodide. So that was Bill Finnegan that gave that information. He’s an expert on cloud seeding. Anyway, it would take 2755 kilograms of Bismuth tri-iodide.

Stephen Salter  38:08

Can you say that again please, two seven.

David Mitchell  38:11

2755 kilograms of Bismuth try iodide to you would take seven days to deplete that much. So that’s like he had CM. I think that’s the Oh, hold on a sec.

Stephen Salter  38:33

It’s very accurate to give it to four significant figures.

David Mitchell  38:38

Of sure it’s a calculation. So it’s really not meaningful in terms of the significant figures, but it’s the half life. So this calculation, said, that’s how much would be depleted from the atmosphere in seven days. Let’s see. I’m just reading some notes here. So I’m a little unorganized. Says so every seven days, about 2755 kilograms of Bismuth tri-iodide would be, would need to be introduced into the seeding layer of the atmosphere above 45 degrees latitude in both hemispheres. If you were seeding both hemispheres.

Stephen Salter  39:25

That’s just less than three tonnes.

David Mitchell  39:27

Right,  yeah, it’s it goes a long way, because that size of the particles are about 0.1 microns.

Blaz Gasparini  39:35

But even now a challenge for the engineer, all good, much particle. So how do you distribute them over the whole Arctic in the right concentration because if you happen to have too many of those, then we would actually get probably just the opposite effect form with too many ice crystals and I think this is really something that actually nobody ever looked into?

David Mitchell  40:02

I agree.

Metta Spencer  40:03

Are you thinking in terms of a plane flying through and just throwing a handful out of the window? Or are you, maybe a balloon that stays up there for a long period of time, and can do it gradually? Or how are you going to distribute this stuff?

Blaz Gasparini  40:22

Well, to be fair, I don’t think anyone wrote a scientific publication on the transport mechanism. So all so just to make sure we have some ideas, right, but now those are just, you know, very, they’re just like, are just our ideas for now. and I cannot be sure what of the proposed ideas may work best. But of course, you said you mentioned first the aircrafts. and yeah, I think that’s a reasonable first idea to mix in maybe some of these particles in the jet fuel and then inject them. Or balloons, I’m not sure how much you could loft up there. Maybe you may have some drones at this stage, which could even reach levels high enough for a cirrus I mean, it’s hard, I think, but maybe we’re getting there.

Metta Spencer  41:23

You need to do it over a gradual period of time. I mean, not not just one, you don’t throw up a clump of it out and then fly on another 100 miles and throw another clump out, you actually need to, a continuous even dispersal of just a little bit here and there.

David Mitchell  41:44

Yeah, that’s right. So it would have to be a continuous dispersal of these aerosols. And it’s, it’s like Blaz pointed out, it’s it’s there hasn’t been much work or research done on how to get an even concentration of those particles. Because if you have too many of them, they can create the opposite effect. They call that over seeding, making the cirrus cloud thicker and trapping more heat. So there’s a sweet spot, you could say, and how, what the right cause optimal concentration of ice nuclei is in the atmosphere.

Metta Spencer  42:24

I didn’t know that.

Stephen Salter  42:25

It sounds, if you will, rather low concentration over a big area. Is that right?

Blaz Gasparini  42:30

Yeah, yeah.

Stephen Salter  42:33

So something coming out of the back of an airplane would be a much higher dose than you want.

David Mitchell  42:38

 Right? Yeah, that’s the problem.

Stephen Salter  42:41

And you need to…

David Mitchell  42:43

So you might need a fleet of drones, sending up the aerosol in smaller amounts that are more numerous in their coverage. To get a more even concentration.

Metta Spencer  42:55

I assume that what you do is start off small. Because if you think that if you overdo it, you’re in trouble. But I got the impression so far that you’re not exactly sure when that what that tipping point is. And and so maybe you want to do it just a little bit and kind of watch out as you increase it. Is that a reasonable thing?

David Mitchell  43:18

Yeah, that’s reasonable I think. We can monitor some of this from satellite to see whether we’re, what kind of cloud we’re producing. You know,…

Metta Spencer  43:29

Look, if you if you had somebody with with drones right now, who could do this. and you know, we’re reading every day about how many drones are going to Ukraine? Presumably, we’ll get two or three of them up in the Arctic, and have them just begin experimenting? I mean, if you’re doing a very small scale, would that be of pragmatic value? Would you be learning anything from that? Or do you still have to do a whole bunch more investigating at the level of modeling before it makes sense to actually do anything? You’re physically experimental?

