The ‘ghost roads’ driving tropical deforestation

Benjamin Thompson

Welcome back to the Nature Podcast, this week: the impact of ‘ghost roads’ in tropical forests…

Nick Petrić Howe

…and the proteins that may determine left-handedness. I’m Nick Petrić Howe.

Benjamin Thompson

And I'm Benjamin Thompson.

<music>

First up on the show today, we’ve got a story about roads. Now, roads are fundamental to modern society: the networks they create around the globe are vital for economies and livelihoods, helping people and goods get where they need to go.

But of course roads can also come with significant costs to the environment, particularly in tropical regions. Clearing trees to build these roads can create extra emissions, and damage biodiversity-rich ecosystems. And new roads can open up areas where natural resources can be exploited, sometimes illegally.

And the thing is, a lot of the roads that are created to allow these activities aren’t necessarily on the books. It’s thought that around the world there are a huge number of roads like these that don’t appear in official government maps.

This has led researchers to ask how many of these roads are there? And what is their impact? This week in Nature, a group of researchers have been looking into these hidden highways in the Asia-Pacific region, which they call ‘ghost roads’. I spoke to one of the team, Bill Laurance from James Cook University in Australia, to find out more about them. And asked him what a ghost road actually looks like.

Bill Laurance

So a ghost road is our name for a road that's usually illegal, or at least illicit or informal. They can be like a logging track in a forest, they can just be a road that's been illegally bulldozed in some cases that can be paved, but not normally. So they're really diverse, I mean, the one thing is they're not a big official highway that's got highway signs, and all that type of thing. So the legal roads typically would be found on a government map of some sort. And in particular, on a couple of the international road databases, which are supposed to be pulling all this information together for different countries. And one of these is called GRIP, the Global Roads Inventory Project, and the other one's called OSM, which is Open Street Map.

Benjamin Thompson

Do you have any data on who tends to build these roads and what they're used for?

Bill Laurance

Look, there's a whole wide range of actors that would be involved in this. I mean, it could be anybody involved in exploiting natural resources, and the timber, gold and bulkminerals, all that type of thing. And then there's companies that that operate illegally. I mean, the truth is, there's a lot of illegal activity. It's a real mixed bag, but it adds up to a lot when you put it all together.

Benjamin Thompson

So we have these ghost roads then, and you were looking to see in parts of the Asia-Pacific region, namely Borneo, Sumatra, and New Guinea, how many of these roads existed that weren't on the official database or in the datasets. And it turns out that this isn't necessarily a straightforward thing to do?

Bill Laurance

Yeah well, it was a lot of work. We had 210 volunteers and a few scientists, we put in total, about 7,000 hours doing this work. The actual mapping of the road comes from using Google Earth imagery and we then literally just hand map all of the different roads. And that might sound a little bit basic or something but it turns out that the human eye and the human brain is really, really good at mapping roads, and filling in little missing bits and that type of thing. And we thought, look, there's got to be a better way to do this. And so we started looking at using artificial intelligence with the satellite images to try to map the roads and the best category models that we had had about 81% accuracy. And that's pretty good. But for the work that we want to do, the real research work, we need better accuracy than that, we're over 90% with our human observers. But still, we need to be able to right now to map all these roads all over the world, right? We put in 7,000 hours, and we managed to do three islands in Borneo, New Guinea and Sumatra, or most of them. But to do the whole world, all of Earth's land surface, we reckon we'd need 640,000 hours to do it manually, which is nuts, you know, that will never happen.

Benjamin Thompson

You've had this Herculean mapping exercise with your volunteers and your researchers drawing these lines on Google Earth, essentially. What did you find in terms of how the data that you had painted on compared to these two datasets that are generally used to measure how many roads are in a place I guess?

Bill Laurance

Oh, it was just off the charts. We've found between 3 times more roads and 6.6 times more roads that occur on any existing maps, in other words a whacking lot more roads than you would expect based on the roadmaps. We have one area, specifically in Borneo, and we showed this figure in the paper, because you see the Global Roads Inventory Project estimate and there's just like a couple little lines. And then there's the Open Street Map one, and there's a couple more lines. And then here's our version and there's just like hundreds and hundreds and hundreds of lines, which are all roads. So all that was being missed.

