Coral extinctions and chalky unknowns

Summary

Science in Action – “Coral Extinctions and Chalky Unknowns”

BBC World Service – October 23, 2025
Host: Ronan Pease

---

Episode Overview

This episode delves into three major scientific stories:

• The alarming loss and potential “functional extinction” of staghorn and elkhorn corals in Florida, a grim indicator of global reef health as climate change quickens.
• The enormous but uncertain role of microscopic plankton and mollusks in storing carbon—key players in oceanic carbon cycling whose future responses to climate change are unknown.
• An unexpected, encouraging finding: mRNA Covid vaccines seem to boost immunotherapy outcomes in cancer patients.
• A creative, low-tech way to map crops in Africa using GoPro cameras for machine learning and food security.

Ronan Pease guides us through these stories with discussions featuring marine biologist Ross Cumming, biogeochemist Patrizia Zaveri, immunologist Adam Gripen, and remote sensing scientist Catherine Nakalembe.

---

1. Coral Extinction in Florida: Functional Loss Amid Climate Crisis

The Plight of Florida's Coral Reefs

[01:57–13:07]

• The 2025 Global Tipping Points Report has declared the loss of warm-water coral reefs as imminent, especially if global temperatures surpass 1.5°C above pre-industrial levels.
• A new Science paper documents the “functional extinction” of staghorn and elkhorn corals in 2023 across Florida's 560-kilometer reef tract.
• These branching coral species provide essential three-dimensional habitats for countless marine species.

Key Insights:

• Coral Biology and Importance

  • Staghorn corals: Form long, spiky branches—“large thickets... produce this highly complex three-dimensional structure that provides habitat for many other fishes, invertebrates and animals that live on the reef.” (Ross Cumming, 03:00)
  • Elkhorn corals: Form thick, arm-like branches that build dense canopies.
  • Both suffered catastrophic declines from disease and storms since the 1980s; restoration programs have tried to revive them.

• 2023 Heatwave Catastrophe

  • The 2023 marine heatwave exceeded historical records 2.2–4 times over—“90, 100% mortality throughout most of the Florida Keys and the Dry Tortugas” (Ross Cumming, 05:39).
  • Only a few deeper-water survivors noted; but numbers and reproductive potential are so reduced as to be “functionally extinct.”
  • “If there's not enough of them around, those eggs and sperm won't be able to find each other to fertilize and create new baby corals.” (Ross Cumming, 08:23)

• Regional Implications

  • Similar lethal heat stress recorded across the wider Caribbean region.
  • Scientists are piecing together a regional snapshot, but expect widespread losses.

• Restoration Reconsidered

  • Traditional coral nurseries and planting are overwhelmed by relentless heatwaves.
  • New strategies include:
    • Breeding heat-tolerant corals using cross-regional parent stock, e.g., “the first crossbred corals with one parent from Florida and the other parent from Honduras...” (Ross Cumming, 11:34)
    • Manipulating the algae living inside corals to use more heat-tolerant species.

• Ethical Questions

  • “It’s a bit like farm animals or flowers in the garden, which might be very lovely to look at, but they're not the wild thing.” (Ronan Pease, 12:17)
  • Ross Cumming compares coral breeding for resilience to agricultural selective breeding, noting, “It’s the same type of thing if we were to attempt to breed corals for higher resilience.” (Ross Cumming, 12:41)

Notable Quotes

• “We saw 90, 100% mortality throughout most of the Florida Keys... almost complete loss of staghorn and elkhorn corals.” – Ross Cumming [05:39]
• “If there's not enough of them around, those eggs and sperm won't be able to find each other...” – Ross Cumming [08:23]

---

2. Chalky Unknowns: The Uncharted Fate of Carbon in the Ocean

[13:07–22:03]

The Unsung Carbon Cyclers

• Ronan Pease turns focus to microscopic ocean creatures—coccolithophores and planktonic mollusks—that help regulate Earth's carbon by locking up CO2 as carbonate in their shells.
• Patrizia Zaveri’s Science review paper highlights critical gaps in our understanding of these organisms’ roles and responses to current climate change.

Key Insights:

• Coccolithophores: Tiny, Mighty Carbon Sinks

  • Unicellular plankton that both calcify and photosynthesize: “Invisible, but very important for us and for the ocean.” (Patrizia Zaveri, 14:16)
  • Leave immense sedimentary records—e.g., the “Cliffs of Dover.”

• Planktonic Foraminifera and Pteropods:

  • Build carbonate skeletons; forams leave a continuous fossil record, pteropods are planktonic mollusks with fragile, “winged” shells.

• Big Questions, Big Uncertainties

  • Ocean absorbs ~30% of human CO2 emissions, but the precise future of this ‘carbonate pump’ is unknown.
  • Rising CO2 changes ocean acidity, which impacts these creatures’ ability to form shells effectively.
  • Increased CO2 may initially boost calcifying plankton, but rising acidity disrupts biomineralization, possibly leading to declines.

• Modeling the Unknown

  • “We need to have the missing information to better understand and reconstruct the future of the carbonate systems... we don’t really understand it fully for sure.” (Patrizia Zaveri, 19:54)
  • No historical analogy exists for the current rate of atmospheric CO2 increase.

