WEBVTT
00:00:00.080 --> 00:00:02.240
Hello everybody, welcome to the Fire Science Show.
00:00:02.240 --> 00:00:08.640
By now I think you are aware of my stance about where wildfire engineering is going.
00:00:08.640 --> 00:00:21.280
I'm pushing this into the podcast as much as I can because I truly believe that engaging fire engineers into solutions of wildfire problems, engaging with communities, is really an important thing.
00:00:21.280 --> 00:00:24.239
We're the only ones who can do this kind of work.
00:00:24.239 --> 00:00:58.479
But I also see a rising opportunity for the industry at large, as I also see this becoming a product, a product of fire engineering that could be served to different stakeholders like communities, regional, municipal administration, etc., which want to invest in their safety assessments and improve the safety of their citizens through uh science engineering based uh decision making processes uh powered by a very difficult and challenging analysis.
00:00:58.479 --> 00:01:14.000
Therefore, I I try to build up a lineup of talks in the podcast that uh together kind of carve the way towards practical use of fire safety engineering as a framework in the world of wildfires.
00:01:14.000 --> 00:01:28.239
And of course, uh I'm just a podcaster, I'm not researching this, there are brilliant people who research this uh worldwide, and I was super happy when I saw a keynote at Ljubljana conference earlier this year by Dr.
00:01:28.239 --> 00:01:34.400
Pascale Vacca from University Politechnica Catalunia UPC, and Pascal did show exactly this.
00:01:34.400 --> 00:01:40.560
She shown a workable framework on how wildfire engineering could be performed.
00:01:40.560 --> 00:01:52.640
UPC is known for industrial fire safety, so Pascale seems obviously inspired by risk and approaches coming from the industrial sector, from chemical engineering, etc.
00:01:52.640 --> 00:01:56.239
So it it truly is a workable framework.
00:01:56.239 --> 00:02:15.280
And in this podcast episode, I ask her some deep deeper questions about how this framework looks like, how it can be applied, what are the challenges, what are the missing links, and uh what actually is the role of fire safety engineers at different stages of executing this.
00:02:15.280 --> 00:02:24.879
So, uh, while fire engineering is a future, it is a part of our profession already and will be an increasingly large part of our profession in the future.
00:02:24.879 --> 00:02:29.280
I think we all need to be aware how this works and how can we do it the best.
00:02:29.280 --> 00:02:32.400
So let's learn from those who are actually doing it.
00:02:32.400 --> 00:02:35.199
Let's spin the intro and jump into the episode.
00:02:39.919 --> 00:02:41.599
Welcome to the Fireside Show.
00:02:41.599 --> 00:02:45.280
My name Wojciech Wegrzynski , and I will be your host.
00:03:18.879 --> 00:03:23.360
Applications for OFR's 2026 graduate program are now open.
00:03:23.360 --> 00:03:30.000
If you're ready to launch your career with a supportive forward-thinking team, visit OFRconsultants.com to apply.
00:03:30.000 --> 00:03:36.240
You will join a worldless organization recognized for its supportive culture and global expertise.
00:03:36.240 --> 00:03:40.800
Start your journey with OFR and help shape the future of fire engineering.
00:03:40.800 --> 00:03:42.240
Hello everybody.
00:03:42.240 --> 00:03:46.560
I am joined today by Pascale Vacca from University Politecnica Catalonia.
00:03:46.560 --> 00:03:48.159
Hey Pascale, good to have you in the podcast.
00:03:48.400 --> 00:03:50.319
Hi, thank you so much for inviting me.
00:03:50.639 --> 00:03:51.199
And thank you.
00:03:51.199 --> 00:03:52.960
Thank you for coming to the Fire Science Show.
00:03:52.960 --> 00:03:57.840
I've really enjoyed your keynote in the recent conference in um Ljubljana.
00:03:57.840 --> 00:04:01.520
So that was very nice, and immediately I invited you for this.
00:04:01.520 --> 00:04:03.680
You took the invite, so here we are.
