Exploring the effects of climate change on weather

Transcription:

Patti Harman (00:07):
Hello and welcome to the DigIn podcast. I'm your host Patti Harman, Editor-in-Chief of Digital Insurance. It's no secret that we're seeing more frequent and severe storms, flooding, wildfires, and other weather-related events. We often hear that the shifts in weather are related to climate change, but there are other factors that come into play as well. And here to discuss some of them is Dr. Lou Gritzo, Chief Science Officer at FM, where he focuses on developing technical strategies to handle emerging and underserved risks. Thank you so much for joining us today, Lou.

Lou Gritzo (00:46):
Great to be here, Patti. Thank you.

Patti Harman (00:48):
So weather, it's certainly wreaking havoc around the globe and it is easy to attribute some of these factors to climate change, but what are we really seeing in terms of climate these days?

Lou Gritzo (01:01):
Well, the biggest overriding factor on all the things that we're seeing that affect hazards and risk is this overall very slow general trend of an increasing global temperature. And when I say global, it's the temperatures measured all around the world averaged over the entire year and you get one number and that's the one and a half degree C that you sometimes see referred to when we're talking about changes in the climate or certain things relative to the climate. So that's just one number and it smears everything together, but it's definitely going up and it bounces around a little bit, but the general trend is it's going up. So that kind of overall very slow trend beyond seasonal or annual variability is the big change in the climate. And that general temperature trend can influence natural hazards. Now it's not making any new hazards, it's just making some of the hazards that we've always faced more severe and even fewer of 'em more frequent. And so that's the challenge that we have in front of us.

Patti Harman (02:13):
Okay, that's a great way to explain all of that. I've spoken with multiple experts across the insurance industry and better managing climate risks is a major priority for them, but how are we able to apply science and data to managing risks like hurricanes and wildfires and earthquakes and other types of natural hazards?

Lou Gritzo (02:37):
There's so much talk about the climate and we see all the time there's hazards. We see them in the news and different things that are popping up that are affecting risks that we all need to manage. And it's essentially just overwhelming. I mean, there's just so much happening. So I think that an approach to this which risk managers can use, and it's basically our science approach here at FM, is we're looking at what we call the three legs on the stool. We have, what do we have is technically sound data that tells us what's happening, what's happening in the past. Of course it's data and what's happening now. Do we understand the physical mechanisms in the world that are driving what we're seeing in those data? Not just well data show this, we have no clue why. And then most importantly, do we start to have some models and some insights from the data and those physical considerations that we can use to say, okay, with those changes in the data and what we understand about the physical phenomena, here's what's important to us.

(03:42):

And that's the key thing I think for risk managers and managing climate risk, is to understand the risks that are most important to you and your business, where you're located and what you're doing, and be able to prioritize what you're going to do around the things that are most relevant to you and your business. And I think that is the key. We know enough now for those kind of actions to be something that are reasonable, something indeed is actionable. It's not a guesswork, but understanding those kind of top priorities and they vary a lot by location and there's a lot of variability in them, but in a general sense, you can still establish those priorities.

Patti Harman (04:27):
Okay, good. As I said earlier, I've spoken to a lot of experts and I am just, it's very interesting to me to see how many insurers are just really making climate and dealing with its after effects a major priority for their initiatives in the coming year. You wrote an article climate change, though quite real, isn't spawning More hurricanes, and I read it, it was such an interesting read. How did you reach that conclusion and when you share that perspective with other people, what's generally their reaction?

Lou Gritzo (05:05):
Yeah, Patti, I mean it's a little counterintuitive and I think we think about the physical mechanisms. We know that warm water is the energy that fuels hurricanes, and we also know that those sea surface temperatures are rising and this climate of rising temperature that we're seeing. So there's a natural kind of response that given that's what we know is the overriding factor for hurricanes, we have warm sea surface temperatures, they're warmer than they've ever been and therefore we're going to have more frequent and more severe hurricanes. And so the purpose of the article was to kind of dig into this three-pronged approach with data physics and models and say, okay, what do we really know about this? And is that really the case? Because it's more complicated than just simple rising sea surface temperatures. There's a lot of other influencing factors that go into it, and there's a lot of natural variability.

