265 - How fires can lead to a nuclear winter? with Stephen Welch

Why Nuclear Winter Is Fire Science

Wojciech Węgrzyński 0:00

Hello, everybody. Welcome to the Fire Science Show. The topic of today's episode are the nuclear winters following fires caused by a nuclear exchange. That's kind of a depressing scenario to be considered, but well, world gives us reason to worry about those things once again. And actually, from the fire safety engineering perspective, this topic, uh, crossovers many, many interesting and relevant problems we would find in fire safety engineering every day. That's wind, that's fuel loads, that's compartment fires, that's soot production. All of those combined are important for the thing that we are considering today, which is, can the planet Earth face a catastrophical scenario following fires caused by a nuclear exchange? I did not come up with this topic on my own. I, it was the guest who proposed this, and the guest is Stephen Welch from University of Edinburgh. And at first I was surprised when Stephen told me, "Hey, we should perhaps talk about nuclear winters." I'm like, "Okay, fine. That sounds, that sounds interesting." But in fact, it is really in-depth, beautiful fire science combining so many different, uh, things from our discipline to find a solution to the question, are those still a threat, and how, how big the threat is? So I, I hope I've sparked some curiosity. My promise is that you'll find a lot of useful things in the episode. And, uh, if you would like to learn more, stay with us. Let's spin the intro and jump into the episode.

Sponsor Message From OFR Consultants

Wojciech Węgrzyński 2:06

The Fire Science Show podcast is brought to you in collaboration with OFR Consultants, a multi-award-winning independent consultancy dedicated to addressing fire safety challenges. OFR is the UK's leading fire risk consultancy that this year celebrates its 10th anniversary. As experts in fire engineering, they are fully committed to delivering preeminent expertise to protect people, property, and the environment. With over 30 chartered engineers and a team of fire researchers at their core, they continually explore the challenges that fire creates for their clients and society so that the best research, experience, and diligence can be applied for effective tailored solutions. In 2026, OFR will grow its team once again and is keen to hear from industry professionals who want to collaborate on fire safety features this year. Get in touch at ofrconsultants.com. And now back to the episode Hello,

Meet Stephen Welch And The Question

Wojciech Węgrzyński 3:07

everybody. I'm joined by Stephen Welch from University of Edinburgh. Hey, Stephen.

Stephen Welch 3:11

Hello, Wojtek.

Wojciech Węgrzyński 3:12

I have to start by giving you an apology. Uh, when you first came with the topic of, you know, nuclear winters and, and soot, et cetera, I thought, okay, that's, that's like I'm diehard Fallout fan, so I vibe with, with this. Uh, maybe that's not good, but, but yeah, fine. I've done episodes on researchers, uh, researching, uh, how monkeys discover fire. I research-- I had a, professor from Spain who did an, fiction novel on fire. Every now and then a nice non, uh, kind non-serious, I'm dropping air quotes in here, you know, topic is, is great to have in the podcast. And then I went into a, a rabbit hole of this stuff that you sent me. And man, this is like serious and this is actually very relevant for our engineering because to solve the problem, you... that's like hardcore fire science out there. So sorry, man. I, I, I belittled it, but, uh, now I'm gonna, uh, go for, uh, atonement of some sort

Stephen Welch 4:07

Thank you, Wojtek. Yes, it's, it's a serious topic. Yeah, it, it scares people somewhat sometimes when you mention it, but, it's important as well, so that's why I kinda got em- uh, embed- embroiled in it. And, um, having learnt a bit about it, uh, I'm quite keen to tell other people, about the relevance and, and the importance of it. So, um, thank- thanks for the invite

Wojciech Węgrzyński 4:27

Yeah, super. I'm, I'm actually happy to talk about it. I've also, you know, I've learned, uh, some of this, uh, nuclear research when I was, uh, writing my wind and fire reviews.

Stephen Welch 4:38

Oh

Wojciech Węgrzyński 4:38

Uh, yeah, because, because the, the wind effects on fires, a lot of that was studied in the '40s, in the '50s,

Stephen Welch 4:46

Yeah

Cold War Research And Outdated Inputs

Wojciech Węgrzyński 4:47

strongly in relationship to nuclear fires. Like fires started by, by nuclear weapons. Like that was quite a recognized threat in the '60s. Uh, may- maybe we can briefly speak a little bit on, on the history o- of that. Did you, did you go into that going into your res-

Stephen Welch 5:05

I mean, ab-absolutely. I mean, this, this is one of the most fascinating things about the, the big topic. I mean, we're gonna come on in a minute or two to talk about firestorms, and the idea of the s- the storm is, uh, the wind that's, that generated by the fire. So it's, you, you've got your energy release coming from, from the fire, which will be orders of magnitude higher than any energy released by the nuclear weapon itself. So that's the first thing to realize. so, it's possible to have firestorms from, from various sources, including, uh, nuclear weapons. You could equally have them from, conventional, bom-bombing campaigns, and we'll maybe talk about some of the history o-of that as well. but yeah, you're absolutely right. Um, this was very topical back post-war, when we had the Cold War. Um, there was a lot of research being done and, n-not, not necessarily about the effect of the wind, but a-about fuel loads, which is kind of where I came into the problem.

Wojciech Węgrzyński 6:00

Mm-hmm.

Stephen Welch 6:00

almost everything that has been done on predicting the climatic impacts of these incidents is based on, very outdated historic data that were pr-pr-predominantly gathered in the '70s and '80s when the, the main kind of, concept of, of nuclear winter was kind of recognized and being propagated. Uh, and then I think towards the end of the Cold War, the in-interest kind of diminished a bit, maybe funding sources dried up a bit. and then there have been, unfortunately since, uh, reasons to, you know, get interested in it again. But, uh, yeah, I mean, w- the big question that we need to, to have a, a talk about later is, um, you know, where, where do you get your fuel loads and, and what's the impact of, of modern, uh, fuel loads on, on soot? And how does the soot then affect the climate at the end of the, th-this long chain? yes, it's very topical, always in the news these days,

Wojciech Węgrzyński 6:55

Yeah.

Stephen Welch 6:56

nuclear news.

Wojciech Węgrzyński 6:57

If you, if you had to frame the problem broadly speaking, what is it about? Like consequence of the nuclear explosion itself, the consequences of fires that started, or it's even broader about some, I don't know, uh, point where the, the, the whole planet goes out of balance? Like d- do you even

From MAD To Self-Assured Destruction

Wojciech Węgrzyński 7:18

consider-- W- which scale are we talking about?

Stephen Welch 7:20

it, it is a multifaceted problem, though ultimately it comes down to the climate impacts and the potential,

Wojciech Węgrzyński 7:26

Hmm.

Stephen Welch 7:27

you know, vulnerability of, of human existence on the planet, as a consequence of that. So,

Wojciech Węgrzyński 7:32

So that's the concept of nuclear winter

Stephen Welch 7:34

yeah, of, convecting soot into the stratosphere, which is the, the key. You know, we could talk about what the stratosphere and troposphere are, but, um, it's, it's high into the atmosphere, uh, and then it's distributed, globally. So it has global, uh, weather impacts and climate impacts. you know, some of the early models done in the '80s, uh, demonstrated the, the potential temperature could be of the order of 10, 20 degrees and, you know, the k- the kind of impacts could, could last for months or years. So i- it's, it potentially is a, is a threat to life on Earth. So it's, it's a, it's a critical thing to, to understand and, and explore and, you know, you could say we've been successful so far. There haven't been nu- nuclear exchanges. but i- is that partly because of, uh, the good work that's been done in, in, in raising awareness of, of this kind of concept? in terms of, uh, I think e-even as an engineer I can understand the, the basics of, nuclear, retaliation and, and, and how we need a, a balance. So, you know, y-you, you're not gonna attack someone if they'll attack you back, and we referred to this historically as, as MAD. This is mutually assured destruction. You attack, they'll attack you back and, you're destroyed. but nuclear winter, concept, tells us we need, uh, to, to renew that. We need to revise it. It's, uh, now gonna be we don't need the retaliation. It's SAD, self-assured destruction.

Wojciech Węgrzyński 9:03

Slabo šneka

Stephen Welch 9:04

once you've launched the, the missile, that's it, you know. I mean, uh, global, humanity, is in peril and, and might be wiped out regardless of, of, of whether there's anyone there who can retaliate.