Notable Quotes

• “They leave a really incredible sedimentary record since many, many millions of years, so they can be studied also in the Earth history. So it's a very unique group of organisms.” – Patrizia Zaveri [01:19, repeated at 14:16]
• “This is big experiment we are doing at the global scale is quite new... and we don't really understand it fully for sure.” – Patrizia Zaveri [19:54]
• “The carbonic chemistry... is really fundamental, not only for life in the ocean... but also for maintaining the atmospheric CO2 and the carbon cycle... in a balanced way for many, many millions of years.” – Patrizia Zaveri [21:27]

---

3. An mRNA Vaccine Bonus in Cancer Therapy

[23:16–32:13]

Surprising Medical Synergy

• Adam Gripen reports research showing that patients on immune checkpoint inhibitors for cancer who happened to receive a Covid mRNA vaccine lived nearly twice as long as those who didn’t.

Key Insights:

• Immunotherapies, including “immune checkpoint inhibitors,” only work if the patient has an active immune response against their tumor.
  • “If you don't have any flame, you can fan all day and you will never create a fire.” (Adam Gripen, 24:55)
• mRNA vaccines (including for Covid) act as a potent “siren” to activate the immune system—potentially “resetting” immune cells inside tumors to attack cancer.
• Initially found in “accidental” observations, confirmed in retrospective studies of over 1,000 patients and then in animal models.
• The effect was so robust that median survival for those receiving vaccines doubled.
• Mechanistically, both the mRNA and lipid nanoparticle components appear necessary.
• Clinical trials are planned to validate these results.

Notable Quotes

• “The COVID mRNA vaccine acts like a siren to activate all the components of the immune system throughout the body. And when this happens inside a tumor, those immune cells... start looking around for things to kill...” – Adam Gripen [25:50]
• “Their median overall survival was nearly double what we saw in patients who didn’t receive this vaccine.” – Adam Gripen [27:20]
• “These are very early days in the idea and there have been many false dawns in cancer research. Let’s hope this isn’t one of them.” – Ronan Pease [32:13]

---

4. Mapping Crops in Africa: High-Tech Meets GoPro Simplicity

[32:13–39:42]

From Satellite Images to Motorbike Surveys

• Catherine Nakalembe uses GoPro cameras mounted on vehicles to collect millions of images of fields in Kenya, building real-world training datasets for machine learning models to better map crops from space.

Key Insights:

• GoPros mounted on car roofs or motorcycle helmets capture 5 million images in Kenya alone, with data processed to identify individual crops (“We were aggressive and we're collecting an image every .05 seconds... we get multiple images per field.” (Catherine Nakalembe, 34:31))
• The images serve as “ground truth”—critical for training AI models to interpret satellite data accurately for crop types and health.
• Crop identification is crucial for food security, yield prediction, and monitoring crops in refugee settlements.
• The method is “low tech and simple”—replicable, cheap, and scalable: “You can just send out an extension agent and they can do it.” (Catherine Nakalembe, 37:42)
• The system has been deployed in Uganda, Tanzania, Zambia, Madagascar, Germany, France, and the US.
• Nakalembe shares the protocol widely: “I call it a kits program. You can fill out a form... and I'll send you a kit.” (38:58)

Notable Quotes

• “Because the ultimate goal is to make a map with satellite data... we need, if we give the model examples, a lot of examples of these different crops in these different contexts, then you can have a voila.” – Catherine Nakalembe [36:26]
• “It does not make sense to do it any other way rather than this way. It’s really fun at the same time.” – Catherine Nakalembe [39:40]

---

Memorable Moments & Quotes

• On Coral Loss: “If there's not enough of them around, those eggs and sperm won't be able to find each other to fertilize and create new baby corals.” – Ross Cumming [08:23]
• On Ocean Carbonate Cycle: “We need to have the missing information to better understand and reconstruct the future of the carbonate systems.” – Patrizia Zaveri [19:54]
• On Vaccines and Cancer: “When this happens inside a tumor, those immune cells... start looking around for things to kill, and they start programming other immune cells to kill the cancer.” – Adam Gripen [25:50]
• On Mapping Crops: “Low tech and simple... you can just send out an extension agent and they can do it.” – Catherine Nakalembe [37:42]

---

Timestamps for Key Segments

• [01:57] – Introduction & Florida coral crisis
• [03:00] – Coral morphology and habitat (Ross Cumming)
• [05:39] – 2023 marine heatwave devastation
• [10:35] – Restoration efforts and new approaches
• [13:07] – Ocean carbonate cycle, plankton’s role (Patrizia Zaveri)
• [17:38] – Ocean acidification impacts
• [19:54] – Limits of current knowledge/modeling
• [23:16] – mRNA Covid vaccines and cancer immunotherapy (Adam Gripen)
• [27:20] – Improved survival rates in patients
• [32:56] – GoPro cameras to map African crops (Catherine Nakalembe)
• [34:31] – Mass image collection and data use
• [37:42] – Simplicity and scalability of the method

---

Tone & Takeaway

The episode’s tone is thoughtful but urgent—often somber in the face of climate-driven losses but buoyed by stories of resilience and innovation, both technological (cancer vaccine synergy, crop AI) and natural (corals’ former glory; planet-wide plankton). The human spirit in science comes through, whether despairing over reefs or joyfully assembling a GoPro field kit. The overarching message: As the world changes rapidly, deep understanding and creativity are both essential tools for the future.