(06:02):

So what we did is we went back, used reanalysis data, went back far as we can and focused on land falling hurricanes because there's very good records for land falling hurricanes, the ones that are out in the ocean. We didn't have the ability to see those from satellites until fairly recently. And with low numbers of hurricanes, thankfully a year, a small window of data really doesn't do the job. And so if we look at those data very clearly, very crisply and understand where the data came from and their limitations, you'd see that there's really no big change in the number of land falling Atlantic hurricanes in recent times or even in past times. So what that means is yes, there are some changes in the hazards that hurricanes pose specifically in terms of moisture and some increased severity, especially late season. And we're seeing that right now late in 2024 while we're talking, but there's no big change in frequency of land falling hurricanes in the Atlantic. What it doesn't mean is that we can just rest on our laurels and that hurricanes aren't important. I'm certainly not saying that, but what it means is in this notion of prioritizing the risks you're going to manage, make sure you manage to the current hurricane risk, maybe a little bit of potential severity increases in the future, but don't go overboard. You don't have to.

Patti Harman (07:32):
Okay. That's a great segue into my next question because I was going to ask what changes are you seeing when it comes to storms? Because based on our coverage, it seems like we're seeing a lot more of those 100 year floods or storms and they seem to be increasing in severity. And is that what you're seeing from your perspective? And if so, why then?

Lou Gritzo (07:57):
Well, the biggest change in natural hazards is the change in flooding hazard. It's always been the world's most frequent natural hazard that frequency is increasing, the severity is clearly increasing. And so we can certainly see that in the data. We understand the physics behind that in a generally warming world, which means that at certain periods of time it's going to be much warmer because overall in average it's warmer, that warmer air, if there's moisture available to be captured in it, in other words, moisture, that you could evaporate from the oceans or other ways that warmer air holds more water for every one degree C or about 1.8 Fahrenheit warmer, the air is, it can hold 7% more water vapor. And so when it rains, there's a lot more water vapor held in that warmer air and it rains harder and it rains more intensely. We certainly are seeing from the data and our physical understanding and the models this increase in precipitation, especially in the extreme precipitation and the very heavy rains, that's certainly a big factor in causing floods, but there are others as well. Development increases the flood risk. We're taking these things and we're paving over areas that used to be natural fields or capture water naturally and keep it from running somewhere else. And now with development that water has to be managed going to go somewhere, and so we have more water coming out from heavier rainfall, we have fewer places for it to go. And so that's definitely the biggest driver. I think if there's one thing that risk managers should focus on in terms of managing their climate related risk, it should be flood.

Patti Harman (09:49):
Yes. And we've had a lot of examples of that this year in terms of flooding and just places where people I don't think expected it. So how are you using science to better analyze the various hazards and develop tools and strategies to help mitigate their risks?

Lou Gritzo (10:09):
So the approach is taking a really hard look at the data that are available, making sure we understand where those data came from and that we're going to use them in the right way to draw some conclusions, physics-based models, and then of course understanding the physics behind those is really the kind of core approach that we use to develop insights. And those insights then can help risk managers determine how to prioritize on top of those things. We have some very practical ways of providing that insight where we take those three things together, we look at the ensemble of climate model predictions and the different climate scenarios, and given the uncertainty in the models and the variability in those scenarios, we could still do this kind of rack and stack of the different major perils on a location basis using some downscaling models that we've developed and our partners have working with us on.

(11:09):

So what that allows us to do is go through and present a climate change impact report that shows for the major apparels what the percent increase is going to be with some fairly broad bans of variability and uncertainty around it. But it allows a risk manager to prioritize. And that's what we're looking to do is to help say, okay, this is what you need to look for in the future. And then most importantly, be able to then say, okay, and here's what things that you can do to better manage those risks, some cost effective measures to manage those risks. But it all starts with making sure that there's a good understanding of the current hazards, a look ahead best we possibly can given the science that we have now, and then starting to focus on the right solutions. And I will say that in the last 10 years, the ability for climate science models to do a credible job of predicting what's going to happen in the future has really gone from black to fairly white. The models now given data from 10, 20 years ago, and up to that point can predict the current climate fairly well. There's fairly good agreement within the top tier models as to 20, 30, 20 50 kind of changes in perils. Whereas 10 years ago, the climate models, they were truly in very different places. You could get whatever answer you wanted to give depending upon which model you chose. And that's of course not what science is all about.

