Scientists of the world unite

Summary

Science In Action: Scientists of the World Unite BBC World Service | Release Date: May 1, 2025

Overview

In this episode of Science In Action, the BBC World Service delves into critical discussions from the European Geophysical Union's annual conference in Vienna. The episode navigates through pressing Earth science topics, including the potential collapse of the Atlantic Ocean's circulation, the historical impact of carbon dioxide on climate, seismic risks in Istanbul, and innovative methods for detecting natural hazards. Through expert interviews and insightful presentations, listeners gain a comprehensive understanding of the current scientific landscape and its implications for our planet's future.

---

1. European Geophysical Union Conference: A Hub for Earth Sciences

The episode sets the stage at the European Geophysical Union's annual conference in Vienna, uniting over 18,500 scientists from around the globe. This gathering covers a vast array of Earth science topics, from planetary origins to climate futures. Roland Pease, the host, expresses the challenge of encapsulating the conference's breadth but highlights a focus on climate change and its potential tipping points.

"The unifying theme here is Earth Sciences... but there is one topic I really wanted to get to that's the future of our climate."
[02:03] Roland Pease

---

2. Atlantic Ocean Circulation: Approaching a Tipping Point

Interview with Stefan Ramsdorff

Stefan Ramsdorff, head of department at the Potsdam Institute for Climate Impact Research, discusses the critical state of the Atlantic Ocean's overturning circulation—a major current that warms Northern Europe by transporting heat from the tropics. Ramsdorff warns of the current's weakening due to global warming, stressing the possibility of a "tipping point" that could lead to its collapse.

"We fear that the tipping point will actually already be crossed in the next few decades, by the middle of this century."
[05:10] Stefan Ramsdorff

Ramsdorff explains that crossing this tipping point could result in severe climate alterations, including:

• Cooling of Northern Europe: Reduced heat transport would lead to cooler temperatures.
• Extreme Weather Events: Increased frequency and severity of weather extremes.
• Sea Level Rise: Potentially an additional meter rise in the North Atlantic region.
• Shifts in Rainfall Patterns: Disruption of tropical rainfall belts, affecting rainforests and increasing flood risks.

He emphasizes the urgency of reducing global emissions to prevent crossing this tipping point.

"It's not easy, it's inevitable and it will really change the face of Europe."
[05:35] Stefan Ramsdorff

---

3. Historical CO₂ Levels and Climate Change

Presentation by Hannah Jerikova

Hannah Jerikova, an isotope geochemist from the University of St. Andrews, presents research on historical CO₂ levels and their impact on Earth's climate. Her study reconstructs CO₂ concentrations during the Phanerozoic Eon, particularly during the "ice house" interval spanning the Devonian to Permian periods.

"When your CO2 rises, your climate changes. And that's an important lesson."
[14:15] Hannah Jerikova

Key findings include:

• Low CO₂ Levels: Approximately 300 ppm during much of the ice house interval, significantly lower than today's ~430 ppm.
• Extended Ice Caps: Ice covered large parts of the planet, potentially reaching the tropics.
• Abrupt Climate Transition: A rapid rise in CO₂ to about 900 ppm, likely due to volcanic activity, led to the melting of ice caps and a subsequent rise in sea levels.

Jerikova underscores the parallels between past and present, reinforcing the role of CO₂ in driving climate change.

"We know from measurements today that it's doing it today, but it's done that even hundreds, 300 million years ago in Earth's history."
[13:42] Hannah Jerikova

---

4. Seismic Risks in Istanbul: Understanding Earthquakes

Interview with Patricia Martinez Carson

Patricia Martinez Carson, a professor at the GFZ Helmholtz Centre for Geoscience in Germany, provides insights into the recent seismic activity near Istanbul. A magnitude 6.2 earthquake occurred under the Sea of Marmara, highlighting the region's vulnerability.