Transcript

[Ronan Pease]

This BBC podcast is supported by ads outside the uk.

[Sitecore Advertiser]

The customer journey isn't just changing, the journey is change. New ideas spread in an instant, expectations rise overnight, Decisions are made in the blink of an eye. That's why companies need sitecore. We put your brand in the moment right when your customer is ready to act. So every message feels personal, timely and makes your brand unforgettable. The journey is change. Sitecore moves with with it. See [e-mail] Journey in Turkey if you're willing to take a detour, you'll discover the food even social media hasn't got to yet. From Michelin stars and wine in Ola to traditional recipes and the home of baklava in the East, Discover the culinary capital of Gaziantep and talk to the locals. Every dish has its own story, flavors, experimentation and tradition. Turkia has it all. Plan your [e-mail] Foreign.

[Ronan Pease]

Welcome to Science in Action from the BBC World Service with me, Ronan Pease. Weave the threats of global warming and rising CO2 to marine life, much of it precious and invisible.

[Patrizia Zaveri]

They live a really incredible sedimentary record since many, many millions of years, so they can be studied also in the Earth history. So it's a very unique group of organisms.

[Ronan Pease]

There's also an unexpected cancer bonus from COVID vaccines, the MRNA type and how the simple GoPro camera is helping agricultural scientists map out crops in East Africa.

[Catherine Nakalembe]

I know it is vegetation, I know it is in this range, so it's a crop, not a tree. But I don't know what it is. So we need a lot of examples.

[Ronan Pease]

Last week, the 2025 global tipping points Report, cataloguing points of no return in the growing climate crisis, declared we are already crossing the first of these tipping points, the loss of warm water coral reefs around the world over the next decade. Their deaths are 99% probable, they say, if we pass 1.5 degrees of warming, which does seem inevitable. This week in the journal Science, we have an example of the danger, the functional extinction of two species of in the hot summer of 2023 from the Florida Reef, which stretches 560 kilometres from offshore Miami on the Atlantic coast down and round to the Southern Keys, the highland chain that curves beneath the southern tip of the state. As ever, the corals are carbonate colonies built and inhabited by countless tiny animal polyps. And these two, the staghorn and elkhorn corals, deserve their name. According to marine biologist Ross Cumming.

[Ross Cumming]

The staghorn coral forms colonies with these long pointy branches and they're Actually designed to fragment and break and continue growing since they are colonial animals. And as they do this, they can form these large thickets of these spiky branches that produce this highly complex three dimensional structure that provides habitat for many other fishes, invertebrates and animals that live on the reef. You know, if you're swimming over one, they can stretch for, you know, tens of meters or more sometimes. Or they can also form individual isolated colonies, you know, the size of, you know, a basketball or, or, you know, a small bush. So. So it's very variable in size depending where they're growing.

[Ronan Pease]

And they're just sitting there a short distance under the ocean surface, soaking up the sun and capturing small things that float past.

[Ross Cumming]

That's right, yeah. They absorb light and use photosynthesis of the, in their symbiotic algae that provides them most of their nutrition they need to grow and build the reef. But they do also capture plankton and particles floating by in the water column with their tentacles.

[Ronan Pease]

And then the other ones you're talking about in this paper, the elkhorn, I think they're similar, but obviously different.

[Ross Cumming]

Yeah, the elkhorn also branch, but they form these large, thick, sort of arm like structures that will grow, you know, up towards the light, towards the surface, and they will crisscross each other and form this dense sort of tree like canopy in places where they form these large stands, in some places you will see lots of them, in others, you might not, even prior to 2023, see many at all. And that's because they've historically been in decline since the early 1980s due to disease, hurricanes, poor water quality and previous bleaching events had already reduced their populations by 90% or more. But then in recent years, we've also seen large restoration efforts for these species. And so in some locations which have been targets of restoration activity, thousands, even hundreds of thousands of these colonies have been grown in nurseries and then planted back out onto the reef to boost the amount of coral that's present. So depending on exactly where you'd look, you might see a lot of these species in some locations.

[Ronan Pease]

So they were critically endangered. There's been this conservation effort to try and preserve them. But from what you're saying in this paper, all that effort's been sort of wiped out effectively by the summer of 2023.

[Ross Cumming]

Yeah, unfortunately that's, that's exactly what happened. So in 2023 we had this heat wave that was so severe and shattering records, it was 2.2 to 4 times more heat stress than we'd ever seen before in Florida. And this heat stress surpassed the limits of essentially all of the Acropara to survive. So we saw 90, 100% mortality throughout most of the Florida Keys and the Dry Tortugas, which is the bulk of Florida's coral reef. We did see pockets of higher survival on the northern end of the reef tract where the heat stress was less severe. So it was about 38% mortality in the far northern reaches of the reef tract. But most of Florida's coral reef, it was almost complete loss of staghorn and elkhorn corals.

[Ronan Pease]

And it really is just that they have this quite stable range of temperatures which they can survive in. Is that right?

[Ross Cumming]

Yeah, that's right. They're very sensitive to deviations from the temperatures they're used to. And when the water gets warmer than this historical maxima that they're used to seeing, they undergo this stress response called coral bleaching, where they expel their somatic algae, they turn white, and then they're often susceptible to mortality following that. But in 2023, the heat wave was so severe that we actually saw this sort of almost rapid mortality of these corals in. Within days before they almost.