David Mitchell  44:08

Yeah, it’s sort of a controversial question. But I think scientifically, you could argue that you would learn something by doing an experiment like that, in a limited region of enough to measure from space like from a satellite, but small enough to make it harmless.

Metta Spencer  44:25

The reasons for asking you at the same time as as Stephen Salter is that we are literally thinking in terms of proposing that the Canadian government invest some money and doing some practical experimentation with marine cloud brightening. And I’m thinking, he would be working in the summertime. and you guys have been working in the wintertime, but very conceivably, some of the same equipment would work. Be useful for both projects, and as far as I can tell, you would not get an each other’s way, because you’d be working at different times of the year. Would you, could you imagine doing a joint research project in which you try both, both kinds of experiments?

Stephen Salter  45:18

We could certainly both use satellite observations to see what we’re doing. But I would reckon that we could prove no, almost everything that we need to know, just by measuring contrast changes for clouds that we couldn’t see ourselves that the contrast change you need to change to save the planet is well below with the contrast detection of the human eye. So you have to take a lot of images and superimpose them and compare them with somewhere where you didn’t, you hadn’t been treating. and and this is a way in which you could do an experiment, just with satellite. So the satellite measurement could be very, very similar with two things.

Metta Spencer  46:01

Can you use the satellites, even in the dark in the winter?

David Mitchell  46:04

Yes, we have a new satellite technique now that measures the number concentration, effective particle size and ice water content of the cirrus clouds, and it’s unique to cirrus clouds, so doesn’t look at the other types of clouds. So that operates all year long, you don’t it’s in the long wave, in terms of the radiation that it’s measuring, it’s it’s thermal radiation that’s given off by the earth that happens all year long. So you don’t need sunlight for that.

Stephen Salter  46:39

And you get a lot of data for very little money.

David Mitchell  46:43


Stephen Salter  46:44

So we’re free once once you’re up there.

Blaz Gasparini  46:47

Yeah, I mean,

Metta Spencer  46:48

You don’t have any use for for drones. though do you? You’re, you’re hoping to just spray things from from sea level.

Stephen Salter  46:57

My stuff goes from ships, yes, and it’s moved up in just to the bottom of the atmosphere to the troposphere. Because of there’s there’s turbulence caused by the wind blowing over the water, and we can’t get it nearly as high as the, the cirrus clouds. The, we get a fairly even mixture for the bottom, maybe 1000 to 2000 meters. But it’s much harder to get it above that, because there’s much less turbulent.

Metta Spencer  47:30

Well now earlier on, we were kind of joking about using your wafers, your nozzles for both purposes, but taking it seriously. How it sounds as if maybe that’s not a bad idea that that you’re going to need to disperse this stuff in such small amounts and so thinly, that it might be comparable to the kind of concentration that Stephens nozzles would be aiming toward. Am I making stuff up something up? Or is that uh,…

Stephen Salter  48:05

Well, the, the concentrations in the places that we’d like to work are about 50, 10 to 50, maybe 100 per cubic centimeter. And what we want to do is to double the number there to get a contrast change of about five or 6%. So doubling the number of drops in a cloud gives you five or 6% increase in reflectivity. And if you want to get more than that every doubling is twice as much work, about twice as much effort, twice as much cost. So we want to go places where there’s initially a very small concentration perhaps in air that’s been over the Antarctic for a long time and has had snow cleaning it up. But the, I think I think the manufacturer of the aerosols would be very different. I don’t know the first thing about bismuth tri-iodide, but I’m I’m much happier to work with a liquid going in. Is bismuth presumably bismuth is a solid, is it?

David Mitchell  49:19

Correct? Yes.

Stephen Salter  49:22

And I think I’d like to work with either a gas or, or a liquid.

Metta Spencer  49:28

So there’s no percentage and trying to get us use the same equipment year round?

Stephen Salter  49:34

Yeah we would only  use it for the observation. Making fine powders is is really a lot harder than pumping liquids through nozzles. And I think a lot of fine powders will want to stick together they clump together.

David Mitchell  49:54

They disperse them with pyrotechnic flares. That’s how cloud seeding Enterprise has done it in the past.