Wojciech Węgrzyński 9:15

That's, that's, that's pretty sad actually. Uh

Stephen Welch 9:18

So yeah, no, I just, I just thought maybe, may-maybe it could be called "Mad or Sad" or something like that, but I think maybe just stick with "Nuclear Winter,"

Wojciech Węgrzyński 9:25

Uh, w-when I, when I've ever, ever thought about or contemplated nuclear winter and outcomes of, of, you know, nu- use of nuclear weapons, I didn't consider fire to be a primarily hazard in that, because

Stephen Welch 9:39

Yeah

Wojciech Węgrzyński 9:39

I, I know that there was like over 2,000 nuclear weapons detonated in the world. Like,

Stephen Welch 9:46

Eh?

Wojciech Węgrzyński 9:49

nuclear weapons since the first nuclear explosion in 1945. That's 2,000 nukes we bombed our planet already with.

Stephen Welch 9:56

Yes

Wojciech Węgrzyński 9:57

a lot of nukes. Like you would say, if you ask, ask a random person, "Oh, how many nukes you, you think we need to drop?" They're-- could give you like a ballpark number of 100, 200, 500, 1,000. Like we've already bom- shot like 2,000 of them and no winter yet. So, so indeed, it, it appears that it, it's considered more, more as a trigger in this rather than the main method of, of creating this climate consequence. So, so indeed the fire is the main threat in here

Stephen Welch 10:25

Yeah, I mean, obviously it depends, uh, those, uh, nuclear tests, I mean, not, not all of them were nuclear tests. There were a couple of examples of nuclear detonations. Uh, but, but the nuclear tests obviously would be planned to avoid the consequences of, developing

Fire As The Main Global Trigger

Stephen Welch 10:40

fires. But, yeah, we know from the Hiroshima and Nagasaki bombs that, the consequences in terms of fire were, were immense. Um, they, they, they generated firestorms. Uh, and as I said, you know, the, the energy release from, from the fire is, is the dominant thing. Uh, that's certainly in terms of the climate impact, where, where the energy, uh, is coming from. And, know, another, another, fact that struck me early on was, uh, when we're working all, all our lives in, in, in the, in the building fire kind of space, thinking about limited number of But typically in a, in a nuclear kind of fire scenario, you might be, uh, you might be talking about, say, 100,000 ignitions. So I know, yes, in, in, in wildfires, we have, spotting fires and embers and all the rest of it, so you have lots there potentially in, in, in, in wild- in, in the wildfire space as well. but, uh, uh, in, in terms of, you know, something like a, a nuclear exchange near an urban area this, this is the, the mechanism is rates of ignition of multiple, uh, sources, you know, both, both external to buildings and internal to buildings

Wojciech Węgrzyński 11:49

How much of those fires do we need to have to start triggering, uh, effects at the global scale? Or is it just one good fire that reaches high enough?

Firestorm Definitions And Megafire Evidence

Stephen Welch 12:00

Yeah. So this is, uh, the-- there's some historic data on the extent, of those kind of, um, firestorms that resulted during the war. So, um, we, we actually have a, a formal definition of, of a firestorm in terms of the min-min-minimum burned area. So this was due to, uh, Gladstone and Dollman back in 1977. They said you need at least, uh, one point three square kilometers. Typically, those firestorms in the war were, were significantly larger than that. they also said you need to have half of your structures, on fire simultaneously. So you can imagine that, um, over one point three kilometers at least, even, even, even at the small end, that's a lot of stuff that's burning. fire load densities, uh, comes in. Well, that's obviously one of the, the critical parameters we need to talk about later. they need to be of the order of four grams per square centimeter, so that's actually translates directly to forty kilograms per meter square, which is our, our office fire load. and ambient winds need to be, less than three point six. So, so the, the firestorm generates its own wind. Um, but yeah, obviously it's affected strongly also by weather conditions. So then in terms of how big does it need to be, I think, uh, this is again, why the problem has become again topical,

Wojciech Węgrzyński 13:18

Hmm?

Stephen Welch 13:19

in, in the last kinda decade or so. Because, uh, we now have, much more knowledge about the impacts of wildfires in terms of soot injection into the, in-into the stratosphere, and this is coming from satellite data. Uh, there's really fascinating stuff there, uh, analysis of the, the Canadian wildfires in 2019, um, Black, uh, Saturday Australian, fires and- the data we have there is that these immense, modern, we call them megafires, uh,

Wojciech Węgrzyński 13:50

Hmm

Stephen Welch 13:50

are injecting of the order sort of 0.5 to five teragrams of soot,

Wojciech Węgrzyński 13:55

Error grams

Stephen Welch 13:56

into the stratosphere. Uh, yeah, so we're, we're talking about, what is that? 500 tons or something of, carbon. So i- immense amounts of carbon getting up into the stratosphere. Now, with modern satellite technologies, of course, we can then track the, climate impacts and, and, and they, they've been measured. We, we, we already know that these kind of, uh, megafires can depress global temperatures by, I think, of ballpark numbers about point, 0.1 of a, a degree. So, not big, but or equally not, not nothing. So they, they, they are measurable. so yes, I, I think that, that's why it's, uh, also become very topical and, and it pro- it, it also provides a kind of, validation case. You see, you, you can, you can build your, your urban fire storm model, uh, and kind of, uh, do comparisons. Obviously, the fuel load density is completely different in, uh, typical wildfires to typical urban, fires. But, um, in, interesting comparisons to be done. So I'm not sure I've really answered your question as to how big it doesn't need to be. But yeah, uh,

Wojciech Węgrzyński 14:55

no,

Stephen Welch 14:56

the

Wojciech Węgrzyński 14:56

you gave a data point. Like if, a megafire can deliver already measurable decrease of the, of the global temperature, so like a single event, a very large event, but a single event that's already measurable. I assume in a scenario, uh, of, of some sort of, nuclear exchange, we're rather talking about multiple fires, at, at the same time, which adds to each other and probably it's not a simple addition, but the, the, the effects probably build up on each other, uh,

Stephen Welch 15:24

Uh,

Wojciech Węgrzyński 15:26

right, right there.

Stephen Welch 15:27

so certainly some of the, the outputs of these very detailed CFD-based models show that it's a nonlinear kind of thing.

Wojciech Węgrzyński 15:34

uh, does the height matter? so it's One is how much of fuel burned, which kind of drives your overall number of how many tons of stuff you've emitted, because that, that's a fuel calculation.

Stephen Welch 15:46

Uh-huh.

Wojciech Węgrzyński 15:47

But how far it went, my intuition tells me that's gonna be related to the intensity of the fire,

Stephen Welch 15:54

Oh, yes, absolutely. Yes. No, I mean, most of it doesn't get into the stratosphere.

Wojciech Węgrzyński 15:59

But,

Stephen Welch 15:59

that's the thing.

Wojciech Węgrzyński 15:59

but for the scenarios that you are considering, it's needed to be in the stratosphere.

Stephen Welch 16:03

It needs to be in the stratosphere

Wojciech Węgrzyński 16:04

so what was the deal with the, with the, with the layers?

Stephen Welch 16:07

um, competence awareness, yeah, I'll declare that I that's

Wojciech Węgrzyński 16:10

No

Stephen Welch 16:10

not my expertise, but this, this is kind of, I understand the basics about you need, you need to get it high enough. Uh, there's a lot of debate about what kind of particles, you know, is it carbon, carbon particles or, part- particles with, adhered hydrocarbons sulfurs and nitrates and all that kind of thing, which I'll leave for the, climate specialists. But yeah, the, the, the o- the other point I, I just wanted to mention before, uh, forgetting is, an- another data point is the Gulf War fires. So we, we, we know some of those fires were, multiple fires of, of order 300 megawatts. Um, but the distance between them was too great. So they, they were n- nowhere near, uh, what is required to produce, Which is, which is what you tend not to get with, with nuclear strikes, your fires tend to be concentrated. So that's, that's one of the big issues.

Wojciech Węgrzyński 17:00

The, the, there, there was also, uh, in terms of data points again, we also have some data from volcanic eruptions which, which inject a lot of stuff into upper layers of atmosphere.

Stephen Welch 17:09

they'd als- uh, and they also have climate impacts, uh, but the things, uh, yeah, ob- obviously the mechanisms in the up- upper atmosphere are, are different.