"The time is unknowable. We know that it will most likely happen at some extent, but we don't know when."
[23:58] Patricia Martinez Carson

Key points discussed:

• Fault Behavior: The Marmara Fault exhibits both creeping (slow movement) and locked segments, with the latter posing significant earthquake risks.
• Stress Redistribution: Recent earthquakes alter stress distribution along the fault, potentially triggering larger quakes.
• Energy Propagation: Studies indicate that seismic energy tends to move northeast towards Istanbul, increasing the city's exposure to strong shaking.
• Historical Context: The North Anatolian Fault has seen a series of large earthquakes migrating westward, with the Sea of Marmara segment remaining a concern.

Carson emphasizes the importance of continuous monitoring and preparedness, given the unpredictable nature of seismic events.

"Seismic waves might be larger in the direction of Istanbul if a large earthquake nucleates."
[23:35] Patricia Martinez Carson

---

5. Innovative Detection of Geomorphic Events Using Bio-Inspired Algorithms

Presentation by Joseph

Joseph, a researcher from Potsdam, introduces a novel method for detecting and locating geomorphic events such as landslides, debris flows, and volcanic eruptions. Inspired by animal behaviors, Joseph's algorithm enhances the ability to interpret complex seismic signals.

"It's ready to be applied in early warning systems. It needs a little bit more stability that comes with testing it in more data."
[30:21] Joseph

Highlights of the approach include:

• Bio-Inspired Agents: The algorithm mimics behaviors like a falcon's dive or a whale's migration to efficiently search and locate seismic events.
• Simplified Processing: Unlike traditional machine learning models, the algorithm is lightweight, requiring minimal data and computational resources.
• Real-Time Application: Capable of running on live seismic data feeds, enabling rapid detection and early warnings.
• Practical Implications: Potential to provide critical lead time (e.g., 45 minutes) for evacuations and disaster response, thereby saving lives.

Joseph underscores the algorithm's efficiency and its readiness for integration into existing seismic monitoring systems.

"It's both accurate and fast, which is crucial for rapid perception of natural hazards in an accelerating climate change condition."
[28:57] Joseph

---

Conclusion

This episode of Science In Action provides a compelling exploration of current Earth science challenges and innovations. From the looming threat of disrupted ocean currents and historical climate lessons to the seismic vulnerabilities of major cities and breakthroughs in hazard detection, the discussions underscore the intricate interplay between scientific research and global sustainability. As climate change accelerates and natural hazards become more frequent, the insights shared by these scientists highlight the urgent need for informed action and technological advancement.

Notable Quotes:

• Stefan Ramsdorff: "A tipping point is more like when you lean back on your chair and you lean back too much... and we fear that the tipping point will actually already be crossed in the next few decades."
  [05:10]

• Hannah Jerikova: "When your CO2 rises, your climate changes. And that's an important lesson."
  [14:15]

• Patricia Martinez Carson: "We know that it will most likely happen at some extent, but we don't know when."
  [23:58]

• Joseph: "It's ready to be applied in early warning systems. It needs a little bit more stability that comes with testing it in more data."
  [30:21]

---

Stay Informed: For more detailed discussions and updates from the European Geophysical Union conference, visit the Science In Action webpage at bbcworldservice.com.

Transcript

[Roland Pease]

This BBC podcast is supported by ads outside the uk.

[Amica Insurance]

At Ameca Insurance, we know it's more than just a car or a house. It's the four wheels that get you where you're going and the four walls that welcome you home. When you combine auto and home insurance with Amica, we'll help protect it all. And the more you cover, the more you can save. Amica Empathy is our best foreign.

[Ryan Seacrest]

And I was on a flight the other day playing one of my favorite social spin slot games on chumbacasino.com I looked over the person sitting next to me and you know what they were doing? They were also playing Chumba Casino. Coincidence? I think not. Everybody's loving having fun with it. Chumba Casino's home to hundreds of casino style games that you can play for free, anytime, anywhere, even at 30,000ft. So sign up [e-mail] to claim your free welcome bonus. That's chumbacasino.com and live the Chumba life. No purchase necessary VGW or prohibited by law.

[Amica Insurance]

See terms and conditions 18.

[Joseph]

Hi, I'm from Vietnam but now I'm working in Germany. Yeah, my study is related to the atmospheric chemistry. Hello, my name is Joseph, I'm from the University of Toronto.