00:04:03.680 --> 00:04:16.319
So let's try to give a deeper dive into the keynote topic, which was integrating wild fire urban interface, fire risk management into the fire safety engineering practice, challenges and opportunities.
00:04:16.319 --> 00:04:19.360
I'm not sure if I want to start with challenges or opportunities.
00:04:19.360 --> 00:04:20.480
That's a tough choice.
00:04:20.480 --> 00:04:27.600
But maybe maybe the first question to ask is why do you think we need to integrate fire risk management into fire safety engineering practice?
00:04:27.600 --> 00:04:28.959
Perhaps let's start there.
00:04:29.279 --> 00:04:47.120
So, with all the events that we are seeing, you know, every summer and also in winter, as we've seen in LA, I believe that fire safety engineers can greatly contribute in basically reducing risk and reducing the consequences of wildfires when they reach urban communities.
00:04:47.120 --> 00:04:53.839
So that's I think that's it's a great opportunity for fire safety engineers to contribute to the to the problem.
00:04:54.319 --> 00:05:00.000
Do you think I I many times I've said in the podcast that I see this being a real job eventually?
00:05:00.000 --> 00:05:05.600
Like I even had people already emailing me, voice like I am already a wildfire uh engineer.
00:05:05.600 --> 00:05:11.920
Uh but do you think the wild the fire engineers could contribute at all parts of the process?
00:05:11.920 --> 00:05:13.920
How big the contributions could be?
00:05:14.160 --> 00:05:21.680
Yeah, so when we talk about uh the WUI fire issue, so the Warner Nurban Interface fire issue, we talk about different scales.
00:05:21.680 --> 00:05:31.439
So we have the landscape scale, that's the biggest scale, and then we can go down to the community scale, and then even smaller to the homeowner level.
00:05:31.439 --> 00:05:42.240
So when it comes to the landscape scale, this is where fire safety engineering would say has less of uh input because here it's where forest engineers are the experts.
00:05:42.240 --> 00:05:47.519
Um, but we can greatly contribute at the community scale and at the parcel scale.
00:05:47.519 --> 00:05:52.000
This is where buildings are located, this is where we can look at evacuation.
00:05:52.000 --> 00:06:09.600
Uh so this is kind of where uh our expertise would be really useful because we we're experts, of course, in a fire uh inside a building or in an industrial facility, but we can translate this expertise to you know what happens if the fire is coming from the outside and it's reaching the building.
00:06:09.839 --> 00:06:11.120
I share your opinion here.
00:06:11.120 --> 00:06:27.680
I think this though, I still uh see uh you know with the forest scale, with uh the developments like Winity and you know the tools that the colleagues around the world are building to uh manage like large-scale evacuations and fires, but I think there's still uh perhaps some preparedness work to be done.
00:06:27.680 --> 00:06:31.759
But I did indeed we can shine in the community scale.
00:06:31.759 --> 00:06:42.800
Um, you being in UPC, I know UPC always has been a place for industrial fire safety, you know, and it's it's made its name uh in this industrial fire safety is risk management.
00:06:42.800 --> 00:06:45.920
And uh I also see risk management in your approach.
00:06:45.920 --> 00:06:47.759
So are those two things connected?
00:06:47.759 --> 00:06:53.120
Are you like following the ideas of industrial uh risk management in in this space?
00:06:53.439 --> 00:06:54.240
Yeah, exactly.
00:06:54.240 --> 00:06:55.120
You're right.
00:06:55.120 --> 00:07:02.319
The the group Intertech is uh let's say most known for industrial fire safety, industrial risk.
00:07:02.319 --> 00:07:08.079
We also worked, I think now there's about 20 years of experience in wildfires.
00:07:08.079 --> 00:07:16.639
Um they started looking at fire retardants because we were in the chemical engineering department, so you know then you there's some penciling there.
00:07:16.639 --> 00:07:19.680
And then we move down to actually looking at risk.