Stephen Salter  50:02


Metta Spencer  50:04

Say it again explain what’s that, I don’t get it?

Stephen Salter  50:06

Well it’s a bit like a firework. We did an experiment with that. Discovery Channel did this. and it worked amazingly well for producing clouds from a completely clear sky. And it’s used for rain making, and that works if you can choose the right times and places to true, which is difficult. But if you do if you get it right, it’ll certainly he increase rainfall.

Metta Spencer  50:37

Okay, well, how much do you think let’s say, we wanted to do a very cautious experiment with cloud thinning to just I don’t know how much information you would get from a very cautious experiment. But you know, only from the state standpoint of not wanting to overdo it and cause bad effects. I would assume you want to go go slow and start off small. But if you did the small project, and you know, really starting in the next few years, would this be expensive? Or is it something that administratively could be run by the same staff of people who run the cloud brightening and possibly we might even want to experiment with the, with Franz Oeste’s iron salt aerosol, or titanium aerosol project, spraying in the Arctic? It would it, would it be compatible,and how expensive? Do you imagine a a modest little experiment would, would, how expensive would it be? And how much information could you get by doing it if you started with the basic knowledge that you have right now?

David Mitchell  52:10

Well, based on this, these notes I made about 10 years ago, based on the cost of bismuth tri-iodide back then. There was a seeding all the latitudes poleward of 45 degrees was 20 million per year. I’m sorry, 6 million per year.

Blaz Gasparini  52:31

Okay, sorry David this is not very close to zero. What do you say it’s a it’s an idealized scenario with which we play. It’s it’s a back of the envelope calculation?

David Mitchell  52:42


Blaz Gasparini  52:44

The public doesn’t even need to know about that. I mean, this is very nice. You put that out. But but but just, just just to make it clear, we were talking about modifying one of the most uncertain components of the climate systems. And there’s a lot of things that we need to do before we go and seriously do that bigger scale. I believe, however, that we may get some information from a very small scale experiment, which, in which we would maybe inject in a controlled sort of environment.

Stephen Salter  53:21

A cloud chamber, yeah.

Blaz Gasparini  53:22

Oh, yeah, cloud chamber experiments first, but then even if you would go in the field, you could maybe think of injecting some of this dust, aerosols or whatever, aerosols and see what happens to the ice nucleation. But I believe we’re talking here about scales, which are way below the scales where that would have any measurable climate impact. We’re really talking about the fundamental processes, how these clouds form, which, unfortunately, are not known well enough. And another thing, marine cloud brightening or Stratospheric Aerosol injections, they both of these methods have natural analogons that we can study. Ship tracks for marine cloud brightening. Well, that’s not a natural analogon it’s a human made analogon, but we really could learn a lot from, from from that type of perturbations, while with Cirrus we are unfortunately, not really having any, like, natural, clear natural analogon or  way when which nature would would be clearly changing serious properties. I mean, we do have something but it’s a bit of a messy business.

David Mitchell  54:36

Yeah. Well, one thing…

Blaz Gasparini  54:37

We can say that we have to be cautious when talking about that it’s a super exciting topic. It’s so uncertain. That’s why we are I think here, and there’s so much still that we need to learn before thinking about the cooling effect of that. I mean, we think about our cooling effect thing in our scientific meetings, and it’s correct and we should do that. But just from more of a broader public policymaking aspects we were far from from from these methods to be, you know, really in the portfolio of ways how to tackle our climate crisis for now.

Metta Spencer  55:23

I get it and I, I’m, you’re gonna have to forgive me, because I’m not a scientist. I am an activist, and I am somebody who’s scared to death about climate change. And when people say we’re in an emergency, they talk about the catastrophic effects that we’re all going to be suffering, in fact, we’re suffering from right now. Excuse me, I live in Canada, where we’re having wildfires caused by global warming. And it doesn’t take much to reinforce the reality that there’s going to be even more catastrophe coming our way unless we do some serious changes right away, which to me means we, we do not have the time available, that a scientist would normally require in order to proceed through all of the logical steps and take the cautious, careful approach that that you all take. I think if you’re in an emergency, you throw everything you got at it, and see what works. And and so I’m much more ready to go out there and try doing something with, if there’s any chance that it will have any value. So that’s why I keep saying, you know, can we, could, could you learn anything of value within five years, if we, if, let’s say the Canadian government wanted to fund research project in the Canadian north?