Wojciech Węgrzyński 17:17

Yeah, but it's just again, a single event. there was a year without the summer after,

Stephen Welch 17:21

yeah

Wojciech Węgrzyński 17:21

an explosion in Indonesia. So

Stephen Welch 17:24

Yes

Wojciech Węgrzyński 17:24

very, uh, obvious cl- climate impact, right?

Stephen Welch 17:27

Absolutely, yes

How Firestorms Generate Extreme Winds

Wojciech Węgrzyński 17:29

Okay, uh, tell me about the mechanisms of a firestorm, because I assume that firestorm is a condition which is necessary to reach this intensity that allows you to break the layers of atmosphere where the, where the suit becomes dangerous to the climber

Stephen Welch 17:43

So, um, I, I gave some of the, the definitions. I think they're fairly arbitrary, but it, it needs to

Wojciech Węgrzyński 17:48

One, three kilometer, uh, four grams per

Stephen Welch 17:51

the, the

Wojciech Węgrzyński 17:51

centimeter

Stephen Welch 17:51

historic definitions of, of what you're gonna call a firestorm. but the, I think the key is, is, is the wind. So, as well as having, potentially extremely, high numbers of ig- ignition sources, they generate their own climate. So the, the, the wind speeds in, in a firestorm might be the order of 100 meters per second. This kind of, you know, way above what we get in nor- normal weather patterns. So you can, in some senses, you can see it as a, as almost like a forced convection problem. I know it's a, it's a nat- it's a natural... It's driven by the, the energy release at the heart of the fire. but you're blowing, wind towards the center. You, you, you create a kind of mushroom cloud kind of, uh, scenario. so obviously one of the big questions there is, uh, how, how do those fires spread? it is difficult at the edge where you, you're, your, your fire spread is in, in the opposite direction of, of a strong wind. but bear in mind also that you've got your convection of, debris, embers, stuff, you know. You, uh, one, one of the things that, really struck me when I, I first got into this topic was seeing the video footage of the Durango fire, the Oakridge fire in Colorado, back in 2002. you can find some of that online. Uh, it's a forest fire, but it generated a, a tornado kind of conditions. Uh, and you can see big chunks of trees, you know, trees that are sort of several meters across that were, were ripped up, know, and, and they went up into the atmosphere in some kind of fashion. You, I mean, the video, the video is the, the examination of the, the debris after. And, and, and those, you enormous conifer, uh, trees were completely broken up. And that just kinda, uh, helps you to get a feel of, a- actually this is kind of outside our domain of experience. You know, we, we do fire experiments. We do, sometimes we do large fire experiments. We see, uh, large pool fires that produce a lot of heat. But we're talking about something that's kind of another level up and, and, and just kind of being able to appreciate the kind of severity of these fires is, I think, one of the big challenges in, in working in this topic. So yes, they're very intense. They, they generate their own wind. the whether, whether or not they can spread is, you know, depending upon the substrates, what kind of materials are burning. If they're in urban environments, you potentially, have, building collapse. So y- y- you potentially got that going alongside the ignition events and the burning fires. Uh, so some of the models have what's called rubblization parameters. Again, it's, it's quite hard for us to, to comprehend these dramatic events, uh, kind of outside our experience. You get pictures of, war damage, uh, y- maybe gives you a feeling for it. But, yeah, real- really drastic, uh, am- am- amazing kind of scenarios with, with intense energy release, um, and, and damage

Wojciech Węgrzyński 20:43

A- a- actually, there was one fire experiment where they did a firestorm. I don't know, did you, did you, did you seen that one?

Stephen Welch 20:50

Which one is it?

Wojciech Węgrzyński 20:51

In 19-- in 1980s, it's a paper called "Intense Atmospheric Vortices Associated with a Thousand Megawatt Fire."

Stephen Welch 20:58

Okay,

Wojciech Węgrzyński 20:58

And those mad lads, they've made a 1,000 megawatt fire just to s- to verify the hypothesis, can they make the atmosphere start spinning? And they j- j- just had, you know, dozens of giant diesel, powered burners,

Stephen Welch 21:14

Yeah

Wojciech Węgrzyński 21:16

and they just, like, fired them at the same time, and they measured the, the, the atmosphere started spinning around them. Like they, they've l-

Stephen Welch 21:23

fantastic. Yes

Wojciech Węgrzyński 21:24

literally made a 1,000 megawatt in fire experiment.

Stephen Welch 21:27

getting permission to do that now might be quite tricky

Wojciech Węgrzyński 21:30

No, I mean, it's like you, you would be burning like a hun- 100,000 quid per hour in diesel to, to make that even to start with. So we're talking about expensive research out there. But

Stephen Welch 21:41

Yeah

Wojciech Węgrzyński 21:41

it just, uh, helps to, investigate the hypothesis can, can the fire generate its own weather effects? And I also-- Uh, what you've described with the trees, et cetera, we had a large wildfire in Poland in, uh, it was in 1991. I know some, uh, people who, who attended that fire. They told me it was like a, like literally, literally a hurricane. And you, you-- There are pictures after the fire where you see giant trees or j- okay, large trees which are snapped in half and not burnt. So it's, it's-- They didn't break because of being burnt, they broke because the wind broke them, and they were very large trees. So, uh, in this setting, it creates a very, um, challenging, uh, wind pattern, and we know that largest fires are associated with very strong wind.

Stephen Welch 22:26

Absolutely, yes

Wojciech Węgrzyński 22:28

if this is a natural event, a fire growing to, to such an, a thing, then what you mentioned about spread, ability of the fire to spread upwind is probably very relevant because it eventually, must reach a point at which it cannot spread upwind anymore because the wind it created is too...

Stephen Welch 22:47

Yeah.

Wojciech Węgrzyński 22:47

So, so it's

Stephen Welch 22:48

I think

Wojciech Węgrzyński 22:48

self-confining. uh, but after a nuclear explosion or after a natural disaster like a earthquake, there was a, a great, Kanto fire after Tokyo earthquake in '23.

Stephen Welch 22:58

Indeed. Yeah

Wojciech Węgrzyński 23:00

also a fire whirl. in this case, the ignitions already have happened. So this kind of... You already have the fire on a very huge chunk of the terrain with probably very much destroyed urban environment, which is now ready to burn or already burning, and now you inject a massive wind into that kind of, you know, gathering all the energy into a central plume, which can go up.

Stephen Welch 23:27

that, that, that's absolutely true. Um, but I think a lot of the time maybe the spread is, is a minor aspect of the problem. So, well, uh, as engineers we have to always c-come up with kind of ways of dealing with uncertainties and, and probabilities. Uh, so I think, y-you know, we, we don't know the size of the nuclear exchange, you don't know the extent of the exchange, you don't know the context of the, the urban environments, the, the shape of the landscape, topographies, all that kind of thing. So, yeah, I mean, there, there certainly is an argument that, that's, that some of the fires will not be of interest. They'll, they'll burn themselves out. I, I guess you de-de-decouple those. and yeah, may-may-maybe when you're going into those probabilities, yeah, may-maybe there's something there that says, well, what is the potential for any particular fire then to go on uh, of significance for, for a nuclear winter kind of, uh, context?

Modern Fuel Loads And Dense Cities

Wojciech Węgrzyński 24:18

Okay. Le-let's talk about the, the, the fuel problem. two things. One, in the real nuclear, uh, explosions in cities, Hiroshima, Nagasaki, we're talking about 1940s architecture, largely timber, but also not a huge cities for the modern scale. and, and, and Nagasaki is also in mountains, so it it's like very specific, uh, terrain as well.

Stephen Welch 24:41

Yeah

Wojciech Węgrzyński 24:43

then you say 1970s, '80s, most of the work that was, uh, focused on, quantifying the fuels, et cetera. W-when you say that, you know what I see? I see a FSRI video, grandmother couch versus modern plastic couch and how they burn.

Stephen Welch 24:59

Yeah

Wojciech Węgrzyński 24:59

and this is your 1980s data point that, that you have. So how do you even approach the topic of estimating what could be the fuel load today? and, w- where did you get trying to, to figure out that number? Because I know you tried.