[Roland Pease]

I study Cloud microphysics and I'm coming over from Canada. Hi, I'm from China Macau.

[Stefan Ramsdorff]

My major is the Hydrology science and.

[Roland Pease]

I'm going to do some research to.

[Stefan Ramsdorff]

Help the risk of the flood Risk mitigation. My main discipline is Atmospheric sciences and.

[Roland Pease]

I'm from the Netherlands.

[Joseph]

Me I am a physicist.

[Patricia Martinez Carson]

I have done my PhD in atmospheric.

[Joseph]

Sciences and currently I'm working in the.

[Patricia Martinez Carson]

National center for Meteorology.

[Roland Pease]

I'm from University of Estonia in the.

[Patricia Martinez Carson]

Baltics, Northern Baltics and I'm working in the boreal peatlands and tropical peatlands and.

[Roland Pease]

I'm a broadcaster from the BBC World Service. With 18 and a half thousand people from all over the planet milling around this conference center, we could fill the whole of this edition of Science in Action, detailing each of the participants credentials. But the unifying theme here is Earth Sciences. This is the annual conference in Vienna of the European Geophysical Union where they're covering everything from the origin of the planet to the future of our climate, from the rocks beneath our feet to the sun and planets above our heads, from fossils to forests. And the only trouble is that there's only one Roland Pease and I can only get to a tiny fraction of what's on offer here. But there is one topic I really wanted to get to that's the future of, of our climate. In particular, a debate that we've been covering on science and action that's sharpening over whether changes in Atlantic circulation could have dramatic impacts in coming decades.

[Stefan Ramsdorff]

My name is Stefan Ramsdorff. I'm head of department at the Potsdam Institute for Climate Impact Research in Germany and I study the Atlantic Ocean overturning circulation.

[Roland Pease]

The current that brings warm water to the north of the Atlantic and warms the Northern hemisphere.

[Stefan Ramsdorff]

And particular, yes, this is the big heating system for Europe because it releases huge amounts of heat out of the ocean into the atmosphere in the northern Atlantic, which then warms Europe.

[Roland Pease]

So we were talking to you, I guess a couple of years ago, the last time we spoke to you, and your concern is that the changes in the climate because of global warming are going to stop that happening, which will have big climatic effects. You were just bringing us up to date with your latest results.

[Stefan Ramsdorff]

We know this current is weakening and this is the reason why. There is already a cooling in the northern Atlantic sea surface temperatures. It's not overland yet, but it's already affecting our weather to some extent. And there is a prediction by climate models that it will weaken further. And we know there is a tipping point that it could cross where it then essentially collapses so that this heat transport, heat release to the west of Britain stops. And that would have a massive impact not only on European climate, but around the world.

[Roland Pease]

I mean, I saw that you've now taken these projections beyond the year 2100, which is where things look quite dramatic.

[Stefan Ramsdorff]

Yes, the IPCC has said that it's unlikely that this current will collapse by 2100. But some of the model runs have been extended beyond the year 2100. And quite a few of those actually do show the current collapses, especially all of those for high emissions.

[Roland Pease]

There's a key word that you've used, tipping point. Understand that this is not like putting your foot on the accelerator or taking it off. If this collapses, you can't just sort of restart it again, this current.

[Stefan Ramsdorff]

Exactly. A tipping point is more like when you lean back on your chair and you lean back too much, up to a point, then it tips over and you can't do anything about it anymore. And we fear that the tipping point will actually already be crossed in the next few decades, by the middle of this century. And then it becomes inevitable that the further collapse basically unfolds over the next 50 to 100 years.

[Roland Pease]

Okay, you're saying that you can see that this circulation is weaker now than it was say 100 years ago. And that already seems inevitable. And this is a slow process. So even if it started, it's not that you expect all the impacts to be visible tomorrow.

[Stefan Ramsdorff]

That's right. It's not happening within 10 years or a few weeks like in that movie the Day After Tomorrow. It unwinds slowly, but it then is inevitable and it will really change the face of Europe.

[Roland Pease]

So what I'm wondering is, when is that point of no return that you're talking about?