Stephen Salter  56:52

Yeah. So if you, if you don’t have enough money, and you don’t have enough time, you’re very likely to fail, and once you’ve failed, it’s much, much harder to rescue something. It’s the whole lollipop in the gutter, nobody picks them up out of the gutter. So this is why we’re reluctant to do something which has a high probability of failing, we want we want to be confident that there’s a reasonable chance that it’s going to work before we…

Metta Spencer  57:21

Within a few weeks, I’m going to be writing a report to the Canadian Pugwash group, saying whether I think based on the conversations we’ve had so far, whether I think that we are ready to reasonably recommend to the Canadian government, that they look into seriously investing a number of billion, million dollars, I mean, like 30 for you and another six per year for them. And who knows what for the iron salt aerosol experimentation, that they start investing money in practical, realistic, experimentation. and at this point, I am going to have to say no, it looks like everybody is, all the other scientists are saying we’re not ready yet. But what I am inclined to do is say to the Canadian group, no, they’re not ready yet. But let’s keep this thing open and continue the conversations, because in X number of years, they think there’ll be ready. Now you tell me how many X number of years it will take you before you feel that such a practical experiment might be worthwhile for the Canadian government to fund.

Blaz Gasparini  58:41

I think it would be worthwhile if the Canadian government would fund some field campaigns, which would increase our knowledge on how these clouds form to start with that actually costs a lot of money. I’m personally working with climate models. That’s my expertise. But I do get extremely frustrated at times, because we don’t have enough information to constrain our models with. We may have some information, David is working on that, but that is not enough. And I guess with a budget of 10 million per year, we may be able to sample those clouds in a systematic way and learn a lot about that, and learn whether there’s potential that perturbing them would really yield some cooling. and then as a second step, you we could, you know, go and try to make some very small scale experiment if the previous results would indidcate.

Metta Spencer  59:51

You’ve upped the price tag, a minute ago it was going to be $6 million a year and now it’s 10. Okay, I’m well willing to consider making any kind of proposal that, that would actually appeal to you guys that you really think this would be of value? My only one of my concerns is that I don’t know any Canadian scientists who are working in this area. I’ve actually been looking around and I don’t, I can’t find anybody. and you would need some Canadians because I don’t think they’re going to is going to just pay to have you know, you folks come from, from from Nevada and Vienna, and even from from Scotland, unless there’s some Canadian involved. So that’s one of the things, but if I could say, I mean, I left him a report, say, let’s continue our conversations for another three years, and evaluate them at the end of three years, whether or not they’re ready to do anything, or in the meantime, the Canadian government should be spending some money on doing this kind of study, which would inform them and provide the kind of knowledge that will speed this whole thing up. So what would that be? You say you don’t have enough basic knowledge? Is Canada in a position to generate that knowledge?

David Mitchell  1:01:20

Yeah, one conversation. There was a workshop at that, where I first met Blaz the Institute for Advanced sustainability studies in Potsdam, Germany, where we were talking about possible field campaigns, you remember Blaz? And Andy Parker was in charge of that and there are, I’ve been helped. I’ve been collaborating with some grad students at ETH Zurich. On this, where we, one of the documents that we came up with was a way to test this idea from, with satellites. There are clean cirrus clouds over the Rocky Mountains. We see that with the satellite retrievals that shows plenty of evidence that there’s these clean cirrus are homogeneously nucleated Cirrus over the Rocky Mountains.

Metta Spencer  1:02:17

Sorry, including the Canadian Rockies?

David Mitchell  1:02:20

Yes, including that. In fact, they’re more over the Canadian Rockies than the United States Rockies.

Metta Spencer  1:02:27


David Mitchell  1:02:28

And there’s dust storms in Asia like over the Takla Makan Desert and Gobi Desert and so on Mongolia. So that dust goes into the upper troposphere where the cirrus clouds live a lot more effectively than the Saharan dust storms. So these Asian dust storms kind of ride the conveyor belt of the cyclones that are generated over the Pacific and so on. They’re elevated to high levels and where the cirrus clouds can receive them over the Rockies. So if you pick out a dust storm, that’s naturally seeding that dirty. This is the dust particles that we talked about earlier. So they’re going to be seeding the clean cirrus over the Rockies. And you should see a change and the physical properties of the cirrus clouds when that dust plume arrives. That’s something you could do without any field campaign, really. That’s just Mother Nature doing her normal work and you’re just monitoring what nature’s already doing. So there’s no controversy about this, like there has been in the past.