00:07:19.680 --> 00:07:26.959
So when we talk about the wildfire that's reaching the woo, we also use a risk, let's say risk assessment approach.
00:07:26.959 --> 00:07:32.079
Uh we look at what's the hazard, you know, are you located in a hazardous area?
00:07:32.079 --> 00:07:46.959
Uh yes or no, what's the exposure, for example, that the community or reparsel is going to have when it comes to the wildfire, and then we look at the vulnerability of the community or of the building if we look at the smaller scale.
00:07:46.959 --> 00:07:54.800
So we still have this uh, let's say we still look at risk as a whole, so including all its uh parts.
00:07:55.199 --> 00:08:06.720
In the paper that you sent me, uh there's a formal way of doing this called the disaster risk management cycle, which covers the phases of risk assessment planning and and all the stages.
00:08:06.720 --> 00:08:19.519
I'll I'll briefly read through the phases and then we'll go in-depth into them because I really enjoyed that the fact that it's a framework, you know, it's it's not just uh a bunch of thoughts that uh it would be better if we did we did something with the problem.
00:08:19.519 --> 00:08:21.920
No, it's it's it's a it's a workable framework.
00:08:21.920 --> 00:08:33.600
And also um for those who were not in Ljubljana, Pascal has presented a full case study, applying a lot of what is being said in here to a specific uh part in uh uphill Barcelona.
00:08:33.600 --> 00:08:39.919
So we will not only cover the the theory, but we'll also try to discuss the case study, I hope.
00:08:39.919 --> 00:08:41.120
We can right.
00:08:41.120 --> 00:08:48.159
Okay, so let me let me just quickly read the the phases of disaster risk management cycle DRM.
00:08:48.159 --> 00:08:54.879
Assessment, risk management planning, risk prevention, emergency preparedness, emergency response, and recovery.
00:08:54.879 --> 00:08:57.440
How about we try and tackle them one by one?
00:08:57.440 --> 00:09:04.480
So, how do you see this integration of we risk management into fire safety engineering practice in the assessment stage?
00:09:04.480 --> 00:09:07.200
Like what is the stage and what engineers could do in it?
00:09:07.440 --> 00:09:08.240
Yeah, I think so.
00:09:08.240 --> 00:09:13.519
Risk assessment, I think this is what we are most comfortable with as fire safety engineers.
00:09:13.519 --> 00:09:24.159
We are used to do fire risk assessment when it comes to events, fire events inside buildings, so we can translate this to when a fire is coming from the outside.
00:09:24.159 --> 00:09:29.600
So basically, here is where we look at all the all the different components of risk.
00:09:29.600 --> 00:09:34.240
Um, so again, we look at the hazard, we look at the exposure, we look at the vulnerability.
00:09:34.240 --> 00:09:41.519
And this, the assessment, the typology of the assessment will depend on the scale that you're looking at.
00:09:41.519 --> 00:09:51.840
If you're looking at the community, you might want to look at different criteria and for different indicators compared to when you are looking at the parsley level.
00:09:51.840 --> 00:10:08.799
So, for the community, for example, uh, when it comes to the exposure, we might look at how the fire can percolate, can spread through the community, reaching the entire community, so not just the what the part that's facing the wineland.
00:10:08.799 --> 00:10:24.159
And this has also several components with, for example, uh the topography will help uh spreading the fire, the amount of vegetation or amount of fuels that inside the community uh can help spreading the fire.
00:10:24.159 --> 00:10:30.480
So at the community level, we might look at things more in general of the community.
00:10:30.480 --> 00:10:38.320
If we we might also, for example, include the construction materials of the most of the homes inside the community.
00:10:38.320 --> 00:10:49.279
It's not the same if we have a situation such as in what happened in LA or in Hawaii, where we had most of the houses made out of timber.
00:10:49.279 --> 00:10:53.519
So this also helps the fire spreading through the community.
00:10:53.519 --> 00:11:00.159
And it's different if we are, for example, here in Mediterranean Europe, where buildings are made out of concrete.