[Ronan Pease]

So you were sort of going out there and seeing corals a few days ago were sort of struggling. But you can tell while you're out there, can you, this kind of stress they're under?

[Ross Cumming]

Yeah, there were observations of, you know, tissue sloughing being observed directly on these corals, and as this heat wave sort of took off, and there were many, many divers observers in the water observing these corals, collecting this information on mortality and documenting these losses.

[Ronan Pease]

This was 2023, I presume you've been back in 2024 and this year. I mean, what's the current state?

[Ross Cumming]

Yeah, we've been back and, you know, nothing's really changed with the Acropara. We haven't seen any coming back. One glimmer of hope is we have found a few more survivors out there. There's one deeper location where we looked and found survivors. Maybe they were buffered from the heat stress in the deeper waters. But, yeah, largely the. The story hasn't changed.

[Ronan Pease]

And that's why you say functionally extinct, even if there are some survivors, because they have this spawning method, I think they just send their eggs and sperm into the water to mix. And if there's not enough.

[Ross Cumming]

That's right, yeah. If there's not enough of them around, those eggs and sperm won't be able to find each other to fertilize and create new baby corals. And we've documented reproductive failure of these species in Florida actually for quite some time due to the historical decline in their numbers. So now, following this catastrophic loss in 2023, there's even less hope that they can actually recover on their own.

[Ronan Pease]

I feel this is a story not necessarily with the Florida Keys that we've covered over the years on the program, but certainly in the Caribbean, the stresses and the bleaching events that corals are under. How indicative, should I say, is this particular episode that you're talking about in terms of coral health around the region?

[Ross Cumming]

Yeah, so we don't totally know that yet. And scientists are working now to sort of combine and integrate data from across the Caribbean to produce a regional picture of what's happened to these species over the last couple of years. But we do know that the heat stress that we observed in Florida that led to these levels of mortality, similar levels of heat stress were also observed throughout most of the Caribbean in 2023 and in 2024. So, you know, it's likely that these species have been heavily impacted throughout the Caribbean, although we've, you know, anecdotally heard about some places where, you know, mortality was very high and other locations where mortality was lower. So there may be, you know, pockets of survival out there. And we're working now to sort of produce that whole picture of what's happened.

[Ronan Pease]

I mean, I presume the nurseries that you were talking about where you're actually deliberately growing new coral and then you can try and re precedes the reefs. That effort I may be continuing, but it feels to me like it's going to be pretty hopeless because with global warming, you're just going to get more events like you got in 2023 and you're just going to be seeing basically more death.

[Ross Cumming]

That's right. So I think we're at a point in restoration where we know we need to change the approach. And that means finding ways to integrate interventions that can boost coral heat tolerance into restoration. And there are scientists and restoration practitioners working on many of these. Now, some methods include trying to breed more resilient strains of coral by using heat tolerant parent corals, or even sourcing new parent colonies from other populations outside of Florida, for example, that might already be adapted to higher temperatures elsewhere in the region. And by cross breeding corals from Florida with more resilient corals from elsewhere, we might be able to produce corals that have a better chance of surviving the next heat wave in Florida. So that's one intervention approach that is being tested, actually Right now, they're the first crossbred corals with one parent from Florida and the other parent from Honduras. In an area where elkhorn coral is naturally thriving in warmer waters. Corals were crossbred and planted out onto reefs off of Miami earlier this year. So that's being piloted. Another intervention approach could be manipulating the symbiotic algae in corals. So there's different species of these single celled algal symbionts that live inside the corals. And certain symbiont species are more thermally tolerant. And by manipulating those species, we might be able to increase the heat tolerance thresholds by 1 or 2 degrees. That could give corals the leg up they need to survive the next heat wave in Florida.

[Ronan Pease]

I mean, it's a very interesting question in the ethics, shall we say, of conservation, if in a sense there's this human intervention to make artificial species or artificial variants. It's a bit like farm animals or flowers in the garden, which might be very lovely to look at, but they're not the wild thing. I presume you have these discussions.

[Ross Cumming]

Yes, of course. And you're right. We've already do these types of things and we've been doing these types of things for a very long time with farm animals and agricultural crops. And, you know, we, we breed them for traits that we deem desirable or that will be beneficial. And so it's the same type of thing if we were to attempt to breed corals for higher resilience.

[Ronan Pease]

Ross, coming off the John G. Shedd Aquarium. And I guess we'll be hearing more reports like that as the oceans keep warming, the carbonate those reefs build themselves from ultimately comes from the carbon dioxide dissolved in the ocean waters, which itself comes from CO2 in the atmosphere. They are, in effect, mineral banks of carbon drawn down from the rest of the biosphere, though not a big one, a few tens of millions of tons per year. I've read the big players in this game are tiny organisms in the vast ocean expanses, plankton and miniature mollusks, which in their huge numbers cumulatively turn billions of tons of CO2 into carbonate. But the exact fate of those billions of tons is uncertain, which matters when we're adding billions of tons more to the atmosphere and that ends up acidifying the ocean water. Patrizia Zaveri and colleagues have laid out the state of our ignorance in another paper in Science and told me about these unappreciated critters. Firstly, the coccolithophores.