Metta Spencer  1:03:48

It also sounds like it might be cheaper.

David Mitchell  1:03:49

Yeah, a lot cheaper, you just need to develop a satellite method for doing this. We already have ideas on that you did it to use the geostationary satellite. But the new [GHOSt] satellites have a channel at 13 microns that would should allow you to do this.

Metta Spencer  1:04:07

Okay, do you know anybody in Canada who would be interested in doing it?

David Mitchell  1:04:11

There should be I don’t really know, the satellite remote sensing people in Canada. But I’m sure there must be somebody out there.

Metta Spencer  1:04:19

Yeah. But there would not be the same people who are doing the climate cloud studies normally.

David Mitchell  1:04:27

I don’t know.

Metta Spencer  1:04:28

Yeah, okay. Let me now

Stephen Salter  1:04:30

You accumulate a large number of satellite images, you can detect a very, very small change in contrast, which is what is all that you need. So it’s a it’s a, I’m sure that’s the right thing to do. Anything you can do to collect the materials that are in the cut to different kinds of cirrus clouds would be would be valuable. And then I think trying to do a test tube experiment with a tiny quantity of the chemicals that you controlled, will tell you a great deal as well. Have you, people do very well with chemistry and test tubes rather that they’re in charge of what went into the test tube. If you’ve got 9 billion people doing their own experiments in different test tubes and not telling you what they’re doing, it’s a lot harder. So test tubes, like Cloud chambers are what you want. For that side of things.

Metta Spencer  1:05:26

Is there any connection between what he’s just proposed about satellites, and any information that would be of value to you, Stephen?

Stephen Salter  1:05:36

I certainly would like to get access to satellites for measuring cloud contrast, but I think this is so easy already. Because we can see the ship tracks, and I don’t think I’d need any special satellite information. Something that can tell me about the size of aerosol would be interesting. And I picked up my ears when you mentioned that. The size, size of aerosol that we want to do is very important, we got to get the size bang on, right, because we know that if it’s too small, we work in Norway so that if it’s too small, it can heat. Or if it’s too big, it can clear clouds, and therefore, perhaps by making rain, and that will work in the wrong direction as well. So we want to be I think the best size is about 10 to minus 14 grams of sea salt, per aerosal particle. That’s the target at the moment.

Metta Spencer  1:06:40

Is there any value in exchanging information that it let’s assume that both of you or get some way of getting the Canadian government to do something with satellites, and with collecting samples and so on?

Stephen Salter  1:06:55

I hope there would be but at the moment, I can’t see anything except for the satellite observations.

Metta Spencer  1:07:01


Stephen Salter  1:07:01

As I think it would be very, very, very similar.

Metta Spencer  1:07:04

Now, from what you know, about the iron salt aerosol project, which I think we have to take seriously too, and of course, I didn’t try to get them today. But we could do it another show in which we put everybody together and see what, what you gain from working, you know, doing your studies, sort of under the same roof. We could do that, and I just simply am now at the point where I have to decide what I’m going to suggest to the Canadian Pugwash group. And I think I have to say that the scientists are not ready to do any experimental, any, at least costly experimental work on the ground, or in actually flying through the clouds yet. But, but we want to continue these conversations, for another year or two, and see whether or not the state of knowledge is, has advanced enough to be able to seriously recommend a specific project that Canada should fund. Is that reasonable?

David Mitchell  1:08:22

With that satellite experiment I was talking about? Having a aircraft in the clouds where the satellite was looking would be very helpful, I think. So you could have aircraft involved it just wouldn’t be doing any cloud seeding. It would just be measuring what’s already there in the cloud.

Metta Spencer  1:08:44

Would drones do work for that?

David Mitchell  1:08:47

Possibly, yeah.