[Stefan Ramsdorff]

That's the billion dollar question. We don't know where this point of no return is, this tipping point, but we used to think it is very unlikely to cross it this century. Now the more recent results suggest that that was too optimistic. And actually we are on track to cross it already in the next 20, 30 years or so if we don't rapidly reduce global emissions.

[Roland Pease]

It's not just that Europe would cool down because of the loss of this transported heat, but that actually the weather systems change, which I think is probably much more important to people.

[Stefan Ramsdorff]

We will be seeing extremes that we haven't seen before. The sea level will rise by an additional meter, up to a meter around the North Atlantic. In addition to the global sea level rise that we're also witnessing, the tropical rainfall belts will shift to the south so that the tropical rainfall is not where it used to be, which is really bad for rainforests. And then the tropical rainfall will be elsewhere where people are not used to it, the drainage systems are not used to it. So that will cause problems, problems with flooding. So this is really in Earth history. We have seen similar things happen during the last ice age, and those are the most dramatic climate changes that we know from paleoclimate that can unfold in over a question of decades.

[Roland Pease]

It is very apocalyptic. We've spoken to you about it before with these new models. One, are you more certain than you were before that this will could happen? I'm not quite even sure how to phrase this because it's still a hypothesis in a sense. And recognizing the detail of our approach to this seems to be incredibly important.

[Stefan Ramsdorff]

Yeah, it is a risk and we used to consider it a low probability, high impact risk. So like less than 10% chance of this happening. But because the impact would be so severe, we need to take it very seriously now. We don't consider it low probability anymore, unfortunately. I would say it's like almost 50, 50 that we will reach this tipping point in the next decades, depending on what we do to emissions.

[Roland Pease]

And last of all, we did speak recently to a team who have A different view from the Met office in the uk and they're saying that there are outside influences which will moderate or could moderate the kinds of processes you're talking about. They discuss that here, this conference as well. Is what's the source of this difference of opinion? Is there a way of, Is there some intermediate ground somewhere or how do.

[Stefan Ramsdorff]

You resolve that there is more a difference in terminology? There's not really a difference in climate impacts. They are the same models these people are using and they just pointing out that this overturning circulation will not go to completely to zero in what we call a collapse. But the climate impact is the same. It will become very weak, very shallow, not reach so far north. So that heat release that warms Europe and what you are talking about, that will still basically collapse even though the overturning in the ocean doesn't completely vanish, it just greatly weakens.

[Roland Pease]

Stefan, thank you very much indeed. Go and get that coffee. Stefan Ramsdorff, reinforcing concerns of an Atlantic tipping point sometime in the next few decades. A climate denialist talking point is that climate has always changed so that what we are doing to the atmosphere now is unimportant. But in the session on deep climate history, one talk really underlined that adding carbon dioxide to the atmosphere dramatically does warm the earth.

[Hannah Jerikova]

Happy to answer any questions, thank you very much. Hello, I'm Hannah Jerikova, I'm an isotope geochemist. I'm from the School of Earth and Environmental Sciences in the University of St. Andrews. And we published a paper in January this year that reconstructed the CO2 from one of the most dramatic climatic transitions that we have over the last 500 million years. So in the Phanerozoic. So we're talking about a huge ice house interval that happened sometime from the Devonian, span much of the Carboniferous and ended in the Permian. And we reconstructed CO2 using a novel approach that previously was thought that was challenging to apply so far back in time. And it showed that actually for much of this ice house, CO2 remained very low. We're talking about 300 ppm CO2. So today in the atmosphere we have around 430, we're going towards 430 ppm. And so CO2 was much lower and ice extended all the way to potentially the tropics. And this ice house conditions lasted for tens of millions of years and then they abruptly, relatively geologically speaking, abruptly came to an end within only few millions of years. In the beginning of Permian, we specifically pinpointed to 294 million years ago. Because of a rise in CO2 from volcanic activity. So this really solved and kind of put together the kind of climatic conditions that were responsible for this ice house and sort of solved many questions around it.