Patti Harman (12:46):
It's really changed the availability of data for risk managers and allows them to identify risks earlier and really take smart steps to mitigate them, I think, in a lot of processes or places. So we're going to take a short break now and we'll be back in just a few minutes. Welcome back to the DIG podcast. We're chatting with Dr. Luso, chief science officer at FM about weather climate change and more so we hear so much about the implementation of artificial intelligence to gather information and sift through data. What role, if any, is AI playing in helping insurers better manage weather risks?

Lou Gritzo (13:34):
Yeah, AI is a great tool and I think it is helpful to maybe put it in a couple of different buckets. There's first the traditional artificial intelligence, which is being used to different degrees for a long time, and it's looking at the data that's available and drawing trends, developing rules from those data. A great example of this is pattern recognition. I see this phenomenon over and over again. I can recognize this pattern, I can use artificial intelligence to say, yeah, that pattern corresponds to this kind of phenomena. That pattern corresponds to this kind of rainfall or that pattern corresponds to this kind of a hazard. And those things are fairly well developed and they're used in increasingly for scientists and engineers to be able to draw better conclusions from the information the app. And so those are great incremental advances of AI that have just continued to occur.

(14:36):

The challenge we're faced with has now generative ai where given a lot of different information, sometimes from very desperate sources, there's newish information that's developed from that and where we have to be, we as a general community concerned about the risks related to the climate, have to be careful that we don't generate things that are not accurate, that don't match that kind of, does it really match the data? Does it really match our physical understanding? Does it really agree with validated models? And so there's a potential pitfall of putting too strong of a stake in specifically generative ai. And I think at the same time, we have to keep in mind that what we're talking about is developing some fairly broad but still actionable conclusions. None of the science or the models, and given the variability of the models and the uncertainty and on top of that, none of those are highly precise. They're not going to give an answer that is a decimal point or two decimal points out. Nobody can predict the future that well. And so especially when it comes to these kind of complex, interrelated, non-linear phenomena of climate science. But AI is a good way for the traditional AI to be used to provide incremental insights and to continue to build that knowledge. And that's what good science is all about. It's about building knowledge based on what you've learned in the past.

Patti Harman (16:21):
And AI, it's just another tool. And I think sometimes people are so willing to just say, oh, well the AI can do this or it can do that. And I know from a journalism perspective, that's something that we're very much aware of. It can change what you want to say or what the information is, or from a generative AI perspective, it can change images and what you're actually seeing. So that would be something to really keep an eye on.

Lou Gritzo (16:50):
Yeah, there's no doubt that is an envogue to say, I used AI and it's an attractive thing, and I understand that, and it is attractive. As you mentioned, Patti, it's a tool, but the question, and I think the challenge is most risk managers have important things to do rather than dig into this is what did you use AI for? How did you use it? Tell me why that's a benefit to what I need to do, which is my day job of reducing risk,

Patti Harman (17:25):
Right? Yep. So it'll be interesting to see how all of this unfolds and how it plays out in the months to come as we see more development and previously uninhabited areas that seems to be increasing the opportunities for more weather related claims. Are there actions that residents, risk managers and even contractors can take to build structures that will be better able to withstand some of these events like wildfires or severe storms or high winds, those sorts of things?

Lou Gritzo (18:03):
Future development is a huge opportunity to really improve resilience. As you start looking at planning future developments, it's an opportunity to stand back and say, okay, where's it going to be? How's it going to be situated? How's it going to be designed, built, maintained? That's exactly the right time to begin to build true resilience. The challenge is the stuff that's already out there, the current exposure, that's the really big challenge and retrofitting that is harder, and we can talk about that. But the new development really is the time to pause and say, am I really building in the right place? The first thing is, is this really a place where I should be building? What are the hazards expected to be in 2030 or 2050 because nobody wants to develop an area that's going to be gone in a few years. And what can I do as part of the design of my development to make it engineered for resilience in a passive way?