[Patrizia Zaveri]

Yeah, this is a very fascinating marine group of Calcifying plankton, how we describe it in the review. So this group of organisms are unicellular. They're also quite unique because compared to the other two group, they also perform photosynthesis. So not only they calcify, but they also perform photosynthesis. But very, very important, so invisible, but very important for us and for the ocean. And the other also very unique characteristic for us is that they leave a really incredible sedimentary record, geological record since many, many millions of years since the Mesozoic. So hundreds million years. So they can be studied also in the earth history, in the geological record. So it's a very unique group of organisms.

[Ronan Pease]

I live in the shadow of a huge range of hills which are made up of, I think, Cretaceous coccolithophores, caught these long chalk downs which were laid down tens of millions of years ago. So, yeah, I get a sense of that.

[Patrizia Zaveri]

Yeah, this is like the Cliff of Dover. They are a very, very visible example of what this type of organism can store in terms of carbon, for example. So like the example of the Cliff of Dover.

[Adam Gripen]

Yeah.

[Ronan Pease]

And so those are plants, effectively, that lay, that have this sort of carbonate structures inside them. The other two forms that you're talking about, these are animal forms that lay down some kind of carbon shell.

[Patrizia Zaveri]

So these are the two major components, even of course, they are minor compared to Coccolithophora are planktonic, foraminife and sharp theropods. They all build a skeleton and they are both zooplankton. They are quite different in the way they look and the way also in terms of their fate of calcium carbonate from the production to the sediment. For example, we know foraminifera, the same as coccolithophorae. They have a very, very beautiful and long and continuous geological record. We also see how efficient they are in terms of being exported from the production layer and then sedimented. And the other groups that study very much in relation to ocean acidification, they're also very unique because they are mollusks. They don't look like organism that they can be planktonic. They look more bentix. They look like they could be bentic. So attached to a surface, but evolutionary, they develop this kind of wings and you like in the. In the planktonic environment.

[Ronan Pease]

So in the open, they're more mobile than the others. The others are sort of floating. And I guess this is why what you're reviewing here is so important, because looking at the numbers, I mean, these are incredibly tiny creatures, but there are so many of them. If I'm right, they are Absorbing from the ocean every year, billions of tons of carbon. And therefore they can have quite an influence on the fate of the carbon dioxide that we're putting into the atmosphere. So that's the question is how much help they can be and whether we're going to damage the kinds of things that they are maybe doing at the moment.

[Patrizia Zaveri]

Yeah, understanding really the response or the vulnerability of these different groups of organisms to climate change, ocean acidification, all the different changes that are happening now very rapidly in the ocean, is critical for understanding really what will happen to carbon sequestration. Because we know that about 30% or so of carbon that we are emitting is actually trapped and exported and sequestered by the ocean. So we really need to understand the fate of this organism to better constrain also the carbon sequestrations.

[Ronan Pease]

I, I mean, the most simplistic version I would imagine, is if they need this carbon dioxide to make their shell parts, and the, the coccolithophores, which are plants, they need the carbon dioxide for photosynthesis and so on. They ought to be loving what we're doing to, to the atmosphere because this is giving them more carbon dioxide. But I guess that's not quite true.

[Patrizia Zaveri]

Not exactly. Not exactly. So we know that astray coholutophores are quite sensitive to change to high, actually to projection of high CO2, because basically they are taking this element for calcification. And we know, for example, with acidification, so with increased hydrogen in the upper seawater, is actually altering the process of biomineralization on the cells. And so the organism need to adjust quite quickly to these rapid changes, and he has trouble to adjust to these high CO2 concentrations. So in the first place, you see the increase in production because they are enhanced production to the high CO2. But then we also see some effect of high CO2 in terms of, for example, in the physiology, in the biomineralizations.

[Ronan Pease]

And in terms of helping us out of the mess we're making of the atmosphere. The important thing would be that they're all sinking to the bottom of the ocean and making the next generation of carbonate reefs and chalk hills and so on. My impression from your paper is that you're saying we simply do not know the capacity of the system to do that.

[Patrizia Zaveri]

Yeah, we need to have the missing information to better understand and reconstruct the future of the carbonate systems. But also, of course, the carbonate system influences the carbon sequestrations in the atmosphere. So we understand a lot, but not enough, because this is big experiment that we are doing at the global scale is quite new. It's new, in fact, and we don't really understand it fully for sure.

[Ronan Pease]

I mean, a way it's shocking because the unspoken part of the global warming story is we know all about the carbon dioxide we put into the atmosphere. We don't talk very much, certainly on programs like mine, about the fate of a lot of it. The service, effectively, that things like the ocean do in removing a large part of it and what's going to happen in the next hundred, 200 years is incredibly dependent on those details.

[Patrizia Zaveri]

Absolutely, yeah, absolutely. I mean, I think there are very good models now. We know that what we are doing is, is a very, very anomalous in Earth histories and we don't have any analogous time period that we can use, as you know, to learn more. I mean, we have some time in the past where there were high CO2, but not at the rate, it's not comparable the rate that we are pumping the atmosphere or putting CO2 in the atmosphere. So yes, it's very hard to really anticipate. But one thing that we know, what I said before, that these tiny organisms, they had a major, major role in regulating the carbonic chemistry of the ocean. And the carbonic chemistry, okay, it's a concept maybe not so easy to understand, but the carbonic chemistry of the ocean is really fundamental, not only for life in the ocean, for developing species, but also for maintaining the atmospheric CO2 and the carbon, the carbon cycle in a way, in a balanced way for many, many millions of years.