[Roland Pease]

And so this ice house condition. So this was. The world was basically largely frozen, the ice caps spread over much of the planet until suddenly you're saying there was a huge rise in CO2 and that melted it.

[Hannah Jerikova]

Yes, yes. So one of the problems is that the evidence we have for the existence of the ice caps comes from the geological record, and the geological record is not complete. So there are potential gaps in the evidence of glacial deposits existing. But we do have relatively good evidence from many continents that ice kind of extended almost towards the tropics. So we had an ice cap on the south. We're not sure about ice on the North Pole because we don't have many good records and there was not much continental mass on the very north. But we would have, we are expecting quite a large sort of ice cap, at least in the. In the south.

[Roland Pease]

So, I mean, the reason I was really interested to see your talk is because, as I understand it, you're saying that it was the carbon dioxide that came from the Earth that changed the atmosphere, that changed the climate, which is, of course, what we're doing now. But people are skeptical about that. But it's deep geology you're talking about. Big, deep time.

[Hannah Jerikova]

Yes, that is true. I mean, we really were able to show that as soon as the CO2 rose. That's the time when the ice house interval disappeared. We also see evidence in the geological record for rise of sea level. So your ice melting and sea level rising. One thing to really bear in mind, though, is that kind of these sort of geological reconstructions, they happen over a long time. So we have slightly different timeline. Well, pretty different timeline to today, but we can really see this tight coupling between climate and CO2, and that CO2 plays a key role. We know from measurements today that it's doing it today, but it's done that even hundreds, 300 million years ago in Earth's history.

[Roland Pease]

And how much do you think it rose by? You're saying it started about 300 ppm? I'm giving away my age. That's probably not far off where it was when I was born some decades ago. How much did it go up to make this transition?

[Hannah Jerikova]

So we, our records show that it went towards 900 ppm. Again, we're talking about huge amount of CO2. Yes, that's right.

[Roland Pease]

So you probably trebled.

[Hannah Jerikova]

Yes.

[Roland Pease]

The amount of CO2 in the atmosphere.

[Hannah Jerikova]

But it happened over long periods of times, millions of years. So it's a bit hard maybe to.

[Roland Pease]

Imagine we're doing it way faster. The changes we're talking about have been over 50 years. Your geological record is over millions of years. So that's quite a lesson, I think.

[Hannah Jerikova]

Yes, yes, absolutely. I mean, it's very hard really to sort of comment what this, you know, we're talking about today about changes of 200 years. So it's very hard to sort of compare that as an apple and apple to the geological record. But one thing we know for sure from the geological record, that when your CO2 rises, your climate changes. And that's an important lesson.

[Roland Pease]

With unseasonably warm weather in Vienna outside the conference, all bright, sun bathing the city, it feels a bit odd contemplating the lessons from that talk as it echoes in my ears. But that's what happens when you take science in action out to an international geoscience conference.

[Amica Insurance]

At Ameca Insurance we know it's more than just a car or a house. It's the four wheels that get you where you're going and the four walls that welcome you home. When you combine auto and home insurance with Amica, we'll help protect it all. And the more you cover, the more you can save. Amica empathy is our best policy.

[Roland Pease]

Foreign.

[Ryan Seacrest]

It is Ryan Seacrest here. There was a recent social media trend which consisted of flying on a plane with no music, no movies, no entertainment. But a better trend would be going to chumbacasino.com it's like having a mini social casino in your pocket. Chumba Casino has over a hundred online casino style games, all absolutely free. It's the most fun you can have online and on a plane. So grab your free welcome bonus [e-mail] sponsored by Chumba Casino.

[Amica Insurance]

No purchase necessary vgw group void we're prohibited by law 21 + terms and conditions apply.

[Roland Pease]

Last week, just before we made our way here, a strong though not devastating earthquake under the Marmara Sea by Istanbul reminded me of. Concerned seismologists have long held that a larger part of the fault there might fail catastrophically someday. I was sure someone here could give me a better understanding of what they mean and what just happened.