Stephen Welch 25:13

Well, yeah. I mean, we, we-- the, the-- I should give credit to the latest project student who, who created a transparent, scalable kind of, architecture for, for doing this. Uh, I wouldn't cl-- He, he, he himself wouldn't claim that he's computed that number. So it's, it's a mechanism, uh, into which you can feed various inputs and then study the sensitivity of the output. So I think the interesting thing here is it's, uh, you're absolutely right. A lot of the historic, data from the scenarios and from the data gathering in the '80s, is, st-strongly, cellulosic heavy. So that, that's kind of, there have been environments that we had. Um, I mean, there are, there are other materials there as well. and sometimes it's consideration of things like fuel storage or, you know, when, when you've had a, a nuclear, weapon strike, you'll have release of gas and other, other fuels and all that, needs to be patched in as well. so the the data we have about the types of fuel, is, is obviously outdated, but it then impacts potentially, uh, we can update that, but it impacts the problem in two ways. So one is, uh, how will the fires develop? and then you, you take your compartment fire as a kind of microcosm of the problem and potentially multiply it by times to scale it up into, well, this is an urban fire. will that spread or can we study it at the compartment level? What about glazing failure? And we can talk about all of those things. So, the, the nature of the fuel is, is obviously very important because it affects the, the fire development, the potential for the fire to interact with the glazing and cause, uh, y- a more, uh, open, over-ventilated, uh, condition. so, of the interesting things that Ab Hass, uh, discovered in his work was, just how significant the under-ventilated contribution could be.

Wojciech Węgrzyński 26:57

Mm.

Stephen Welch 26:58

Okay. So that's, you know, we, we can talk about knowledge of, uh, soot, which is a parallel, uh, stream of, of, of interest, and we can get soot yields from tourism. We use these kind of, uh, soot yields in our, in our fire models, or we can try and predict, uh, soot in, in fires. but often if you go to an under-ventilated fire context, you, you might be producing, lot more, even an order of magnitude more. Um, you could see extremely high, uh, measurements of, of, of soot in, uh, under-ventilated compartment fires. I think you're probably familiar with some of those, kind of sources, so that's what strikes me as being of interest. And, and actually, I, I got into this problem because, there was a, a Future of Life Institute call for, for new research on, uh, nuclear winter, which came into our geosciences expert. So, so one of my colleagues here at University of Edinburgh, Simon Tett, who, done work in the impacts of, sulfur aerosols on, warming, for example. Uh, he realized he needed an e- uh, an expert, well, uh, someone who knew something about soot produced by fire. So along to me and said, "Well, w-what can you tell me about what kind of soot, uh, it is that we might get?" And, and, uh, you immediately realize this is a very complex, um, and interesting problem.

Wojciech Węgrzyński 28:13

Hmm

Stephen Welch 28:13

So yeah, that, that's kind of how I, I first got into it, and and, and this is the, the, the kind of angle that, uh, ARP has pursued is, is you, you can't just take yield data. You know, it's not, it's not a building fire model of, of, of a

Wojciech Węgrzyński 28:25

Yeah

Stephen Welch 28:26

fire where you can just use yields. It's com- you're interested in the soot for a different reason than, than visibility, where maybe, you know, uh, once you get past a certain amount, that it doesn't matter

Wojciech Węgrzyński 28:37

Nezmazateľné

Stephen Welch 28:37

Uh, here it does. Ab-absolutely. You need to know exactly, um, how much soot you're generating and,

Wojciech Węgrzyński 28:43

Uh, we-we-we'll, we'll come back to soot just in a second just to close the loop on the, on the fuels.

Stephen Welch 28:48

Yes

Wojciech Węgrzyński 28:49

also the-- compared to the, uh, '40s, '60s, 70s, the modern cities are much more dense in... Like i-if your vessel is the compartment for your compartment fire inside, you have an awfully lot more compartments per square kilometer of your city right now.

Stephen Welch 29:06

Yes

Wojciech Węgrzyński 29:07

that most likely is a quantifiable number of how many-- Like that, that, that's a, broadly speaking, a, a first rough assumption how many compartments you have. Assume like, 10 megawatts per compartment and just, uh, how big a fire you would get in a city. Like that, that's most gonna yield you thousands and thousands of megawatts with just a block or two of, city of flats that burn, right?

Stephen Welch 29:31

Absolutely. And, I mean, there's, there's parallel work going on using things like digital twins. Y-y-you know, we, we, we have modern cities, uh, some, some of them are very, three-dimensional models to very high resolution. So you're, you're right. We know, we know that kind of stuff. I think there's even, global, uh, kind of mappings, three, three-dimensional urban mappings available now. It's probably started at the city level and has expanded. Uh, they, they might have lower resolution, but you're right, absolutely right, uh, uh, Wytech. Um, in, in high-rise, you know, the, the problem is, is, is a very different problem. You're, you know, what's your 40 kilos per meter squared? How meaningful is it when you've got 50 stories? And, what kind of fires might you get? you I think you're, you're now realizing this is actually a, pretty nasty, tricky problem for an engineer. You know, what, what, what can you do to describe that? You, you obviously have to make simplifications about your ignition, and simplifications about is there gonna be fire spread or is it ignition of every compartment? How likely is that? uh, so this was our kind of motive for We had, had a vision to, to kind of create an engineering framework for doing this. Um, actually finding realistic numbers for each part of that is, is, is a big challenge. It's, it's something that will, will go on for a number of years, to just de- deal with the kind of big picture, the geometry part of the problem before you get to how, how the fire burns and, and, and what kind of soot it, uh, produces.

Wojciech Węgrzyński 30:58

Th- Though it also reassembles to some extent the post-earthquake fires because you will also, uh, deal with the damaged building stock, I assume

Stephen Welch 31:06

Yes. Yeah. But, to what extent is, is that damage caused by the, the nuclear, strike, uh, and to what extent is it, consequence of the fire development? So, um, so A- Apas's, uh, framework that he developed for his IMFC thesis project was, um, kind of two-stage model. So you, you had a pre-glazing failure, uh, model, which is often dominated by smoldering. You- energy release rate is low, but you might produce, concentrations of soot. Uh, the, the volume flow rate matters. and then a post, post-glazing failure when you're, you've, you've got probably a split into well-ventilated and, and under-ventilated

Wojciech Węgrzyński 31:42

Fine. Uh,

Soot Modeling Versus Handbook Yields

Wojciech Węgrzyński 31:43

l- let's venture, through the suit and, uh, again, here your, uh, background, uh, plays a role, so, uh, l- you can introduce, uh, us to the glorious times of, early suit modeling of Stephen.

Stephen Welch 31:56

Yeah, thanks very much. I mean, th-this is one time the university systems worked 'cause you, you all have an academic profile and, and you, and you write there what you've done in the past. So, um,

Wojciech Węgrzyński 32:05

Follows you.

Stephen Welch 32:06

after sleeping silently for 25 years, I guess, a PhD in soot modeling in diesel engines actually, uh, got k-kicked in as being relevant. So, um, yeah, I mean, I start, I started my, my work on, um, uh, soot in, in, in measuring, uh, using, um, optical techniques on diesel exhaust. And, you know, a diesel engine is There, there's lots to learn from combustion systems, but in a diesel engine, in certain diesel engines, more than 50% of your fuel carbon will, will go through the soot, phase. Yeah. You, you produce soot, and then you burn the soot.

Wojciech Węgrzyński 32:37

Ah, okay

Stephen Welch 32:38

not the same in all diesel engines, but, um, yeah, it's, uh, quite extreme fires where we get up to that kind of level. So you make lots of soot in diesel engines. Uh, I'll tell you a funny story about my, uh, my MSC project. So I built this, um, optical system f- to put on the exhaust pipe of the diesel engine with a port and a, an air pur- quar-quartz windows and air purge to keep the soot de-deposition off the window. And, ran the, it was sort of, A Ricardo experimental, uh, engine, for, you know, just doing experimental studies against, you know, amount of, uh, s- diesel spray and, RPM and, and so on. And got the first results showing a lovely spike on each, uh, engine cycle and, went excitedly to my supervisor, who was, uh, Professor Moss at the, uh, Cranfield, uh, Group, uh, involved in soot and gas turbines, and, uh, said, "Look, here's my results, and very exciting, I seem to be measuring the he, he looked at it and said, "Hang on a minute. I- isn't, uh, every second cycle is the exhaust stroke? The, the, the other one is the intake stroke." And I said, "Oh, yes, you're right."

Wojciech Węgrzyński 33:42

Mm-hmm.

Stephen Welch 33:42

so what was happening obviously was the engine was vibrating and, and the, port was, cutting the laser beam. So,

Wojciech Węgrzyński 33:48

Oh.