00:11:00.159 --> 00:11:03.039
So they're not gonna burn as much.
00:11:03.039 --> 00:11:14.799
Um, so we can look at these criteria, and then at the at the smaller scale, so the parcel level, uh, we can really look in detail at what are the vulnerabilities of a building.
00:11:14.799 --> 00:11:38.159
There's a lot of work that's been done already to identify which are the most vulnerable elements of a building to a fire that's coming from the outside, and then we can actually assess um what my case study did, what I did in the case study is that I assessed um the vulnerability of the building using a performance-based design approach.
00:11:38.159 --> 00:11:48.320
So I set some performance criteria, I had the fire spreading towards the building, and then I had a look at how the building is gonna react to the fire.
00:11:49.360 --> 00:11:53.039
We can actually touch a little more on that because it was an interesting case of modeling.
00:11:53.039 --> 00:12:07.600
So, from what I remember in the in the keynote, you've run uh FDS simulations for fire spread outside of the building and some fire approaching a building, and then uh you were investigating a facade, like a glazed facade, if I recall.
00:12:07.600 --> 00:12:13.759
Uh so maybe maybe give us some more detail about how you've used the fire engineering tool set in in this problem.
00:12:13.759 --> 00:12:18.879
Because like for me, I would simulate the building and see what's happening outside, you know, and you reverse the problem.
00:12:19.120 --> 00:12:28.000
Yeah, so the issue came because the area of uh Barcelona that's uh located at the WUI is quite a problematic area.
00:12:28.000 --> 00:12:50.240
So the firefighters came to us uh because they uh wanted to know whether they could use our civic senator, it's called basically a community senator of a neighborhood, if they could use it as a community shelter in case of a wildfire, because they knew that the situation of the homes located in the neighborhood was not really great when it comes to preparedness.
00:12:50.240 --> 00:12:54.159
So this was a real problem that we had.
00:12:54.159 --> 00:13:08.159
And so we decided to approach the issue first simulating a wildfire approaching the buildings because we wanted to know okay, what's the intensity of the wildfire that's gonna arrive at the building?
00:13:08.159 --> 00:13:19.360
We kind of took a worst case uh scenario and we ran a faresight simulation to know what would be the fairline intensity, what would be the rate of spread close to the building.
00:13:19.360 --> 00:13:24.879
And then once we have this information, this is our input for our FDS simulation.
00:13:24.879 --> 00:13:29.519
So this is we need some wildfire inputs to place in FDS.
00:13:29.519 --> 00:13:41.519
We don't have uh in when we do performance-based design for buildings, there are some PBD codes and guidelines that tell us, okay, with this type of uh occupancy, you have such heat release rate per unit area.
00:13:41.519 --> 00:13:43.279
And the fire is gonna be this big.
00:13:43.279 --> 00:13:47.200
We are not there yet when it comes to away fires.
00:13:47.200 --> 00:13:48.879
So this is how we did it.
00:13:48.879 --> 00:13:54.480
We needed some fire inputs with the outputs from Farsight to then input in FTS.
00:13:54.480 --> 00:14:05.120
And then we simulated a fire approaching with the rate of spread uh that far sight send, with the intensity that faresight send, and we could see it approach the building.
00:14:05.120 --> 00:14:17.440
Of course, you you also have to put the meteorological conditions in there, so the wind is really important because it's gonna push the flames towards a certain direction, depending on where the wind is coming from.
00:14:17.440 --> 00:14:24.320
Uh, and and this building had a very big glass facade just facing the wild end.
00:14:24.320 --> 00:14:30.879
Um, so we thought, okay, we know the light blazing facades are vulnerable to fire.
00:14:30.879 --> 00:14:36.799
Um, so we wanted to specifically simulate this because it was the most vulnerable element of the building.
00:14:36.799 --> 00:14:50.720
And then through the FDS simulation, you know what's the heat flux that's reaching the glass, what's the glass temperature, a surface temperature, and then you set some performance criteria, and then you see whether uh the situation.