Blaz Gasparini  1:08:50

It’s harder than lower lying clouds. But there’s actually another suggestion, there was a very recent publication, which of course, just one, who knows if it makes sense or not. But in principle, since you guys are thinking about marine cloud brightening, in winter time, as you as you mentioned already, you actually want less of those clouds. And a lot of those clouds in wintertime over the Arctic are actually formed from well are sort of mixed phase so are partially formed by ICE partially by liquid, or many actually still formed by liquid but liquid at cold temperatures, temperatures below way below freezing. There’s an idea of how you could get rid of those clouds in winter, which is actually what you want to get because in winter also those clouds sort of warm the climate, or like a blanket, and you could do that by maybe putting some of these same ice nucleating particles that we talked about before, but just put them in these clouds, which are, you know, at altitudes below two kilometers, basically the same clouds that you’re trying to tackle in summer, just that under colder winter conditions, they will behave differently. And they will behave differently to perturbations from such aerosols, and you may get something out of them.

Stephen Salter  1:10:28

I did a paper about that, which I submitted to Nature and Science and it was rejected by them, but I will send you a copy of it. Using it, it’s putting out very small aerosol smaller than we want, which might be taking, not forming a proper cloud drop, but could still take out liquid water vapor to prevent other clouds drops forming. If you have enough very, very small aerosol, he can remove a large mass of lick of liquid water, that’s too small to be a good cloud reflector. This is based on the observations from Norway by [inaudible] Stier and Christiansen that aerosol in the what’s called the Aitken mode would work in the wrong direction by removing water but not making a proper cloud drop. And the question is whether we can use the same technology to make much, much smaller aerosol. And I think the best way to do this might be to use the same hardware but to use, add freshwater to the salt water so that there’s less salt going up. And it’s the salt that’s doing the nucleation. So very shortly,…

Metta Spencer  1:12:00

Wait, Michael Diamond was on one of the shows earlier, and he sent me a reference to a paper about mixed phase cloud thinning, which it looks like it’s about the only thing that’s been done on that. So I didn’t think to include that in today’s conversation. Should I have done so is there something there that needs to be brought into this discussion?

Stephen Salter  1:12:30

Anything that will get rid of clouds in the winter would be good. It doesn’t have to be a very high up cloud. Clouds at any height we’ll be acting like, like the blanket in the winter. Or one of the worries about stratospheric aerosol is whether you can stop in time for the winter. If, if there’s any stratospheric sulfur around in the winter, it’ll be doing long wave reflection, and that’ll be bad, and the stress effect, people don’t like to be told about this. They’re claiming it’ll all be falling away, which has been a bit difficult to believe given that it’s being dropped off in the regions of the jet stream. And I think it’ll be you don’t really know where it’s going to be, it could be all over the place in the same hemisphere.

Blaz Gasparini  1:13:26

Metta, the paper you mentioned is the same paper that I had in mind. I’m also a co author in that one, and I think it’s an interesting and it’s a different idea from what Stephen mentioned. So So Stephens idea is also interesting, but is a different mechanism. So that mixed-based mechanism is really playing based on this so called Wagoner [inaudible] process where at temperatures below freezing, if you have a ice crystal, that ice crystal will grow on the, on the count of liquid water droplets. and then these ice crystal will become big and will basically deplete the cloud and fall out of the atmosphere. But this is very early stage research. So a lot of questions, I believe.

David Mitchell  1:14:19

Yeah. One, one thing Blaz about that, that we’ve been doing some radiation transfer calculations here and not for the Arctic. And when we’re looking at the cirrus cloud thinning effect, and so we get about five watts per square meter of cooling based on the set when we initialize this radiation transfer model with satellite information on cirrus cloud properties. But then when we add low cloud, it completely blocks out the cooling effect. So even if we have ice, liquid water content of point 01 grams per cubic meter.  That really low liquid water content is still for like, say a one kilometer thick cloud, low cloud, it still neutralizes all the cooling effect. So that’s something to consider when with this mix phase idea.

Stephen Salter  1:15:18

There’s an awful lot of cooling.

David Mitchell  1:15:21

Yeah, yes, we did look at the Arctic, and in the wintertime that literature seems to show that there are not a lot of low clouds around in the wintertime in the Arctic. So that was good news in a way. Maybe 10 to 30% cloud cover from low cloud.