[Ronan Pease]

Patrizia Zaveri of the Autonomous University of Barcelona, talking to Science and Action from the BBC.

[Sitecore Advertiser]

The customer journey isn't just changing, the journey is change. New ideas spread in an instant. Expectations rise overnight. Decisions are made in the blink of an eye. That's why companies need sitecore. We put your brand in the moment right when your customer is ready to act. So every message feels personal, timely and makes your brand unforgettable. The journey is change. Sitecore moves with it. See [e-mail] journey Most travelers stick to the highlights. But in Turkey, if you go off the beaten path, you'll experience a whole lot of adventure. Like the St. Paul Trail where cyclists chase the thrill, or Urla, where you can kite surf to your heart's content for exploring on your feet. Hike up to Mount Nemrud or walk in Patara through ancient civilizations that might reveal the history of humanity itself. Plan your [e-mail] for a gripping adventure.

[Ronan Pease]

After half a programme of gloom, let's Cheer ourselves up. First, a bizarre story about the MRNA Covid vaccines boosting some cancer treatments, which has got a lot of attention. I confess, there's so much nonsense spread about the MRNA vaccines. I was wary of this one, which came out of a cancer conference in Germany, but people I trust seem to rate it, and the details have also just appeared in Nature. So we asked presenter Adam Gripen from the MD Anderson Cancer center to explain how a vaccine against a virus might bolster cancer therapy because it's not a direct connection. First, the class of treatments he's talking about are those that encourage our own immune system to attack a tumor. Immunotherapies.

[Adam Gripen]

The idea with cancer immunotherapy is we want to unleash the power of the immune system to kill cancer cells. We have a few ways of doing this, but the most common way is called immune checkpoint inhibitors, which we think of as really taking the brakes off the immune system and allowing it to kill cancer effectively.

[Ronan Pease]

But if I've understood this right, there are some cancers which you think ought to respond to this kind of therapy, but somehow still there's something holding the immune system back.

[Adam Gripen]

Yes. So for immune checkpoint inhibitors to work, the patients need a pre existing immune response against their tumor. If they don't have that immune response, the immune therapy is not very effective. I think of it as if the immune therapy fans a flame of an existing immune response. And so if you don't have any flame, you can fan all day and you will never create a fire.

[Ronan Pease]

And so the immune response, I'm thinking of this in terms of things like T cells or antibodies, but these are things which recognize something about the tumor that says, hang on, this is alien, this is dangerous, I'm going to go in and kill it. Is that the bit that's sort of missing?

[Adam Gripen]

That's exactly right. And what we think about is really the T cells being the primary driver of these. These are the killer cells of your immune system. When they're trained appropriately, they can identify and kill tumor cells throughout the body.

[Ronan Pease]

Okay, so then we get to the really puzzling part, which is that you think that somehow MRNA vaccines somehow trigger that now, but it's not the sort of the message that's encoded in the MRNA or something, but there's something about them that says, hey, go and do something to the immune system. So this is on top of the checkpoint bit that you were talking about earlier.

[Adam Gripen]

Exactly. So what we found is that the COVID MRNA vaccine acts like a siren to activate all the components of the immune system throughout the body. And when this happens inside a tumor, those immune cells that are residing in the tumor start looking around for things to kill, and they start programming other immune cells to kill the cancer.

[Ronan Pease]

And you saw this first of all, as I understand it, in experiments with mice. So this was a very deliberate attempt to see how that would work, is that right?

[Adam Gripen]

Actually, this all began in 2016 with a study in which we were trying to show how important it was to personalize an MRNA cancer vaccine that we currently have in phase one clinical trials. And we ran a control where we didn't personalize the vaccine. We used an RNA that didn't have anything to do with the tumor at all. And what we were surprised to find was that that RNA unleashed the immune system against the tumor, even when it wasn't training the immune system deliberately to attack tumor cells when the pandemic hit. We were able to ask the question in patients of whether an MRNA vaccine targeting a non tumor antigen, here, Covid, would boost responses to immune checkpoint blockade. And so we looked back at over 1,000 patients treated with immune checkpoint inhibitors. And we found that those patients who happened to receive a Covid MRNA vaccine around the time they started immune therapy lived significantly longer than patients who didn't receive the vaccine. And that really proved the principle that this could be relevant in patients. And then we turned to our animal models to really understand how this works.

[Ronan Pease]

What kind of benefit were you seeing for the patients?

[Adam Gripen]

So we were excited to see a very significant improvement in overall survival in these patients. When we looked at patients with non small cell lung cancer and melanoma who happened to receive this vaccine, in our window of interest, their median overall survival was nearly double what we saw in patients who didn't receive this vaccine, suggesting that there could be a strong association between vaccination and immune responsiveness.

[Ronan Pease]

I take it that you don't have any data at all, apart from survival from the patients. So this was basically comparing health records, I take it you're doing, but you were able to go into mice and then start to see, as it were, the mechanisms that are going on. Because I have to say, when I first saw this, I was thinking, you know, this is more nonsense about the MRNA vaccines, but you're getting some kind of sense of it, are you?