00:14:50.720 --> 00:15:00.080
What we did was analyze the situation as it was, so see whether the building would resist the passing of a wildfire or no.
00:15:00.080 --> 00:15:04.080
And in our case, the result was that that the glazing would break.
00:15:04.080 --> 00:15:11.519
So we need to put some protection or we need to identify some mitigation strategies so that uh the glazing will not break.
00:15:11.679 --> 00:15:14.000
Yeah, but that's for the further phases.
00:15:14.000 --> 00:15:15.120
We'll come back to this.
00:15:15.120 --> 00:15:19.279
Uh we need to clarify what's far sight and how does one use that?
00:15:19.519 --> 00:15:20.080
Uh yeah.
00:15:20.080 --> 00:15:23.360
Okay, far sight is a wildfire spread model.
00:15:23.679 --> 00:15:24.000
Okay.
00:15:24.320 --> 00:15:26.559
It's a semi-empirical model.
00:15:26.559 --> 00:15:32.399
So it uses rotor mill to identify how the fire is gonna spread.
00:15:32.399 --> 00:15:55.679
Uh, you set an ignition point, you have to know the topography of the landscape, you set the meteorological conditions, and then with this, and and of course, very important, you need to have a fuel map uh so that we actually know what's burning, what's the vegetation uh that's burning, and then you can get lots of outputs from the simulation.
00:15:55.679 --> 00:16:00.639
Um, you can get arrival times, you can get rate of spread, fire line intensity.
00:16:00.879 --> 00:16:05.279
And the one that you can translate into something usable by FDS is which one?
00:16:05.279 --> 00:16:06.080
The intensity?
00:16:06.080 --> 00:16:06.960
The fire line intensity?
00:16:07.200 --> 00:16:13.360
Yeah, so what you can do is the um that there's also an empirical equation to calculate the flame depth.
00:16:13.360 --> 00:16:17.440
This is basically how deep the the flame of the firefront is.
00:16:17.440 --> 00:16:20.639
Um the fine line intensity is in kilowatt per meter.
00:16:20.639 --> 00:16:23.519
Okay, and then you divide by the flame depth.
00:16:23.519 --> 00:16:26.080
This is one of the one of the options.
00:16:26.720 --> 00:16:31.360
That's the interesting part, you know, because it's in principle it's only very uh logic and easy.
00:16:31.360 --> 00:16:36.080
You take one model, oh, it gives you an output, and then you put it into another model, and here you go.
00:16:36.080 --> 00:16:38.159
It's your normal everyday Far Safety engineering.
00:16:38.159 --> 00:16:46.960
But um, there are you know those conversions uh and the way how the models work is it's not one unified set of tools like you have to resort.
00:16:46.960 --> 00:16:56.320
And we could probably go deeper on Farsight uh with what's Rotoman model, how it has been developed, and how how does wind and fuel go into that, but that would be an entire podcast episode.
00:16:56.320 --> 00:17:02.960
So I'll probably uh stop with the details of the Farsight in here because I have other questions for the assessment process.
00:17:02.960 --> 00:17:18.559
Um, it's a risk assessment, so I wonder how the quantification of the probabilities look like, or or are like first of all, do you even care about ignition or you look only on the probability that something spreads in certain conditions?
00:17:18.559 --> 00:17:24.640
Because I have an intuitive feeling that ignition will be very, very difficult to quantify.
00:17:24.960 --> 00:17:26.240
Yeah, uh of course.
00:17:26.240 --> 00:17:29.680
So for us, we assume that ignition will happen.
00:17:29.680 --> 00:17:53.519
Yeah, um, luckily in Catalunya, so we are analyzing the specific cases in Catalonia, uh, the firefighters did a great job in collecting past uh fire perimeters, past ignitions, and they kind of identified for each small area of Catalonia the typology of fire that is most probable based on what happened in the past.