Blaz Gasparini  1:15:39

But David you could also affect and dissolve those lower lying mixed phase clouds with sedimenting particles from Cirrus. So you may be able to actually get rid of those because imagine…

David Mitchell  1:15:53


Blaz Gasparini  1:15:54

Those sedimenting big ice crystals and they would fall in the lower lying mix face and then they would sort of freeze them and rain them out or snow them out of the atmosphere. So

David Mitchell  1:16:06

The Bergeron effect.

Blaz Gasparini  1:16:07

Exactly. There’s a ready actually this cirrus cloud thinning studied by Gruber et al, a German group modeling study were very funny. But their biggest relative impact from perturbing Cirrus was not in cirrus clouds themselves, but because they actually perturbed or removed those mixed phase clouds out of the atmosphere, that they got some substantial cooling effect.

David Mitchell  1:16:39

So the mixed phase was more powerful than the cirrus clouds?

Blaz Gasparini  1:16:42

Yeah, because Cirrus are just relatively thin. and their relative contribution, you know, is not super big, while mixed phase seem to be a bit bigger player in the [rate of] budget of the Earth, at least in wintertime, Arctic.

David Mitchell  1:17:03

Right, I read that paper too. Yeah, it was very interesting.

Metta Spencer  1:17:08

So let me say, ask how promising is this whole thing? Is this just very abstract theoretical, maybe, you know, in a million years something might have come of it? Or is this something that we should take seriously and say, look, we’ve got to cool the planet. and this is a realistic approach. And should everybody should do what we can to advance the state of knowledge about how to do it? I mean, I’m willing to take your advice of, of saying that it’s premature to go out and, and do any actual experimentation by spraying things around. But I mean, I’m assuming that that’s, you’re just not ready for that. But what can we do now and is it really worth doing?

David Mitchell  1:18:10

Well, my personal view, and it’s only my view, it’s not, I’m not representing the Desert Research Institute about this. But, I would definitely say it’s premature for any kind of deployment, you know, like actual trying to go operational and cool things down. But I would not say that a small scale field experiment would be a problem, because I don’t see how that would affect the climate. If it suddenly you’re doing some cloud seeding for, say, one or two weeks, over a very limited area. I don’t think that would be a problem for the climate system.

Metta Spencer  1:18:52

Would it be a value?

David Mitchell  1:18:54

Yeah we could learn something from it. Yeah.

Stephen Salter  1:18:56

I think one or two weeks isn’t long enough. I think you need to do it in lots of different meteorological conditions that you could do with a very small quantity.

Blaz Gasparini  1:19:07

Right? I think, Stephen, that’s a good point. You may want to do it for a season at least or come back a couple of years. Or two, three, and repeat that in winter time. Yeah.  Yeah.

David Mitchell  1:19:20

Okay. Yeah.

Metta Spencer  1:19:21

That’s gonna be easier and cheaper. I mean, how much? How much would it cost to do that? Ah,

David Mitchell  1:19:31

Uh, pretty much but I, go ahead…

Stephen Salter  1:19:34

I would think I checked the cost estimates, but it would be a few million dollars to operate a system for two years in a very wide range of places and meteorological conditions. So

David Mitchell  1:19:59

Yeah, I don’t know. really know how, how what the cost numbers would be. But it would be, I think comparable to a typical field campaign on cirrus clouds like Department of Energy does, from time to time.

Metta Spencer  1:20:16

Can you say that this is would be so valuable that because if it turns out that the information is generated, the whole notion of cirrus cloud or or mixed phase cloud thinning is an extremely promising way to solve global warming? Or is it just, you know, who knows?

David Mitchell  1:20:40

It’s undecided right now, we don’t have enough knowledge right now to say one way or the other. But we have done recent research, I think Blaz saw my talk at the EGU meeting. and we have some new methodology using satellite data for estimating the contribution of homogeneous ice nucleation. Or use, you could say clean cirrus clouds, how much of those clouds are out there. and we found that roughly about in wintertime about 22%, in the high, mid to high latitudes in the northern hemisphere, and maybe about 37% in the southern hemisphere mid to high latitudes are around. So that’s enough to do the cirrus cloud thinning idea, and the other factor is that these clouds that are so called Clean cirrus clouds, those have a stronger impact on radiation than the other types, the dirty cirrus clouds. And we know that by the optical depth of the, or the extinction coefficients and the optical depth of the clouds. So these clouds that are the clean cirrus are optically thick clouds, and they have a bigger impact on the radiation balance. So even though they might be only say 25% of the overall cirrus population, their radiative impact might be more than 50%. It could be 70%, or something like that, because there’s so optically thick. So I think there’s a tremendous amount of new research that needs to be done on this. But it doesn’t look like a showstopper. Like there’s some reason that we should not do more research. I think the research looks promising.