Stephen Welch 33:49

I needed a bit of, needed a bit of redesign. But eventually got there and mea- mea- managed to measure, uh, soot in diesel engines and, and then after that, went much more into, into the modeling, And I think you're probably familiar with some of these models. The Moss, uh, Brooks model I think is still used. These are kind of two-parameter models where you, you, you model the, the number density of the carbon particles, and then the ma- the mass evolution and explicitly are capturing nucleation, coagulation where the particles l- lump together into quasi spherical particles.

Wojciech Węgrzyński 34:23

Mm-hmm.

Stephen Welch 34:24

Agglomeration where they kind of form chains. Uh, and then the surface growth, uh, that obviously overlaps with some of those. And then oxidation potentially is, uh, less kinetically controlled. It might be kinetically controlled in some conditions, but often is, is often turbulence controlled. So, yeah, there's, there's a lot of interesting stuff there from the combustion community, not just in diesel engines. I, I, I found, That's how I actually got into fire. So

Wojciech Węgrzyński 34:47

Uh, uh, giving

Stephen Welch 34:49

in a diesel engine, I found I could use the same modeling in a fire simulation.

Wojciech Węgrzyński 34:53

But give, give, give a new background. Uh, uh, for your usual fire safety engineer, they would probably never go that far as e-e-even Mossburg's, um, model. Yeah, be-be-because we are so yield-based in, in, in fire science, right?

Stephen Welch 35:09

Uh-huh

Wojciech Węgrzyński 35:10

so may-maybe you tell me where do the yields come from? You, you mentioned work of Towerson. I, take my yields from the handbook. You tell me where they got there, how they got there.

Stephen Welch 35:21

I take my yields also from the handbook, but we did, and, and obviously those are small scale, well-ventilated fires, uh, and we know the kind of sensitivity is, quite high, so up to things like acetylenes and foams, you c- up to about .2 or some, some, some very high values. Uh, but your typical, hydrocarbons are, are much lower and, um, in, back, back in, earlier, as well, I, I did some studies with these flamelet-based, uh, models, comparing them with yield-based models. So, um, had a paper on that in the fourth, IS, uh, f- Fire and Explosion Hazards, uh, series. and we created flamelets for soot for heptane, and we compared them with yield-based models. And heptane's quite a nice one 'cause it's kind of intermediate. It's kind of mid-range in, in sootiness. Uh, and we were able to demonstrate what kind of prescribed yields, uh, you could use to, to match the, the flamelet predictions, and then explore the, kind of scaling of that problem because obviously the, the timescale matters, um, when, when you're, w- when you're making soot and the timescale effects. So, um, I mean, that was, that, that's interesting, uh, work where you can, we can use more sophisticated models. Uh, you, you generally, in, in fires want your soot to, if, if you care about the radiative loss, you want your fire, your, your soot concentration to be correct where you've got your peak temperatures. You don't really care near the, near the, the fire source, so this, this is why, I mean, in FDS it doesn't matter too much if you inject a lot of soot at the fire source, 'cause it has no impact,

Wojciech Węgrzyński 36:55

it also has a correction mechanism for, uh, for radiation, uh, near the fire vicinity

Stephen Welch 36:59

You, you can use that, but you, you can also turn that off if you,

Wojciech Węgrzyński 37:03

Oh yeah,

Stephen Welch 37:03

So

Wojciech Węgrzyński 37:03

if you're, if you're adventurous.

Stephen Welch 37:05

Uh, but yeah, and then downstream, it, it doesn't matter so much, uh, for our problem. Obviously, if you're interested in visibility, that's a, that's a different scenario where the yield, you know, the yield curve

Wojciech Węgrzyński 37:15

in, in visibility, we've, shown with Gabriel that once you get beyond 0.1 soot yield, it, it becomes a vanity number because it's so dense smoke.

Stephen Welch 37:25

Yeah

Wojciech Węgrzyński 37:26

lose it so quickly in the relevant range. Of course, if you're interested,

Stephen Welch 37:30

Uh-huh

Wojciech Węgrzyński 37:30

interested in change of visibility be- between half a meter and a meter, there's a huge, uh, change in that. But if we're, ballpark interested in 10-meter-ish visibility in a compartment scale, there's like just, uh, you know, no change of the outcomes

Stephen Welch 37:47

Yes.

Wojciech Węgrzyński 37:48

after significant investment of time to find the correct number. Be-

Stephen Welch 37:52

no, I absolutely appreciate that, and I think this is why it's worth emphasizing that, the- these are parallel streams of activity and, and, and obviously the, the significance of the high SOOT yields

Wojciech Węgrzyński 38:01

Yeah

Stephen Welch 38:01

much more important in, in, in this environment. And maybe you don't care about it,

Wojciech Węgrzyński 38:05

The, the--

Stephen Welch 38:06

unfortunately for nuclear winter,

Wojciech Węgrzyński 38:08

Yeah. there, there's a med-medical like, practice type of thing that if a, a specific, you know, test or, or medical examination is not gonna inform the procedures to be done, you don't do it because it's pointless. Uh, so, so in, in kind of in the same way, like

Stephen Welch 38:24

Yeah

Wojciech Węgrzyński 38:25

if, even if you know that, ah, my Sutil is 012 or

Stephen Welch 38:30

Mm-hmm

Wojciech Węgrzyński 38:30

doesn't change the outcome. So you could, uh, you would, you could just save time by doing 01, and in the end you also put like this five megawatt number out of thin air, uh, in there as at the, at the start

Stephen Welch 38:41

yeah. Con-consistent level of crudeness. But, but, but that is all kind of assuming you're not in, in the under-ventilated fire

Wojciech Węgrzyński 38:47

Yeah, exactly. So, so that's what I-- where I wanted to go because we- we're so much, you know, in this yield, uh, mentality and the soot being a material property where I am absolutely convinced based on my, primitive observations of fires that I set up every now and then, that it's a very much, uh, fire condition parameter as well, because I'm not changing my fuel in the middle of my experiment, and boy, my smoke, uh, gets darker. so tell me about what's then happening in the under-ventilated fire?

Under-Ventilated Compartments Make More Soot

Stephen Welch 39:18

Yeah. So, um, the, the studies, um, the scientific studies on this problem in compartment fires are, are relatively modest, I think. So one of the main ones I refer to is the Hammons, um, I think it was a paper led by Koo at NIST, back in about 2009, in a reduced scale enclosure, two-fifths scale with, uh, heptane, toluene, polystyrene. and, uh, amazing levels of soot were generated in these compartments. So they talk-- I mean, you can talk about, um, soot volume fractions or, or soot yields, but, with the highest, polystyrene fuel loads, 350 kilowatts in, in, in those experiments, they were getting 0.65, soot volume fractions and point, yeah, 0.6, uh, soot yields. So

Wojciech Węgrzyński 40:05

So that, that's most of your fuel becoming soot kind of

Stephen Welch 40:09

Yeah, nearly all of it's become soot. Yeah. As, as-- similar really to a diesel engine there. So, um, if that then succeeds in evacuating the compartment, so those obviously are, um, the, the worst, uh, positions inside your compartment. Uh, and then you've got external burning. So that, that, that's the challenge then to compute how much of that is then, uh, consumed externally. So I, I'm, I'm not sure that that actually, uh, was part of, that particular study. And I think it's probably a bit of a neglected field in, in fire research. We're interested in what happens in a compartment. but actually for the nuclear, nuclear winter problem, uh, all we care about really, same as you with, with optical is, is what comes out at the end.

Wojciech Węgrzyński 40:51

Mm-hmm.

Stephen Welch 40:51

You know, what g- what rises into the atmosphere outside the compartment. So, yeah

Wojciech Węgrzyński 40:56

H-how do you get those numbers? W- I mean, you could resort to very large scale fire experiments, but we know those are, are inexpensive. So, do, do you model that? Do you use some simplified approaches?