00:17:53.519 --> 00:18:00.880
So, for example, for the area that we analyzed, they identified two different types of fire that would happen most probably.
00:18:00.880 --> 00:18:07.279
One was is pushed by the wind and the topography, and the other one would be a prune-dominated fire.
00:18:07.279 --> 00:18:07.839
Okay.
00:18:07.839 --> 00:18:12.160
I mean, we are very lucky that this work was previously done.
00:18:12.160 --> 00:18:14.960
So we can use it as a base for certain areas.
00:18:14.960 --> 00:18:18.559
In other areas, it might be more difficult to have this information.
00:18:18.559 --> 00:18:26.319
So then you might go more on a deterministic base of like, okay, what's my main wind direction?
00:18:26.319 --> 00:18:34.400
So, based on that, what's gonna happen if I have an ignition in the worst case uh location based on the wind?
00:18:34.400 --> 00:18:37.920
Yeah, it's not easy if you if you have limited information.
00:18:37.920 --> 00:18:46.960
You can look at past fires, but uh I'm I don't know if each country is recording this, if it's publicly available or not.
00:18:47.279 --> 00:18:51.599
It may also be a case of a small sample size, like think think Nordic countries.
00:18:51.599 --> 00:18:59.759
I don't think they can base the future on uh the past fires, like I had Nieves in my podcast, and that was that was the biggest issue for them.
00:18:59.759 --> 00:19:02.000
Uh how about the fuel?
00:19:02.000 --> 00:19:09.599
Because one, I assume you need to understand what biologically is there, like is it trees, bushes, what kinds of them?
00:19:09.599 --> 00:19:13.119
But also I think it's important for you what state they are.
00:19:13.119 --> 00:19:14.799
Like, are they dried?
00:19:14.799 --> 00:19:15.920
Are they alive?
00:19:15.920 --> 00:19:22.079
Like in Poland, they would go through seasons, so that probably also is this something you account for?
00:19:22.319 --> 00:19:35.359
Uh yeah, so when you look at the information that's available, so you have the the fuel map, and of course, this has different resolution depending on where your information is coming from.
00:19:35.359 --> 00:19:38.559
So this also impacts your results.
00:19:38.559 --> 00:19:46.240
Of course, it's it's better to have a finer resolution because then you can really go at a smaller scale, specifically when it comes to our work.
00:19:46.240 --> 00:19:57.039
Maybe at the landscape scale, it's not as important, but when you look at the community level, it's really important to have at the smaller scale as like up to a single tree scale, like no, no.
00:19:57.039 --> 00:20:02.720
For example, I would say we uh in Catalonia we have 20 meters by 20 meter.
00:20:02.720 --> 00:20:09.680
Uh, I think that's uh that looks like fine enough because then you can really distinguish what's the urban area.
00:20:09.680 --> 00:20:13.759
So what's in the according to the fuel map is not gonna burn.
00:20:13.759 --> 00:20:16.640
Then we can talk about this because this will also burn.
00:20:16.640 --> 00:20:28.160
Uh what is urban area, and then you can really distinguish what is agricultural uh fields, uh, what is the wildland, and then you have, of course, the distinction you have shrubs, you have forest.
00:20:28.160 --> 00:20:34.960
And you also have information on the bulk density of each technology and on the moisture content as well.
00:20:34.960 --> 00:20:40.720
So this is where where we uh look at the conditions of the vegetation.
00:20:40.720 --> 00:20:50.079
Uh we can have you know very low moisture for the dead fuels, and then uh we can have live trees as well.
00:20:50.079 --> 00:20:56.400
One of the issues of these fuel maps, however, is that they are not up to date.
00:20:56.400 --> 00:21:10.480
So sometimes the fuel map is telling you, for such as in the case study that we did, the fuel map is telling you that you have shrubland, and when you actually go there and look at the vegetation, you see pine trees as well.
00:21:10.480 --> 00:21:21.440
So the the that is not updated, and also on the other side, the fuel that is selected, for example, in Farside is the one that's gonna be carrying the fire.