[Adam Gripen]

Yes. So we were able to use mouse models to really understand exactly how this works. What we found is that when you administer the COVID MRNA vaccine in mice, it triggers this broad activation of of the immune system throughout the entire body, including inside the tumor, where we See, profound activation of the immune cells that are residing in the tumor. In the normal situation with a tumor, the cancers often convince those immune cells not to kill the cancer. But what we saw with the vaccine was we essentially performed an immune reset to have the immune system start looking for things to kill. And when that happens, it allows the immune system to start killing the cancer cells.

[Ronan Pease]

And as far as you know, this is because of the mRNA, this sort of bit of genetic material that's in the vaccines and nothing else about them, like the sort of oil droplets that they're held in.

[Adam Gripen]

So we know that the RNA and the lipid together create a structure that is a very powerful immune activator. We think that each of those two components are important, and we're investigating ways to further improve this.

[Ronan Pease]

I mean, the great thing about the COVID vaccine is it was very widely distributed and it's now relatively cheaply manufactured. But presumably you'd like to come up with some kind of formulation that does the job better and maybe more simply.

[Adam Gripen]

Exactly. We use the COVID MRNA vaccine because it is widely available, and we're currently planning a phase 3 clinical trial to test the COVID MRNA vaccine in patients as an immune stimulant. But there's no reason to think that the COVID vaccine is the best possible vaccine at producing this type of a response. And so we're also spending a lot of effort trying to develop a next generation therapeutic that would do this even more effectively.

[Ronan Pease]

Well, good luck with that. I mean, but there is the potential, I presume, you know, the study you've done so far is purely what they call observational. There is a danger that actually this just goes away. It's just a sort of one of those statistical chance things, or. I don't know. Are you confident? Shall I say?

[Adam Gripen]

We were very optimistic about this effect because we saw a very strong signal in our retrospective data. And then we were able to tease out the mechanism and show this effectively in animal models. And then even beyond that, the mechanisms that we found to be at play in our animal models, we were also able to show those same things happening in humans. So in healthy humans who get the MRNA vaccine, they respond to it exactly like the mice responded to the vaccine. And even our patients have some of the biological effects that suggest that their immune system is becoming activated against their tumor. With that said, you know, despite the strong association, we need to test any of these hypotheses in clinical trials before we apply any of our conclusions in the clinic. And so we think it is imperative to test this in a randomized phase three trial as soon as possible.

[Ronan Pease]

One of the odd things about this is that before the pandemic came along, as far as I know, most of the effort like the experiments you were talking about in 2016 with these MRNA vaccines were in cancer. I guess because it's an incredibly intractable problem. It does seem to be slightly odd in a way that because of the pandemic it sort of turned around on itself in a way that this is another way in which the MRNA vaccine may in themselves be really beneficial approach to doing immunotherapy for cancers and maybe for other things.

[Adam Gripen]

Absolutely. I think this study really supports the power and the potential of MRNA technology and specifically MRNA vaccines. These are relatively new therapies and there's still a lot more to learn. But what we know is that they seem to be full immune modulators. And you know, my work focuses on cancer, so I'll focus in the cancer space to say that, you know, they, they have a high potential to change the way we treat cancer in the future and I'm excited to continue work in this area moving forward.

[Ronan Pease]

Adam Gripper as ever, a phrase that's been repeated countless times on science and action over the decades. These are very early days in the idea and there have been many false dawns in cancer research. Let's hope this isn't one of them. To close a return to the program of Catherine Nakalembe, born in Uganda, but now at the University of Maryland where she analyzes satellite earth observation data to understand the health and viability of crops around the world to bolster food security. Except this time she's put high tech space imagery aside and has used instead everyday GoPro cameras put on car tops and motorcyclists helmets which were then driven down the roads of Kenya.

[Catherine Nakalembe]

So you can do it in a car. I should say this because I do not ride a motorcycle. So I put it in the car and as I drive I just take pictures. However you have a motorcycle, as you can imagine, like a border, border taxi somewhere in East Africa. They always have a helmet. And what we have is a basic GoPro camera. We bought the cheapest model which has GPS. Very important attribute, we need to know where you are. And we installed the camera instead of facing the typical way you would do it for. If you're trying to do jumps and stuff, it faces the other way, it's facing sideways, it's facing sideways. So you changed how you place the little thing on the top to face the other way. And we're always trying to collect images on the passenger side. That's the side that's closest to the field. And as they drive or take images, obviously we get a lot of useless, random things. You know, it tilts faces the sky. And ultimately, within maybe a year, overall, we ended up having about 5 million images.

[Ronan Pease]

5 million. So getting on their bikes, their motorbikes, they're just driving down the roads of Kenya, is that right?

[Catherine Nakalembe]

So this particular paper that we just published only includes the Kenya data, which we've been able to kind of sift through. And what's kind of interesting or really insane is the first time we tested this out in Kenya is we had two teams. And this time there was not motorcycles, was cars. And they covered all of western Kenya in about two weeks. The first iteration, like the first test of it.

[Ronan Pease]

And how much detail are you getting? I mean, you're getting one picture per field or. I don't know, because when I said.

[Catherine Nakalembe]

That we were aggressive and we're collecting an image every.05 seconds means that you get, you can get. If a field is small, you can get multiple images per field. And in the end, when you want to train a model, it's not that useful. For every location, you don't need to have multiple images. So some of our processing skip a few images, so we don't have a lot of repetition. And another example in Uganda, where we haven't processed most of the data, we have over 2.9 million images. And it was, I think, three to four people that did all of this.