Benjamin Thompson

Well, I will say that I looked at that example, in your paper, and it is quite stark, right? The images are very different. I mean, the data you've put forward almost looks like the street layout in Manhattan, for instance, like there's this grid layout, in some cases — was that a particularly extreme example? Or does that speak to what you saw more broadly in these different places?

Bill Laurance

Yeah, no, that's not really an extreme example, that's an intensively affected landscape, you know, it's a human dominated landscape, and those tight grids that you would have been seeing are probably associated with oil palm plantations. But no, that wasn't anything particularly unusual. It's just sort of pulled it all together. And we're a little stunned by what it looked like.

Benjamin Thompson

And you’ve got your data that you put forward on where you say these roads exist, what sort of landscape were they cutting through in the different areas?

Bill Laurance

It varies. I mean, in some cases, you know, regrettably, they're cutting right into virgin forest that's completely intact. But in some other cases, you know, there may be landscapes that have got some level of development that can be oil palm, it can be wood-pulp plantations, or can be slash and burn farming.

Benjamin Thompson

Of course, deforestation is a big issue in terms of things like habitat loss, biodiversity loss, and increasing emissions. Now, you looked at the effects of these ghost roads on deforestation on the different islands. What did you find there?

Bill Laurance

Well, the deforestation is really associated with the roads. I mean, the basic bottom line is, where the roads go is where the forest destruction happens. And they're just hand in hand. So for instance, we came up with these statistical models, we put in 38 different variables. So you know, how close are you to a city? What's the topography like? How much rainfall do you have? Rainfall makes it hard to burn forests. And when we put all that into this statistical model, the roads, you know, jumped out as the hugely most significant variable of all, by far, overwhelmingly in this forest destruction process that we're looking at. Because some people say this, like, oh, well, you know, roads aren't really causing deforestation, you'll get deforestation happening, let's say agriculture. And then they put the roads in afterwards. And we said, nah we don't think that's right. We think the roads go in first. And so it's really critical to know where they're going in and why and, you know, that type of thing. So we looked at this, and turns out that, you know, almost like 93% of the cases, the road goes in first, you know, and then in like, 2% of the cases, yeah, there's some deforestation and then the road goes in secondarily. But it's overwhelmingly strong.

Benjamin Thompson

And what sort of impact does this network of ghost roads have on the local environment and biodiversity and what have you?

Bill Laurance

You know, we did an analysis of the so-called human footprint, which is this map designed to sort of use different colours to show how degraded or affected different parts of the world are. So we did one map of a part of Borneo using the old road-databases, you know, you got some orange, you got some red, which of the danger colours. Then we did the same exact analysis, same exact area with all the ghost roads in there and kaboom, then you don't just get a little bit of red and a little bit of orange, the whole map sort of jumps out at you. So what that's telling you is that area, that region is a lot more degraded than one would think. And also, there are pockets that are still relatively intact, but you've need to focus in on those right away because they're extremely vulnerable.

Benjamin Thompson

And what is knowing that these roads exist mean then? How can this data actually be used, andwhat can it be used for?

Bill Laurance

Yeah, well, it can be used for law enforcement for one thing, I mean, you would really like to see a lot of these roads either closed or certainly enforcement activities so that there's not additional roads proliferating all over the place. Let's say you put in a major highway or you know, paved road somewhere, what you commonly get is a whole series of illegal roads coming off of that it's just a common pattern that you see. We have a plant here in Australia called hairy mary, it's a vine and coming off of the vine are all these sort of lateral big spikes, like little roads coming off. So we call this the ‘hairy mary effect’. It's associated with land speculation and illegal logging and a lot of the other kinds of activities that we talked about. So one of the messages we want to say to governments is be really careful where you develop your infrastructure – it's absolutely critical. And a lot of our arguments are about ways of actually finding techniques to improve the economic efficiency and lower the construction costs of roads, that type of thing. So we do try to approach this in a pretty practical way, you know, all the time being aware of just what a daunting challenge it is.

Benjamin Thompson

And your work in this paper focuses on a few areas in the Asia-Pacific region. But do you think your findings are likely to be replicated elsewhere? Or is there something specific about this area which has led to the development of roads like these?

Bill Laurance

Absolutely, because we've seen these roads proliferating all over the place, no doubt in our mind at all. And we also did a little bit of checking just to see what the numbers were, you know, in other countries and what we could find out just simply. We found pretty much the same pattern, I think we looked at the Solomon Islands, Cameroon, and I think the Brazilian Amazon, and they were all in the same range of numbers as what we're getting in this Asia-Pacific region.