Stephen Welch 41:07

Yeah, I, I think, the validated, uh, models that have been developed for s- the s- the, the two-equation models, potentially would be very useful for exploring this, this problem. you know, we're exploring conditions in, in the timberline compartment, you can see in great detail, you know, which bit of the fuel, from your burner or from your, your wood is, is where in the domain and then potentially then underpinning the, the formation of, of soot. So, you could definitely, run, computations of this, but the, the million-dollar question then is the, the validation, the trust in, in the outputs of those models. Um, I think it's a relatively unexplored topic. I could be wrong. Um, I know people have looked at large outdoor fires, pool fires, and so on, uh, measuring, optical measurements of, of, of soot in, in those. Uh, but here we're talking about under-ventilated compartment fires, um, and the external consequences of that. So yes, I think, a natural, approach would be to use, uh, models. So the other thing to say about those models, and it's something I explored in, in that early, fire explosion hazard paper that I mentioned, is, um, they tend to be for pure fuel. So you, you use kinetics CHEMKIN or something to compute, um, the underlying, chemistry, uh, in, in the-- against mixture fraction. And then on top of that, you build your soot model and, and you can't do that for, for your real building, which has plastic and, and cellulose all mixed together. So you you typically have those things from ethane, ethylene, you know, heptane, if you're lucky. And, you then have to start using some engineering corrections. So, certainly in the work that I did, we, we used the yield. So you looked at the yield and said, "Okay, so the whole thing needs to be scaled up or down by such and such." Uh, but you're obviously that, that's not gonna be, um, perfectly it might be useful, but how accurate it is an open question.

Wojciech Węgrzyński 42:58

W- which is an interesting kind of problem, right? Uh, that,

Stephen Welch 43:01

Interesting. Interesting

Wojciech Węgrzyński 43:02

mean, I had a similar comments about, investigations of, fire scenes, uh, from colleagues from UL when they were investigating the toxicity of fire scenes after a fire, and then recreating some of the stuff in the lab and the cons- and then the conclusion was, yeah, the lab is too clean. Like, we don't get, you know, the exact, you know, same kind of profiles out there.

Stephen Welch 43:23

Yeah

Wojciech Węgrzyński 43:23

and indeed, you know, our fuels are extremely simplified, but yeah, consistent level of crudeness as like you said. Like, if I go with five megawatt fire out of thin air, I can just drop 0.1 soot yield on that because that's the same level of, of consistent engineering assumption and, just drop it as an alpha T squared fire and be done with it. Uh, while in reality my bedroom is gonna be a very different fire from my kitchen in terms of the produced, materials.

Stephen Welch 43:50

Absolutely. Hmm.

Wojciech Węgrzyński 43:50

and then now you even get more exotic things into the mix with all the lithium-ion batteries, with PVs, with the in- insane amount of cables, et cetera. That's, that's like, that gets complicated and complicated

Stephen Welch 44:02

Yeah. So I mean, obviously plastics are gonna be dominant here, and that's I think where we need to focus. It's, um, you know, it's, it's not, you, you can use a, um, a weighted mixture, but, um, if the bit that matters is, is the plastic bit, then maybe the cellulose bits can, can be neglected or the uncertainties will m- be much lower. But that's not always the case. If it's an unventilated, uh, cellulose fire, obviously it also can produce a lot of soot. It, it all depends.

Wojciech Węgrzyński 44:29

I, but I, I've seen, experiments, uh, in which like a very little addition to heptane... I, I've, I've done it myself actually. Like, if you burn pure chemical heptane

Stephen Welch 44:40

Yes

Wojciech Węgrzyński 44:41

you just add a little tiny pinch of toluene into the mix, a little tiny pinch, the s- the amount of soot you're creating is changing tremendously. No, nowhere to the extent of those milligrams of or, or grams of toluene you've added.

Stephen Welch 44:54

Yeah.

Wojciech Węgrzyński 44:55

so

Stephen Welch 44:55

So

Wojciech Węgrzyński 44:56

it kind of is that, like, averaging you could lose these effects of, of those outstanding chemicals that they just change the, the whole mix tremendously

Stephen Welch 45:05

Yeah. No, I agree that that's a fact. I, I did think about the analogy with, with flashpoints. You know, if you introduce a very small amount of contaminant,

Wojciech Węgrzyński 45:13

Mm-hmm.

Stephen Welch 45:13

obviously it dominates your flashpoint and, it's not just, it's not just a linear correction. It's, it's the bit that, you know, is, is then going to influence the, the overall measurement that, that matters. And there's probably something similar going on with soot. Um, probably needs some more, detailed, combustion, experiments and to really explore this problem.

Wojciech Węgrzyński 45:37

It's, it's, it's very, very interesting. But now soot is kind of a living organism in, in a way, uh, because when it-- You already mentioned it, the, the physical phenomena when it's created, how the particles glue up together into form a ball, then the balls glue up together to form like a flakes. Uh,

Stephen Welch 45:55

Yes

Wojciech Węgrzyński 45:56

the, the, the, the structure, uh, gets more and more complex. It also acts kind of as a vehicle for other, uh,

Stratosphere, Aging, Rainout, And PyroCbs

Stephen Welch 46:05

It does,

Wojciech Węgrzyński 46:06

products that, that can land on it and then travel together with, with the flake of soot. But, uh, when, when it goes to the atmosphere, like, it stays there. Question is for how long and what actually happens to it in the atmosphere?

Stephen Welch 46:20

Yeah, we talk about, aging and thermal aging and, and different things, but you're, but you're absolutely right. The, the amount of p- pure black carbon that gets up in, into the upper atmosphere is, is a relatively minor proportion of it. It's most of the, the stuff, the particles, uh, have some kind of organic content as well. so, yeah. Uh, the, the, the, these are complex aerosols that, that obviously you're aware of the fact that you, you create these kind of nanospheres of, carbon. so PAH is, linked together and hydrogen disappears largely. So the kind of typical hydrogen content of that nanosphere might

Wojciech Węgrzyński 46:57

Carbon is s- carbon is so good in creating really funny shapes, right? What a, what an

Stephen Welch 47:02

precisely

Wojciech Węgrzyński 47:02

interesting, element

Stephen Welch 47:03

It is a, a wonderful element. and, and it, um, it, it creates these molecules that are very carbon, uh, heavy. So p- typically of a, a carbon, nanosphere is, is, uh, of a, of a soot particle is, is carbon. But then naturally it's in an environment where you've got these other things floating around, organics and PAHs that tend to absorb to the, to the surface. And, uh, that really massively complicates the problem because they have different optical, characteristics then. and, and therefore, different, different potential influences on, global warming. Um, from my understanding, as I said, I'm n- I'm not an expert in this area, but it's the carbon, uh, black particles that are the major contributor. They, they have most, influence on the forcing. Uh, so when they're combined with other things, that tends to mitigate that, that forcing possibly even ch- changes it to, to a negative in, in certain situations. But, but overall, we know that the, the carbon, contribution to this is the dominant one, and, and there will, We trust the, the, the climate expertise here, which is that it will, it will produce, um, cooling effects.

Wojciech Węgrzyński 48:08

I mean, Yeah, in, in terms of Lambert-Birlow, it's also like kind of exponential. So we're also not talking about one-meter wide layer in the atmosphere like your nice densitometer in a fire experiment. It's gonna be quite a big cloud, so it doesn't have to be that much carbon in it. Like you don't have to have, one gram per cubic meter, which would be awfully lot of soot, and you may find that in extreme wildfire. You could maybe have 0.001 gram per cubic meter, but the cloud is 500 meters, so that's, that's already enough

Stephen Welch 48:37

Absolutely. Yes. It's, it's, it's a, it's a direct function of the size, the,

Wojciech Węgrzyński 48:41

B-b-b-b-

Stephen Welch 48:42

and we're talking about probably several kilometers or tens of kilometers

Wojciech Węgrzyński 48:46

depth is such an interesting problem and also not very appreciated in, fire science, in relationship to, heat fluxes, et cetera. But, but that's a, that's a sidetrack. Uh, long does or how long could the-those soot particles live in the atmosphere? Because they're attacked by UV radiation, they're attacked by sun, uh, heat, there's a lot of stuff up there, oxygen there. They like to, oxidate, I assume.

Stephen Welch 49:11

Yeah. the, the climate modeling people do talk about these things, but from my understanding, the kind of timescales that the models look at are of the order of months, um, possibly up to

Wojciech Węgrzyński 49:21

Wow

Stephen Welch 49:22

if you go back to the, early, the, the, as I said, the 1980s work, uh, the, I think Turco was the first, to coin the f- the, the, the term nuclear winter and, and the e- early climate models that were done then. So they typically are on log scales of up to weeks, months, uh, where things begin to tail off, uh, and then potentially even up to years, depending upon the size of the nuclear exchange. So, what actually is happening to the soot particles up in the atmosphere, uh, during those months is, uh, I'm, I'm not sure anyone, anyone knows. Obviously, you've also got, potential, uh, the climate people talk about rainout. So how much of it actually g- gets there in the first place? What about the climatic conditions as the soot cloud is, uh, i- is, is ascending? so there's, a lot of complexity there. From my, my rough reading of, of these things, so, um, uh, moisture is very important there. There's, there's stuff about, uh, the, the self-lofting and, and the role of latent heat, um, and how that affects, potential for these clouds to rise. but I, I wouldn't claim any expertise in that. I think we need to, to take input from the, the climate specialists there.