(19:05):

In other words, just inherently resilient to the hazard. So for wildfire, that's defensible spaces, open spaces for flood, it's high ground for, there's a lot of different ways that development can occur that makes sure that it is sensitive to future hazards and also development can occur to not be a contributor to those hazards. So porous spaces for groundwater to go. And then of course the FM has a whole host of data sheets that are probably available that can be used to determine what do you need to do in terms of non-combustible construction in terms of wind protection and things. There's a great amount of knowledge already out there. Getting it in the development process is absolutely the most cost effective and best place to do it.

Patti Harman (19:54):
Yeah, that's one of the areas that we've been covering are building regulations and that sort of thing and how they can make a huge difference in terms of the ability to withstand some of the winds and flooding and other things like that. And it really, but it kind of starts at a more granular level I think, at that point.

Lou Gritzo (20:14):
Yeah, for sure. And I think that's something that we talked about hazards making news and people seeing it virtually on a daily basis. You can see something that's happened relative to a hazard somewhere in the world. And so it is producing an increased awareness of the importance of these kinds of considerations in development, and I think that's an opportunity for significant improvements in the future.

Patti Harman (20:39):
Yeah, I would totally agree with that. And we hear so much about the importance of flood and storm protection for homes and businesses. What's the best way to try and approach some sort of client resilience, particularly in flood prone areas? I've had the opportunity to see a number of interesting tools and technologies in this space, even ones that FM has developed. So if you want to maybe share a little bit about that.

Lou Gritzo (21:08):
The flood, as we talked about, again, the world's most frequent hazard and certainly a challenge that's increasing. And the first place to start is understanding what I need to be prepared to face and what's the potential flood levels that I need to be ready to address given my location, given the configuration of my building, given where my values are, and then knowing those things, developing a cost-effective way to protect against that kind of flooding. And so understanding that the water can be a meter high at a one in 100-year level. So it's not going to happen every hundred years. It has a one in 100 chance of happening every year, and that's super important to keep in mind then, okay, how am I going to protect against that? And the ideal thing is to keep the water away and do this in a way that it doesn't require a lot of human intervention.

(22:03):

So it can be swales, it can be designs, it can be configuration of the landscaping or other engineered features. If that's not possible, then there are different flood barriers, perimeter barriers that can be deployed to keep water away when required that combined with the good flood emergency response plan to make sure the people are in place to put those barriers up in the right way so that they function. If the water gets past the perimeter, there are opening barriers that can be used to seal off openings to make sure the water doesn't get into a building or cause damage inside or some kind of critical equipment. Then there's also other measures that you can do inside the envelope of the building. Take that valuable equipment, get it up out of the lower level, get it a little higher, make sure that there's a plan in terms of if water were to come in the building to limit the amount of damage it does, and then a response plan to respond afterwards.

(23:03):

So we do have an FM approval, standard FM 25 10. It's also American National Standards Institute and an ANSI Standard for testing those barriers. And the one thing I would emphasize here is that with flood, if you put up a perimeter barrier and one piece of it doesn't hold, everything you've done is for naugh, the whole thing fails once that water gets in through any one little area. We've seen this with levee breaks. It just breaks in one area, then it's as if the levy weren't there at all. So it's really important to make sure that the means that are used are proven, that they're proven to be durable, that they're proven to be effective, they can handle things like wave impact and current impact. Flood carries a lot of debris along with it, debris impact. And so those are important things to consider, not just, well, I'll just put something up to keep the water out, but make sure that it's actually going to work, otherwise we've gone to all that effort and you still experience the flooding, which is not what anybody wants to have happen.