Benjamin Thompson

Bill Laurance from James Cook University in Australia there. To read his paper, head over to the show notes for a link.

Nick Petrić Howe

Coming up, how proteins that build cells’ skeletons may determine whether or not you’re left-handed. Right now, though, it’s time for the Research Highlights with Dan Fox.

<music>

Dan Fox

In 2022, hail the size of baseballs pounded Gerona in Spain. Now, new research has found that the storm, which killed one person and injured 66, was fueled by climate change. The storm occurred during an unprecedented six-week marine heatwave in the nearby Mediterranean Sea, which raised the temperature of surface waters by 3.27 degrees Celsius. A team of researchers used weather and climate models to analyse the heatwave and simulate alternative scenarios. They found that it intensified the storm by increasing the amount of moisture in the air and fueling the rising air masses that produced the hail. They say these extreme weather conditions would have been less likely in the preindustrial era. Read that research in full in Geophysical Research Letters.

<music>

The turbulent motions of the matter inside a star generate waves that make it surface throb like a beating heart. Now, astronomers have detected the smallest of such vibrations known as starquakes from a star 3.65 parsecs away. ε Indi is a dwarf star about three-quarters the size of the Sun. For brighter stars, starquakes can be spotted by measuring tiny variations in a star's light, but that method is insufficiently precise for objects as faint as this. Instead, astronomers used a spectrograph to split the starlight from ε Indi into its constituent colours. The repeated bulging and shrinking of the star’s surface affected the wavelengths of the light it emitted, leading to shifts in that spectrum. Analysis of this change allowed the astronomers to precisely record vibrations of the star’s surface as tiny as 2.6 centimeters per second. Vibrations such as these could provide clues to the star’s interior structure and composition. If you think that's gold-star research, you can read it in full in Astronomy and Astrophysics.

<music>

Benjamin Thompson

Finally on the show, it’s time for the Briefing Chat where we discuss a couple of articles that have been featured in the Nature Briefing. And do you know what I'm gonna go first, this week, Nick. And I've got a story about the beginning of a trial testing a potential way to treat people with liver disease. And it's a rather unusual one, it has to be said, because it involves growing additional mini-livers inside them. And I read about this in Nature.

Nick Petrić Howe

I mean, you certainly intrigued me with a lot of the information here. But why would such a thing be necessary? Like what's wrong with the current ways that we have to treat liver disease?

Benjamin Thompson

Well, that's a great question, Nick. And I will say that the numbers here are actually pretty large, because in the article, it says that more than 50,000 people in the US die each year with liver disease. And end stage disease can cause scar tissue to accumulate that prevents the organ filtering the toxic substances that are found in the blood. And this can lead to infection, and liver cancer. And one of the ways to treat this currently is a transplant. And we've talked a lot on the show before about, you know, there's a shortage of organs. And it turns out that 1,000 folk in the US die each year waiting for a transplant, and thousands more aren't eligible for a transplant. And so researchers are looking for ways to alleviate this shortage. Now, we've talked about some of them in the past, you know things like potentially using animal organs to fill the gap. But in this instance, a company called LyGenesis has really come about this from a rather different way. And so this involves injecting healthy liver cells from a donor into the lymph nodes in the upper abdomen of a person with liver failure. And the idea is that over several months, the cells will multiply and take over this lymph node to form a structure that will potentially take over blood filtering from the failing liver.

Nick Petrić Howe

Right. So, when you said ‘mini livers’ these are mini livers being grown inside the intended recipient.

Benjamin Thompson

Exactly right. And so far only one volunteer has undergone this process, right. That was a couple of weeks ago as part of a clinical trial and the company is hoping to enroll 12 people by mid-2025 into this trial. And what they've done is delivered donor cells from a donated liver through a tube in the throat, injected them say directly into a lymph node near the liver.

Nick Petrić Howe

Right? And is the idea that the liver grows in that lymph node, and then they replaced the failing liver? Or is it just going to take over the function of the failing liver?