[Patricia Martinez Carson]

Alright, so my name is Patricia Martinez Carson. I am a professor at the GFZ Helmholtz center for Geoscience in Germany, although I am originally from Spain. And so basically what happened last week it was earthquake of magnitude 6.2 in the middle portion of the main Marmara fault in the Sea of Marmara near Istanbul, in a kind of transition of the fault between a section that it is slowly moving or creeping and part of the fault that it is fully locked. The slip of the fault for a magnitude 6.2 earthquake can be approximately half a meter of slip. And it can rapture a length of about 30 or 35 kilometers of the fault. And time, of course, it depends a little bit on how far away are you from the source. But at the source, I would say at least between half a minute and one minute.

[Roland Pease]

So this is, as it were, the two bits of the earth, the two plates sort of sliding past each other like that by about half a meter. I mean, these things always sound so dramatic. The amount of rock that's moving and that's what causes the shaking that you then feel.

[Patricia Martinez Carson]

That's right.

[Roland Pease]

But this is much less than the earthquake that people are worried about that could hit very near Istanbul.

[Patricia Martinez Carson]

Exactly. So as we know, the North Anatolian Fault is a major plate boundary. And over the last century there has been a migration starting from the east towards the west of large earthquakes starting in 1939 on the Eastern part in a city called Erzinja. And then almost the entire fault ruptured in a number of magnitude 7 earthquakes that culminated into this Ismit Andurche earthquakes, magnitude 7.4 in 1999.

[Roland Pease]

And since then I've sort of been waiting, when does the next bit go exactly?

[Patricia Martinez Carson]

That left the Marmara fault or the Sea of Marmara part of the fault as the only portion of the fault that has not raptured in a major earthquake since approximately 1756 or so. And that has of course, the potential, potential to rupture.

[Roland Pease]

So this earthquake that happened last week, this was sort of the other side of that locked bit of force, is that right?

[Patricia Martinez Carson]

It's a bit in the middle part of it, because the entire Sea of Marmara could in principle ruptured. But it is true that it occurred on the western part of the fully locked segment that it is in front of Istanbul, the so called Princess Islands segment. And that is more than 20 km long piece of fault that is directly in front of Istanbul and that appears to almost host no seismicity.

[Roland Pease]

Now, I've had this impression in the past that what determines the size of an earthquake is it starts and then it just keeps going, it keeps spreading until it stops, in principle. Was Istanbul sort of lucky last week that this earthquake didn't just go along the whole fault?

[Patricia Martinez Carson]

Yes, well, as you know, the earthquakes stop when they basically Run out of fuel when they run out of energy to continue propagating the Raptor. In that sense, what made this earthquake to stop, we don't know exactly, obviously, but it could have been heterogeneity on the fault that made it a bit.

[Roland Pease]

Of roughness or something, or a bend.

[Patricia Martinez Carson]

In the fall, a bit of geometry constraint that allowed the Raptor not to propagate further.

[Roland Pease]

So it could have been worse?

[Patricia Martinez Carson]

It could have been worse, of course.

[Roland Pease]

But does it, does that then change the conditions on the bit that I'm saying I'm worried about?

[Patricia Martinez Carson]

Yeah. So the occurrence of this Raptor, what it does is that it releases the stress or pain part of the stress on the part of the fault that ruptured. But at the edges, of course, the stress increases because it brings these edges closer to failure in what we call the coulomb stress. So the coulomb stress on these edges increases, and that could have, of course, advanced the stress on this part of the fault that is about to rupture in a larger earthquake.

[Roland Pease]

And when you Talked about the 1999 earthquake, that was at the other end and that also stressed the fault in between.

[Patricia Martinez Carson]

So the Izmit earthquake is to the east, and then there was, on the other western end, there was a magnitude 7.1 in 1912. It's called the Ganos earthquake.

[Roland Pease]

I'm wondering if there was an earthquake at one side of this fault, the Marmara fault, 25 years ago. You've just had another one, this side. Does this mean that the stresses are being added to in both directions?

[Patricia Martinez Carson]

Absolutely, yeah.

[Roland Pease]

And if an earthquake starts, it could start at either end and then just spread across the whole fort. Does that matter, you know, to Istanbul, to the kind of shaking it can get?