Metta Spencer  1:22:40

Can you think of any, you know, some of the early papers, I think were ambivalent about whether it might have some harmful side effects. I mean, you know, I’ve heard people speculate that if you if you do certain things to the clouds in the Arctic, it might cause a monsoon floods in Pakistan or something is can you foresee any realistic reason for apprehension?

David Mitchell  1:23:09

There certainly could be some unwanted side effects. But that’s why we need to do the climate modeling. The problem with the cirrus cloud idea, as we know, so little about the cirrus cloud behavior. The two different ice nucleation pathways, for example, those, that information is not in the climate models right now. So the climate models can’t really inform us very well, whether there’s going to be side effects or not, or if there are side effects, how severe they’re going to be or if they’re even present. So we basically, in my opinion, the climate modeling is way too premature to to really comment on that. Blaz might have another idea.

Blaz Gasparini  1:23:58

I agree with with David. I’m myself a climate modeler working with those big models that are, you know, used to project our future climate conditions. and those very commonly don’t have the level of detail needed to really reasonably well simulate this clouds and how these clouds form, how these clouds dissolve. There’s other types of models, which may be more appropriate for some of the questions that we’re asking here. But those models aren’t. Those models are really used more to understand how these clouds behave in a smaller scale environment. But not so much about the global impacts of put potential cirrus cloud cleaning deployment.

Metta Spencer  1:24:53

So I’m close to where overtime, you know, I had a lot, that I am, as I say, just a few weeks, if that, away from trying to have a report to the Canadian Pugwash group about what its position should be. And I’m, I’m inclined to say, nobody’s ready to do very much experimentally, but I am not sure whether even that is the case. At any rate, my, my inclination is to say, let us have a continuing series of conversations among people who are concerned about the climate in the Arctic, and, and technologies for reducing the crisis there. and and have and then another in a year, have another decision about whether the state of information is advanced enough so that we can recommend that the government undertake a specific project. Is that a reasonable, it tells me if that’s unreasonable, for me to say that?

Stephen Salter  1:26:14


Metta Spencer  1:26:15

It’s not reasonable?

Stephen Salter  1:26:18

Not unreasonable.

Metta Spencer  1:26:20

Not unreasonable. Good. Okay. David, what do you think?

David Mitchell  1:26:24

Same, not unreasonable.

Metta Spencer  1:26:28

Thank you.

Blaz Gasparini  1:26:29

Cirrus cloud research is really in its early stage, and it’s just hard to make any statement about geoengineering potential of these methods.

Stephen Salter  1:26:40

That’s anything that you measure could be very valuable. Things that you didn’t expect, would be very valuable turned out to be amazingly valuable, and to your surprise,

David Mitchell  1:26:53

And what we’ve been talking about is very important for climate modeling, climate science, in general, it’s not just restricted to climate intervention research. The climate models need to know the same information to predict a more accurate future.

Stephen Salter  1:27:11

Especially anything to do with the instrumentation that you’ll be needing would be valuable for other fields.

Metta Spencer  1:27:20

My impression is that this this area of research is extremely important and it can be the most, by far actually, the most promising of the four interventions that we have been investigating for the Pugwash group. But it’s the least advanced and and, and at the moment, I don’t think I can say that we should ask people to develop a particular research agenda or program that we should ask the Canadian government to fund because I don’t think that anybody’s far enough along to be able to say what that should look like. But that I will ask them to, to support the idea of continuing this kind of conversation for at least one more year. and, and then we’ll, we’ll have another assessment. Okay?

Stephen Salter  1:28:19


Metta Spencer  1:28:20

All right. Thank you. and it’s been fun, and I’ve enjoyed it very much and we will, I’m not losing you. I’m definitely interested in what you’re doing. So I’ll keep pestering you for more information as we move ahead. Okay, thank you, bye.



Select the Videos from Right

We produce several one-hour-long Zoom conversations each week about various aspects of six issues we address. You can watch them live and send a question to the speakers or watch the edited version later here or on our Youtube channel.