00:21:21.440 --> 00:21:26.880
So if you have different typologies, you actually are using the one that's gonna be carrying the fire.
00:21:26.880 --> 00:21:33.279
So in our example, it was shrubland because it's the shrubs that are gonna be carrying the fire mostly.
00:21:33.680 --> 00:21:45.200
As we're talking about the opportunities for fire safety engineers, I would like maybe I'm naive, but perhaps like a fire engineer could also do the survey themselves, like take a drone, fly around the area and just map it.
00:21:45.200 --> 00:21:49.279
In if 20 by 20 is your grid that that's sufficient for you.
00:21:49.279 --> 00:21:55.519
I think that's that's quite a manage, like at a community center scale, like you were presenting in in uh UPC.
00:21:55.519 --> 00:22:01.359
I mean, that that's a scale of a part of the terrain and the building which you're comfortably modeling with FDS.
00:22:01.359 --> 00:22:08.240
So uh for any fire safety engineer uh listening, like it's it's it's not really huge, it's like few hundred by few hundred meters that's manageable.
00:22:08.240 --> 00:22:14.319
Could imagine like just going there with a drone and uh and you know mapping it for yourself.
00:22:14.640 --> 00:22:19.440
Yeah, actually we did this in a in a project uh several years ago.
00:22:19.440 --> 00:22:39.759
So we had a drone to and basically we had like a 3D with the through the drone, we had a 3D uh image of the area that we wanted to analyze, and then we specifically know okay, the vegetation is located here, uh, and then we could translate this to basically FDS obstacles, you know.
00:22:39.759 --> 00:22:53.359
Uh this was done, it was like I'm not expert at that voxelization of uh, and then this was really easily imported into pyrocene, and then we had obstacles for FDS.
00:22:53.359 --> 00:23:05.920
Of course, this is a big difference because when you are looking at uh fire spreading through the vegetation, like you don't really care about, let's say, the 3D shape of the vegetation.
00:23:05.920 --> 00:23:16.079
You know, the forest is gonna spread mainly on at the surface, so and then it's of course it's gonna catch, you know, if you have a shrub that's close to the surface, it's gonna catch this.
00:23:16.079 --> 00:23:24.319
Uh, but at the moment we are simulating in FDS, we are simulating the vegetation just as a vent, uh like a burner, like typical.
00:23:24.319 --> 00:23:25.200
Yeah, exactly.
00:23:25.200 --> 00:23:27.039
As a burner, yeah, yeah, yeah, yeah.
00:23:27.039 --> 00:23:28.960
On the 3D on the drone.
00:23:28.960 --> 00:23:35.519
This was uh uh we use this to uh then create virtual reality experiences.
00:23:35.519 --> 00:23:56.640
Uh so this is something that when it comes to the prevention and preparedness side is really interesting because then you can place the people inside their neighborhood or inside on their parcel, and you can have a fire starting, and they can see actually how the fire will spread towards the community or towards their home.
00:23:56.640 --> 00:24:00.319
So this is something that we did with uh David Caballero.
00:24:00.319 --> 00:24:04.240
He's uh an expert on way fires in Spain.
00:24:04.240 --> 00:24:09.599
So yeah, this is also something really interesting where engineers can contribute for sure.
00:24:09.920 --> 00:24:10.160
Wow.
00:24:10.160 --> 00:24:31.599
Well, uh it's it's great to discover that that like those uh ideas and concepts are turning into like uh research and then into tools because it means that this will propagate, and in some years this this will really be uh something that a single engineer, like single fire engineer, like you're delivering a fire strategy to a building today, like you could actually do that study.
00:24:31.599 --> 00:24:37.599
And I I wouldn't mind flying to um to Spain and fly a drone and do some modeling in field.
00:24:37.599 --> 00:24:41.519
Uh that that doesn't sound like a bad job to me.
00:24:41.519 --> 00:24:44.960
If you if it is if it could uh support living.