[Patricia Martinez Carson]

Yeah. So, you know, because of the, as you said it, very good. There were these two large earthquakes on each side of the fault. And that makes that the whole fault has like a difference of stresses that are lower in some parts where the earthquakes occurred and larger in the part that has not ruptured yet.

[Roland Pease]

One reason I'm asking this, we were talking about the Myanmar earthquake just a couple of weeks ago, which was very directional, and it actually hit the cities in Bangkok, the shaking, because the, as it were, the shock waves from the earthquakes was going that direction. I'm just wondering if it matters to Istanbul what way the shaking propagates once this earthquake starts to move. Is that a sensible question?

[Patricia Martinez Carson]

Yeah. So that's a very good question. And thanks a lot for asking that, because we have a recent study on this where we looked at how earthquakes on this main Marmara fault propagate or towards where do they send their larger proportion of energy. And what we saw is that statistically they tend for this part of the fault, they tend to throw the largest amount of energy towards the northeast, which is unfortunately oriented towards the city of Istanbul. And that could mean that if a large earthquake is to nucleate on this part of the fault, ground shaking might be larger in the direction of Istanbul.

[Roland Pease]

When I visited Istanbul to talk about these things 20 years ago or so, it was I think, 11 million people. It's an even bigger city now.

[Joseph]

Yeah.

[Patricia Martinez Carson]

To my knowledge it oscillates between 16 to 18 million of people.

[Roland Pease]

That's why I guess experts like yourself are very worried about what would happen. Well, what will happen? I suspect we just don't know when, if, when this fault goes.

[Patricia Martinez Carson]

Yes, the time is unknowable. We know that it will most likely happen at some extent, but we don't know when. And that's why we need to continue monitoring the fault as much as possible over time.

[Roland Pease]

Yeah. So not necessarily next week or the week after next year. Not even necessarily necessarily the next decade, but you always have to be ready for it.

[Patricia Martinez Carson]

Exactly. That's the case.

[Roland Pease]

We've put a link to the program. I made some time back on the threat to Istanbul on the Science in action [e-mail] Staying with seismic waves, but using them as a way to listen to other earth hazards. This was a talk I just didn't feel I could resist.

[Joseph]

My name is Stefania Orsica from Jeffs at Potsdam, Germany. And we want to basically try to use animal behavior as a blueprint to sharpen our ability to locate geomorphic events.

[Roland Pease]

Such as landslides, sort of computerized animal instincts, as it were. But when you say geomorphic events, can I translate that? That's things like landslides, floods, things like that?

[Joseph]

Yes. From landslide, debris flow, rock falls, rock slides, pyroclastic flows, volcanic eruptions.

[Roland Pease]

I talked to people at previous ETU events about this, that you can use seismic waves that they generate and you can pick them up with seismic sensors. Is that the techniques you're using?

[Joseph]

Exactly. So there are stations worldwide that basically monitor non stop and they do capture geomorphic events all the time. And we are technically able to detect and locate. Now it's a little bit tricky, but what we propose now is a new method that takes inspiration from biology to use a very innovative technique to detect and locate these geomorphic events.

[Roland Pease]

I'm Surprised? I imagined it was pretty, you know, simple to take. You know, a seismometer here detects a bit of a rumble, another one over here does, and you can work out where the event is happening. But it's not that easy.

[Joseph]

It's not that easy. Earthquakes are easier to detect and locate because they are clean, predictable, and they have a crisp wave front. Now the surface events produce an emergence smeared and non linear chaotic signal that's really hard to detect and to locate. But what we can do is to actually learn that by instinctive methods, like biology does, like animals do, like understanding the signal from multiple perspectives instead of just looking at one attribute. This is an algorithm that is very simple and elegant. It does not need a lot of data and no training. And it's not machine learning or AI, but it does use, I would say, biological intelligence to create emergent behaviors.

[Roland Pease]

I guess you'd just better explain that more to me.

[Joseph]

What we do is actually encode mathematically animal behaviors in different agents that try to search complex landscapes. And we basically look at biology that have had millions of years to try and fail to find the best strategies to search and find in such landscapes.