[Ronan Pease]

I mean, you're getting enough detail that you can recognize, I don't know, this is sorghum, this is maize, this is, I don't know, some kind of fruit tree or something.

[Catherine Nakalembe]

Absolutely, because the cameras are action cameras. They're meant to take clear photos at very rapid speed. That's a very important thing. Right. So you can drive normal speed. You do not have to slow down. And so the images are not blurry. So you can clearly see this is a maize field, this is wheat, this is soy.

[Ronan Pease]

And what happens if there are crops planted together? Which I think is one of the things you get.

[Catherine Nakalembe]

So this is absolutely a really good question. So in East Africa, in southern Africa, there's a lot of intercropping. So you have multiple crops in a field. So it's not, you know, homogeneous wheat over, you know, long, extended periods. And when the model does the initial prediction, it tries to predict across all of these crops and gives us an indication whether that these crops exist in this One photograph. Because the ultimate goal is to make a map with satellite data. Each pixel ends up having one value. And so we can either say it's a mixed field or it's a maze field. Oh, it's a maize dominated field.

[Ronan Pease]

So the idea of this project is you want to use ultimately satellite data, which is what I've spoken to you about past, but you need some kind of ground truth so that the as, as it were, the texture of those satellite data will correspond to the real life crops.

[Patrizia Zaveri]

That the reality.

[Ronan Pease]

Yeah.

[Catherine Nakalembe]

Yes. Because from crop type mapping perspective, I know it is vegetation, I know it is in this range. So it's a crop, not a tree. But I don't know what it is. So we need, if we give the model examples, a lot of examples of these different crops in these different contexts, then you can have a voila.

[Ronan Pease]

I think you see reasons I love this story is that there's this whole thing about the high tech of using amazing cameras on satellites and there's the machine learning and the artificial intelligence that's that's involved and then suddenly it's all made sort of much more real by something as low tech as a GoPro camera. Because that kind of technology has changed so much in just 10 or 15 years.

[Catherine Nakalembe]

Low tech and simple also in the sense that once you explain it, you can see someone's eye go like, oh, I explained it to a director in Senegalese Ministry of Agriculture is the director of statistics. And he was like, how much does it cost if you're trying to do an agricultural survey, you have to cover real edge areas in a really short time so you can capture and understand what is happening. And there are other use cases that can be built on top of it. Like in Kenya, a lady who works in the ministry in the statistics department was like, oh, we could do this multiple times. That means we'll have photographs of what's actually happening. Because the photographs are very clear. You can see MESA is not doing great. You can see, you can distinguish the different things that the fact that you can go really quickly, you don't need to be able to buy very high resolution satellite data or fly a drone. You can just send out an extension agent and they can do it.

[Ronan Pease]

Is it the idea, the plan, the paper is about doing this in Kenya. Is this something you plan to roll out in huge parts of the world? Or maybe you're already doing it a lot of places?

[Catherine Nakalembe]

Yeah. So the reason why we produced the first data set as the dataset from Kenya1 is it builds into work that we were doing around not only crop type mapping, but yield prediction with the Ministry of Agriculture. So we're trying to do this at the field scale. But the initial campaigns were done in Uganda, in Tanzania, in Zambia and Kenya. But over time we've deployed the same approach to collect data for a project looking at cropland change in the space context. We're looking at refugee settlements, looking at where crops are growing in refugee settlements. I've done it in Madagascar looking at cacao mapping. So if I want to create a map for where cacao is, to understand cacao conditions, I need examples of where cacao is versus where it's just forest that is in Senegal. In support of this project looking at cropland mapping with the ministry there. And when I say we actually all I do is assemble a box, which I have one right here, send it to the team, explain them the concept. I call it a kits program. You can fill out a form and tell me why you want to do it and that you're willing to contribute to our database and I'll send you a kit. So it's scalable, replicable. We've done it in the us, we've done it in Germany, I've done it in France. I'd say, like it's right now, when I think about doing crop type data collection, it does not make sense to do it any other way rather than this way. It's really fun at the same time.

[Ronan Pease]

Which always goes down well with me. Crop scientist Catherine Nakalembe joyfully doing her bit to improve the world and putting a smile on my face that may fade next week when I present the last edition of Science in Action from the BBC World Service with a focus on on the state of science and the wider understanding of science a quarter of the way into the 21st century. I hope you'll join me. Run Pease and producer Alex Mansfield for that.

[Sitecore Advertiser]

The customer journey isn't just changing. The journey is change. New ideas spread in an instant. Expectations rise overnight. Decisions are made in the blink of an eye. That's why companies need sitecore. We put your brand in the moment right when your customer is ready to act. So every message feels personal, timely and makes your brand unforgettable. The Journey is change. Sitecore moves with it. See [e-mail] Journey this is the story of the One. As a custodial supervisor at a high school, he knows that during cold and flu season, germs spread fast. It's why he partners with Grainger to stay fully stocked on the products and supplies he needs, from tissues to disinfectants to floor scrubbers, all so that he can help students, staff and teachers stay healthy and focused. Call 1-800-GRAINGER Click grainger.com or just stop by Grainger for the ones who get it done.