00:24:44.960 --> 00:24:53.680
Um one more um so so in in this uh set of boundary conditions, I also have weather on my list.
00:24:53.680 --> 00:24:56.799
So so how weather is quantified?
00:24:56.799 --> 00:25:04.400
Like one, do you care about like do you care about the range of probable wins or you just care about the worst ones?
00:25:04.400 --> 00:25:08.559
And how do you quantify which one is the worst, actually?
00:25:08.880 --> 00:25:21.359
Yeah, um, well, this depends on on what the focus of your analysis is because one thing is okay, I want to know whether a building can survive a worst case fire scenario.
00:25:21.359 --> 00:25:31.759
Then you go for the highest wind gusts in uh that happened in uh in a certain day, highest temperatures, lowest moisture, moisture content.
00:25:31.759 --> 00:25:36.079
So it really depends on what uh what the objective of the study is.
00:25:36.079 --> 00:25:40.000
If you want to analyze a worst case scenario, then go for the extreme, you know.
00:25:40.000 --> 00:25:56.319
If if you want to analyze okay, what's gonna happen in my average wildfire day, then you what we did is we looked at a meteor situation in a past wildfire that was uh I think I believe it was one in 2015.
00:25:56.319 --> 00:26:10.240
Uh um this was uh a couple of days for Catalunya where it was they had several uh fires, and so we thought this would be a representative day for for uh you know a wildfire event.
00:26:10.240 --> 00:26:20.000
So um yeah, depending on on the objective, depending on what you want to look at, then you you choose uh different meteorological scenarios.
00:26:20.000 --> 00:26:31.599
And also again, you might want to look at the probability of this fire, you know, it might be a worst case fire scenario, but actually it's an unlikely wind direction for the area.
00:26:31.599 --> 00:26:44.720
So you also need to look at that what's the most likely wind direction, uh, what's gonna happen on a day where I have, you know, my typical uh meteorological conditions for the area, or look at the worst case.
00:26:44.960 --> 00:26:51.519
Yeah, like you said before, you you were supported by your firefighters who already identified the plausible scenarios for you.
00:26:51.519 --> 00:26:55.680
So I guess those were also connected with the wind direction to some extent.
00:26:55.680 --> 00:27:04.559
So it and and you had data from previous fires that occurred, which probably followed those uh those typologies of fires uh identified by firefighters.
00:27:04.559 --> 00:27:07.920
So that's a a great side of references to start with.
00:27:07.920 --> 00:27:14.480
And I can also imagine someone being in a completely blind like this is a village in the middle of uh of a of a country.
00:27:14.480 --> 00:27:19.279
Please tell me like what are the what's the the fire risk management I should do.
00:27:19.279 --> 00:27:29.359
And in that case, I guess you probably would lean towards like a complete parametric study of of different winds and different ignition locations in different fuels and different weathers.
00:27:29.359 --> 00:27:31.359
But I think FarSight supports that, right?
00:27:31.359 --> 00:27:33.920
You can do uh probabilistic in Farsight.
00:27:34.240 --> 00:27:34.640
Yeah, yeah.
00:27:34.640 --> 00:27:40.960
And also Farsight, the the let's say the good thing about Farsight is that simulation run really fast.
00:27:41.279 --> 00:27:41.599
Okay.
00:27:41.839 --> 00:27:47.039
So in in a time of I don't know, minutes, you have your results already.
00:27:47.039 --> 00:28:03.680
So it's not like FDS where you you know a model that then such as one that we run took I think it took close to two weeks to so because once you put a wind in FDS, it slows down everything, you know, because the competition is a very important thing.
00:28:03.839 --> 00:28:06.240
And you also need uh and you need a large domain, yeah.
00:28:06.640 --> 00:28:07.759
Yeah, exactly.
00:28:07.759 --> 00:28:20.079
Um, so the good thing about fire site is you can run lots of simulations uh very quickly, so then you can have an idea of uh, you know, how's the fire gonna spread if I put my ignition point over here?