[Roland Pease]

I mean, would this be something like, let's say a bird of prey, an eagle or something that's trying to find a mouse in the, in the scrub?

[Joseph]

Exactly. So let's take the falcon mimicry, for example. Falcons are known to dive with a spiraling inward movement. We can model that with a logarithmic spiraling in with a certain speed that decays because of air drag and pre evasion. For example, in our case, the prey would be the best location of that geomorphic event. And so we train like an agent that will act like a falcon and will try to locate the prey that is the location.

[Roland Pease]

So you're sort of synthesizing, you're taking in all these different rumbles and you're sort of getting just a rough idea in the first instance, but then you have a way of refining it and refining it.

[Joseph]

Exactly. So we start with exploration and then we get to exploitation. For exploration, we can use something like a migrating whale that is very good in searching vast spaces. Then we can communicate through bees or through fireflies that use light. Then we can micro adjust with a hummingbird, for example.

[Roland Pease]

Have you sort of got these things programmed into your computer or is that just a way of talking to me?

[Joseph]

No, it's not an analogy. These are all mathematically encoded and they are biologically grounded, but with an equation within my algorithm.

[Roland Pease]

So you've learned from Nature, what difference does this make? What's the important thing? Is it how accurate you are, how fast you are, or both or something else?

[Joseph]

Both. And actually, even more than that, biology has this wisdom of actually having the best strategies for search, for memory, for precision, accuracy, and for being fast. And we need all of this for rapid perception of natural hazards in an accelerating climate change conditions.

[Roland Pease]

Okay. Because one of the things I've been really interested in over the years is I think we're getting more of these things called glacial outburst floods in the Himalayas, where you get a rock collapse or something, and then there's a lake that's been held back, rushes down these very steep valleys. They've been very deadly. But there's not been a proper way to let people know the flood's coming towards you.

[Joseph]

I love your example, because imagine a landslide happens or an outburst, but there is a time lag, a precious pause that we can use to warn early people to save lives, to actually intervene in a targeted and rapid way. Now, we can use that if we can quickly detect on a seismic data and locate and let people know they do have, let's say, 45 minutes to evacuate an important town and save probably many lives.

[Roland Pease]

I mean, are you at that point now, or at what point do you go from presumably historic or toy examples that you're training on to trying this in real life?

[Joseph]

Well, that's the beauty of this algorithm. It does not need training. So I don't need a historic catalog. It's modularized. So it's a very simple way of not using parameters or catalogs. You just use instincts, instincts of animal behavior. So basically, it's kind of ready to be applied in early warning systems. It needs a little bit more stability that comes with testing it in more data. But basically it's.

[Roland Pease]

But do you think it will be easy to hook it up to real seismometers? I mean, I'd love to know that this actually works in the real world.

[Joseph]

Although there are 3,000 lines of code, it takes 10 seconds to run it. It's super light. So that means we will be able to put it in an API so it can run live on real seismic data.

[Roland Pease]

If something like this could be used to get faster warnings to people in the path of some kind of disaster or to get the rescuers to the right place as quickly as possible, that's got to be a good thing. And that is it for science in action. From this meeting of the European Geophysical Sciences Union, I'm Roland Pease. The producer is Alex Mansfield. We'll be back in the BBC studios next week with another edition.

[Amica Insurance]

At Ameca Insurance, we know it's more than just a car. It's the two door coupe that was there for your first drive, the hatchback that took you cross country and back and the minivan that tackles the weekly carpool for the cars you couldn't live without. Trust Ameca Auto Insurance Amica. Empathy is our best policy.

[Ryan Seacrest]

It is Ryan Seacrest here. There was a recent social media trend which consisted of flying on a plane with no music, no movies, no entertainment. But a better trend would be going to chumbacasino.com it's like having a mini social casino in your pocket. Chumba casino has over 100 online casino style games, all absolutely free. It's the most fun you can have online and on a plane. So grab your free welcome bonus [e-mail] sponsored by Chumba Casino.

[Amica Insurance]

No purchase necessary.

[Roland Pease]

VGW Group Void where prohibited by law.

[Amica Insurance]

21 + terms and conditions apply.