Patti Harman (24:09):
Yeah, that is so true. I am based in Maryland, and Annapolis has a beautiful waterfront area, and I think at least two or three times this year, the tides rose so high that it flooded all of the storefronts along the main walkway there and everything. And you could see people just trying to, and it's more than sandbags at this point. There are so many other different types of technology that are available that, and you're right, it only takes one leak or one weak spot somewhere, and then it really is kind of all for naugh. So one of the aspects of what you're doing that I find really interesting is the application of science and data to address some of these risks. How has that changed over the years and what do you find particularly exciting at this point in time,

Lou Gritzo (25:03):
Patti? I think the most exciting thing at this point in time is the ability of us to get better data through geospatial information. We have better geospatial data now from satellites, from flyover data, higher resolution in digital formats. So we understand a lot more every year about the earth we live in and what it looks like and what's happening. And that's enabling a lot of progress that combined with physics-based models that use that information and can forecast that into the future. When we look at these climate models, they're basically taking everything from the sea surface or the ground surface up and putting it into little bins and calculating all the heat, mass transfer and air flows and everything through that. And it's very complicated, but those models continue to get better. And the big challenge within those models is that the resolution is still fairly coarse.

(26:08):

The computing progress is making huge advances in that graphical processing unit. Computing is allowing significant speed up in a lot of those calculations. So it's better. It's more than just for video games. The way it processes information is also great for science. And so that's super exciting to happen. And then this kind of convergence on bringing the science together and being able to come up with these common model into comparison projects that bring those models together and give a good set of actionable results from them. That's what we really need to be able to look into the future. And that's what we all need to prepare for. We need to think about what we can do now that positions us for the future rather than to respond to the hazards that are happening before us. And if we can get on top of that, then we'll find ourselves in a place where we're truly adapting to whatever the climate is throwing at us rather than simply responding.

Patti Harman (27:09):
You're right, you need a broader focus and just a more long range approach, I think in money areas. Is there anything about the increase in storm severity or other weather related risks that concerns you or that you're watching to kind of see what's going to develop in the next year or so?

Lou Gritzo (27:30):
The biggest concern across the globe is coastal exposures to flooding. The sea levels are continuing to rise. They're rising more around the equator. They're not rising uniformly. So again, this is one of those situations where it's very location specific, but especially in the lower latitudes, rising sea levels, we have in some cases land subsidance where for whatever reason, whether it's groundwater extraction, the land's actually going down, seas are coming up, water's getting closer. Like your example of storefronts maybe in Maryland that combined with this increased development in those areas and higher precipitation rates from storms, so those same areas of lower latitude have warmer air temperature. Those storms hold more water vapor. It's raining harder, it's raining longer. And so all of these things are just piling water on top of areas around the coast, whether it's storm surge from a tropical storm or a hurricane that's coming in, whether it's that combined with the river that's flooding and the rainfall coming down from the storm and the higher sea levels. This is just kind of an increased development. This is a bunch of factors coming together that I think is probably the biggest sole challenge that we face in dealing with climate resilience right now and certainly into the future.

Patti Harman (29:02):
Wow, that gives us a lot to think about going forward. And we've covered a lot over the last half hour or so. But is there anything related to climate or addressing climate risks that you think is important for our listeners to know, or something that I haven't asked you at this point?

Lou Gritzo (29:20):
I think the most important thing for us all to keep in mind is that there's going to come some times, especially in this coastal exposure issue, but we're going to have to make some tough decisions in terms of development, what areas are going to be developed, what areas are not going to be developed, the areas that continue to show and have in the past, these kinds of hazards that are getting more severe in a changing climate, that it's cost effectiveness of being able to reduce those risks is just not something that is going to work out for us. And so that's the challenge that needs to be faced right now is what do we do about those? How do we protect what we have now and then how do we think about how we make it sustainable in the truest sense of the word, which is ways that we can succeed as a society going forward in the future, into the future. And so I think for any risk manager or manager of municipality or person buying a home or investing in a business, those are the things that I think are super important for us to think about because the decisions we make today are going to drive the big challenges that we have in the future.

Patti Harman (30:35):
Yes, they definitely will. Thank you so much, Lou, for sharing your insights with our audience. Thank you for listening to the Dig in podcast. I produced this episode with audio production by Kellie Malone Yee. Special thanks this week to Dr. Lou Gritzo of FM for joining us. Please rate us, review us, and subscribe to our content at www.digin.com/subscribe. From Digital Insurance, I'm Patti Harman, and thank you for listening.