Wojciech Węgrzyński 50:32

I mean, the, the validation points from the big wildfires as well give us that. You know, that smoke can venture thousands of kilometers and be persistent for a long time. The effects of wildfires are observed, uh, like for

Stephen Welch 50:45

I guess, I guess the, the, the big question there is we have so many wildfires, so what proportion of them get, uh, sort into the, uh, stratosphere. I had another project student who we've not actually talked about the role of pyrocumulonimbus clouds. So these, I, I should have mentioned this back at the start when you said, "What is a firestorm?" So, uh, the, the, the, the megafires are known to create these, these kind of clouds, and the nimbus bit is the storm, and the pyro is the, from the, from the fire. and, and they, they, they rise up high into the, in- into the atmosphere. so I had a project student looking at the, the, empirical models for the, correlation between, fire conditions and the generation of these pyrocumulonimbus clouds, in terms of fuel loads and moisture and wind and all that kind of thing. So there's, there's lots to be explored there. I think there's quite a lot of active research going on in this area because we have, of course, we have, better climate monitoring. We, we know more about these things in the atmosphere, uh, than we did historically. and also obviously uh, satellite imaging of, where, where fires are on, on, on the land. So I think it's a matter of correlating those, those bits of information.

City-Scale CFD And Assumption Fights

Wojciech Węgrzyński 51:59

Do you know any research or maybe you've tried yourself to, given all these frameworks you've built to, to quantifying effects of, such a fire happening in a, a really large city, I don't know, like London's

Stephen Welch 52:13

Uh, well, there has. As I, as I said, this is, it's become a very topical area in recent years. So I think the most interesting, contributions, um, that I'm familiar with in, in the literature are the, the work of, Rising Retail, uh, using the what's called, uh, f- the high grad fire tech models from, from Los A- Los Alamos. So these, these are basically, um, numerical simulations, um, CFD, uh, Navier-Stokes solutions. and these have been used to explore the consequences of assumed, uh, nuclear exchanges of, uh, maybe 115-kiloton atomic bombs. So, uh, the software itself, this is Los Alamos, as I, as, as I said. Um, you're typically using very large grid cells. So this is coupling landscape uh, scale fire processes to the atmosphere. So you... Maybe your, your grid is 20 kilometers and you've got 2,000 cells, so that makes about 10 meters si- kind of cell size. So solving Navier-Stokes o- on that level is, is clearly, it's mainly the climate. It's, it's how the, the, the, the, the stuff is moving into the atmosphere. It's not a, a model of the fire itself, but you need your fire source. So, as I mentioned, that's quite a topical area that Rising published a paper and then there were, there was a, uh, a series of exchanges and discussion about the assumptions that went into that paper, uh, that mainly focused on, the fuel loads. Uh, so this, this, this is, uh, Alan Robock was, was, uh, another of the pioneers of, of this kind of area of research. you know, he, he, he looked at these model outputs and, questioned some of the assumptions, and there, there were some revisions and iterations on, on the model predictions. Uh, which is all very interesting, fascinating stuff because it directly ties then into the kind of things that we're interested in. You, you can't just assume, values for your, for your fuel load densities from given scenarios. th- this model application was for, um, a, densely populated city in India or Pakistan, so it's a potential, uh, Indian subcontinent nuclear exchange. So, obviously those cities, mainly concrete, so maybe quite different from some other sources of fuel load.

Wojciech Węgrzyński 54:26

Is it? Like, at the same time you will get a lot more un-destroyed compartments to give you your under-ventilated fires.

Stephen Welch 54:35

Yes

Wojciech Węgrzyński 54:36

it's such an and understudied area, and I mean, for reasons we don't have those fires. Like, the fire science kind of follows, you know, what's happening around, and we're limited in people and resources, and it's kind of how we, we, we do things. But like to, to, to frame it correctly, you know, the fire science is a thing where everything happens like we thought it's gonna happen, and then we break some threshold we didn't know existed, and we have a disaster, you know? and I wonder, like, is there a, a disaster threshold point for a large city where our assumption that the whole city cannot burn at once, right, kind of breaks down and they actually can? Uh, and, and

Stephen Welch 55:17

Yeah.

Wojciech Węgrzyński 55:17

what happens then?

Stephen Welch 55:18

Yeah, I mean, this is, this, this was our kind of motivation to, to start working on developing an engineering framework, 'cause, you know, there's no point just talking about these problems and saying they're very interesting. Uh, you, you, you want to kind of create some tools to then kind of ask these kind of questions. But yeah, being realistic, it's, it's a long-term venture. We've, we've had maybe five project students so far away at the problem. Um, but it'll take a, uh, take a lot more detailed work to answer the question that, that you're, you're stating there, uh, Wojtek. So, actually, more than that, we've had other project students working on, for example, carbon accounting, so looking at inventories of, Scottish. So yeah, using surveys of,

Wojciech Węgrzyński 56:00

Hmm.

Stephen Welch 56:01

stuff do we have in our homes

Wojciech Węgrzyński 56:03

No.

Stephen Welch 56:03

Scotland in our house. So,

Wojciech Węgrzyński 56:04

Yeah. Well, yeah

Stephen Welch 56:04

uh, gathering modern data of, of what we have here. But how, how will that help us answer the question about a nuclear exchange in India and Pakistan? yeah, a lot of extrapolation, uh, needed

Wojciech Węgrzyński 56:15

you need to talk with those Chinese people who run the data from your, uh, vacuum cleaner that, uh, scans your house every 24 hours, you know? They probably have good fuel load data.

Stephen Welch 56:25

Yeah, we're talking also about, uh, developing AI tools. Some of our colleagues, uh, David Rush here is, is very interested in that area. So,

Wojciech Węgrzyński 56:32

W- I, I think that's amazing. I, I actually it's, it's some people think it's, it's kind of like funny or simple, but I, I find surveying fire logs with, uh, AI, uh, image technology, it's something that we're like, just a little inches away from having a reliable solutions for that. So I'm, I'm... I, I hope that, they succeed. Uh, o- one, one final question.

Natural Paths To A Similar Winter

Wojciech Węgrzyński 56:53

Do you think, a natural disaster could result in something very similar to what we talked like in here? Because here, okay, the nuclear scenario i- i- is like one, it's, uh, we just want to know what happens if humanity goes into this horrible, part where, where we, we cause a nuclear war, which obviously there's gonna be a lot more problems around the world, uh, at that point. Uh, but I, I wonder like if there's any foreseeable scenario in which this happens from a natural causes. And what comes to my mind is, you know, like the heatwaves in London. Remember 2021 was that, uh, the year where there was a massive heatwave and there was like the, the day with the most fire interventions? Uh, uh, I think I gave a-- published a paper on that recently.

Stephen Welch 57:39

Yes, and

Wojciech Węgrzyński 57:41

that, that day the wind was very low.

Stephen Welch 57:43

Yeah.

Wojciech Węgrzyński 57:44

That day it was lower than average, so it was a coin toss, you know?

Stephen Welch 57:47

It's a Swiss cheese model when things line up. Uh, same, same with, same with the Bunsfield fire. How many Bunsfields would've happened if, if, if there'd been, uh, no wind, so

Wojciech Węgrzyński 57:57

but do you think it, it could happen for natural reasons like wildfire earthquake

Stephen Welch 58:00

so, uh, it's actually a, a good prompt for me to mention, the, the concept of uh, Fimbulvetr in Norwegian, which is, um, well, actually what we, we call one of our, uh, proposals in, in this area. So, very briefly, this is an Old Norse term meaning an awful mighty winter, uh, the harsh winter that precedes the end of the world. Uh, and it dates back to a mythology related to the volcanic winter of 536. So you mentioned volcanoes earlier, but yet, uh, over the history of, uh, humanity, we have examples of this. and I don't know how much scientific verification there is, from that, ancient history, but, um, certainly Wikipedia says it resulted in a notable drop in temperature across Northern Europe, because of volcanic reason. And, and we're, and we're aware of current examples a-as well. so that's probably the most obvious one, yes. A volcanic e-explosions that have, uh, climate impacts. Uh, I think down at kind of and landscape scale, obviously wildfires are, increasingly topical. They're happening all around us, even in, in the countries we're sitting in at present. They're more and more relevant. But I think, apart from a few in, in vast boreal forests or, uh, particularly in Australia where they've got eucalyptus that adds a lot of, um, uh, oil into the, i-into what's combusting, uh, I think, you know, most wildland fires are, are not gonna tip a-across the threshold. So it's the megafires that you need to, to explore really and then you're into kind of, well, why do we have megafires? Do we have-- What about, uh, wildland fire management? I think there's massive overlap here and, and potential to learn from the, the wildfire community. Uh, so yeah, I've had, uh, Zach has also been involved, uh, here with some of these projects.

Wojciech Węgrzyński 59:47

Though, uh, i-i-if you think about it, w-we are kind of creating a lot of new vulnerabilities in this, in this space on the city scale, not just like building scale. Uh, yeah, we do. But, you know, uh, the combustible facades, that's one.

Stephen Welch 1:00:02

Uh-huh.

Wojciech Węgrzyński 1:00:02

the green facades, which at the same time are pretty beauti- like in this scenario of, of not nuclear attack, but just the drought, they become very strong vulnerability. Mass timber buildings, which from this perspective, they're quite...

Stephen Welch 1:00:19

Mm-hmm

Wojciech Węgrzyński 1:00:20

And the, the, the, the cities are, like, changing. Therefore, I think that, sleeping well with, you know, oh, yeah, the, the age of conflagration is gone. We haven't had a huge city fire since, uh, whatever was the last one of, uh, like there was a fire in Japan in '70s. There was one in 2013, I think, so that, that's probably a risky one. But you

Stephen Welch 1:00:41

all the things you, you say are relevant, but the, the nuclear-- the ignition assumption changes all of...

Wojciech Węgrzyński 1:00:47

Yeah

Stephen Welch 1:00:47

you don't care anymore about fire spread and the size of the fire. If

Wojciech Węgrzyński 1:00:50

yes, yes. In, in a nuclear

Stephen Welch 1:00:52

100,000 ignitions, you know, you,

Wojciech Węgrzyński 1:00:54

you don't care. Yeah.

Stephen Welch 1:00:55

problem

Wojciech Węgrzyński 1:00:55

absolutely. Absolutely.

Stephen Welch 1:00:57

assumption, starting assumption.

Wojciech Węgrzyński 1:00:58

and, and going down to the, uh, natural, I mean, i-i-if this can be caused by, to some extent, an earthquake, like, like in Japan that I think that the fire was triggered by an earthquake.

Stephen Welch 1:01:09

yeah

Wojciech Węgrzyński 1:01:10

W- well, perhaps due to the fact of using more open flame in the houses though to, to, to, you know, give a broader view. But, now we use a lot less open flames in our houses. But we have lithium batteries too, so

Stephen Welch 1:01:22

Yeah, to

Wojciech Węgrzyński 1:01:22

say, let's say, let's say we're even. Uh, look, y- you have spaces in the world where you would have a very strong fan winds like, uh, Santa Ana winds in California, which are long-lasting. Uh, it's, it's... and it's not random. They, they just-- they're seasonal. So there's a season for those winds,

Stephen Welch 1:01:43

Yes

Wojciech Węgrzyński 1:01:44

uh, earthquakes are completely random. So if you get an earthquake in that season, okay, we're down to probabilities and chances, et cetera, but

Stephen Welch 1:01:55

Yeah.

Wojciech Węgrzyński 1:01:55

that could be a really nasty scenario

Stephen Welch 1:01:57

Mm-hmm. no, I think, I think the wind aspect of the problem is, is very interesting and something I think we've, we've kind of glossed over so far. So the whole debate about your under-ventilated fire and soot production, uh, in a, in a firestorm context, obviously it changes so that those, those velocities external to your compartment could be eventually very large. Uh, obviously not everywhere. You've got periphery and you've got center and got stagnation points or whatever. So, um, uh, that, that's I think one of, uh, one of the areas that we need to, to delve into a bit more, uh, begin to think about a bit more for, the next iteration of the model.

Why Awareness Still Matters

Wojciech Węgrzyński 1:02:36

What a beautiful complex world for, uh, an interesting and non-obvious, uh, problem, Stephen. I, I understand why, why you went into that.

Stephen Welch 1:02:46

Thank you. Yeah, no, it, it has been fascinating. Some people say, "Oh, that's a bit morbid, uh, a bit morbid, a bit, um, a bit of a depressing, topic." But, I mean, I think I was inspired by, I read some words by one of the, one of the fairly recent authors, um, is, um, a guy called, uh, Baum, Seth Baum. uh, not, not, uh, uh, not Ara Baum. Um, uh, and he talks about the historic efforts, uh, uh, raising awareness about nuclear winter, including people like Carl Sagan and, um, uh, the Nobel Prize, uh, winner in chemistry back in, in the '80s. but that actually everyone can, can, can raise awareness of, of this issue. It's not just scientific experts. and, it's, it's something that we, that, that we, that we need to, to, to have more awareness of.

Wojciech Węgrzyński 1:03:34

Th- th- this would make a fantastic pop science documentary movie,

Stephen Welch 1:03:38

Mm-hmm.

Wojciech Węgrzyński 1:03:39

Stephen.

Stephen Welch 1:03:40

Yeah.

Wojciech Węgrzyński 1:03:40

you should, you should try, uh, and get-- build some interest. Ah, this is amazing. Well, thank you again for, for coming to the Sci- "Fire Science" show and, and spending an hour with me talking about all those fantastic, fascinating things, and I hope it's, uh, it's, uh, fit for Toth, uh, also for engineers dealing with, uh, down-to-earth problems in normal buildings. Uh, a lot of what we said relates to your normal ordinary... Yeah

Stephen Welch 1:04:05

students have feel that they've learned something they can use in their careers as well.

Wojciech Węgrzyński 1:04:08

hopefully not directly. That's what

Stephen Welch 1:04:10

directly. No,

Wojciech Węgrzyński 1:04:11

it is.

Stephen Welch 1:04:12

indeed. Yes.

Wojciech Węgrzyński 1:04:13

Thanks, Stephen.

Stephen Welch 1:04:13

Yeah. Thank you

Wojciech Węgrzyński 1:04:14

And

Host Wrap And Next Episode Tease

Wojciech Węgrzyński 1:04:15

that's it. Thank you for listening. It turned out to be completely different than what I thought this interview will be, uh, before I've pressed record. It was very interesting discussion, very much embedded in the fire safety engineering concepts. We've discussed winds, we've discussed firestorms, we've discussed fire spread, compartment fires. We've discussed soot formation and soot modeling actually. That was quite an interesting twist, and all those things combined allow you to approximate if a fire caused by nuclear explosion, can it create this catastrophical outcome, being a direct and measurable significant climate impact or not? You need to have a lot of soot formed in the fires move through a plume formed by the fire into the upper layers of the atmosphere, and it's not that obvious will the smoke go there or not. Very interesting topic. I'm still curious if a catastrophe of this magnitude could happen from natural causes or just simply a conflagration of a large modern city actually. If you think about the amount of plastic materials we have in our houses, the amount of flammable materials we have in everywhere around us compared to what we had in the '70s and '60s when people were actually calculating, you know, the loads and consequences of those fires. I'm not so sure, but maybe, maybe a, a normal catastrophical fire could, have a significant climate impact, measurable climate impact. I'm not sure what would happen if a large city like the one I live in, stood in fire altogether at once. That, that could be quite a catastrophical outcome. I hope we never find out in practice, but I find it also quite an inspiring and challenging engineering task to go through, uh, as a, as a fire engineer. It's good to, to, to g- get some, uh, hopefully abstract, uh, concepts, you know, and, and brainstorm the consequences of those fires. I had great joy contemplating fires on Mars in a extraterrestrial, uh, human habitat one day, uh, 100 episodes ago, so here we go again. Um, anyway, Stephen, thank you so much for bringing this topic up. I found it fascinating and, I already know we are gonna see Stephen in a episode in just a few weeks because we've recorded more than one episode, and the next one will be on modeling traveling fires, so that's also a very practical, uh, insights to be given. Anyway, for today, thank you so much for being here with me in the Fire Science Show, and I hope to see you here again next Wednesday. Thank you. Cheers. Bye.