Benjamin Thompson

Well, I think the idea is that it takes over some of the functions in the short term. Now this has worked so far in mice, dogs, and pigs. And in a 2020 study, donated cells formed these many livers within two months. And they had cellular architecture resembling a healthy liver with some of the kind of bile transporting cells and things like that. So you know, there was some promise there and the company think that growing mini livers in the lymph nodes are a good place to do it, because lymph nodes filter waste from the body anyway, and they're an important part of the immune system. But they have a large blood supply and there are actually hundreds of them in the human body, right. So the thought is that maybe using a few of them won't necessarily preclude the others from doing their job.

Nick Petrić Howe

And so what are they trying to find out with this trial? I guess just that this is safe and effective for the participants?

Benjamin Thompson

Yeah, absolutely right. So there's lots to understand. And say this is a human phase-two trial looking at, say safety, as you say, survival time, quality of life, but also trying to establish the ideal number of mini livers needed to stabilise someone's health. So as I understand it, they're going to test up to five lymph nodes to see if they can work out what the sweet spot is. But of course, this isn't without its downsides. The people involved will have to take immunosuppressants, this is very much similar to receiving a donated organ, and the effects likely won't be seen immediately say in the pigs, it took a few months for these mini livers to grow. And also, I mean, it's unknown exactly, you know what effects this will have. One researcher who's not part of the team suggested it might not relieve all the complications of late-stage liver disease. But I think the company are hopeful that this will potentially buy some time until a transplant liver is available.

Nick Petrić Howe

And this is maybe a silly question. But why couldn't they use the cells from the person themselves? Why are they relying on these donor cells?

Benjamin Thompson

Not a silly question at all. That is something that's discussed in the article. It hasn't been looked at yet but it could well be worth investigating whether a person's own stem cells could be used to generate the cells that seed the lymph nodes, as it was, but of course, a long way to go before that happens. And this trial will hopefully illuminate a bit more about the procedure as an actual kind of thing, I suppose. And I know that the company behind this are looking to, you know, grow kidney and pancreas cells in animal lymph nodes as well. So, time will tell whether it's a fruitful approach or not.

Nick Petrić Howe

Well, that certainly sounds like an unusual approach, but one that maybe will have some promise – we'll have to keep an eye on that one. But for my story this week, I've been reading a story in Nature, about a big genetic survey that is trying to find the sort of basis for why you might be left-handed or right-handed.

Benjamin Thompson

Okay, right… left, I suppose. Um because I mean, I guess, not that many people are left-handed compared with people who are right-handed.

Nick Petrić Howe

Yeah, it's about 10% of people are left-handed versus right-handed. And you know it's an interesting thing for researchers to look into, because it seems to be because of your very-early brain development during the embryonic stage that determines whether or not you're right-handed or left-handed. So if we can figure out why this happens, we could learn a lot more about brain development at that age, which could help us understand sort of various disorders and things like that, that have to do with development. But yeah, this study was looking specifically at the genetic variants between left-handed and right-handedness. And they found that there were a lot more rare variants in left-handed people within a gene that codes for a protein that is involved in building the microtubules that form like the skeletons of cells.

Benjamin Thompson

So yeah, microtubules then, the structure, kind of the scaffold of the cell in many way used to ferry proteins and things about. It seems, I mean, it's not immediately obvious how that affects which someone's dominant hand is.

Nick Petrić Howe

And that is something that is still unclear, but researchers have some ideas as to why this may be because this isn't the first study to find associations between these microtubules and left-handedness and right-handedness. And the thought is that these microtubules help form cilia, which you know from your biology background are like these hair-like protrusions in cell membranes, and these can actually help direct flows of extracellular fluid during development. And this can actually lead to an asymmetric expression of genes. And so you could have different gene expression in your right hand of your brain versus your left hand of your brain, and that can be associated with left-handedness and right-handedness, but as I say, that's not quite clear yet. But there does seem to be this association between these tubulins and right-handedness and left-handedness.

Benjamin Thompson

So what you're saying then. So these cilia and these kind of hair things are waving fluid in a different way outside the cell. And this is ultimately leading to a different expression, different turning on of genes inside the cell. Well, that's, I mean, wow, okay, right.

Nick Petrić Howe

I mean, that that's the idea. As I say, there's still a lot that we don't know about this. But certainly, from the study, it seems that these tubulins have a role here. And so it could just be that there are particular rare variants in people who are left-handed, which make the cilia form differently, and therefore lead to different asymmetries in their brains that then eventually lead to left-handedness.

Benjamin Thompson

And this is one of those things, it's kind of an interesting finding, or could this, you know, be expanded out into development more generally? Which I think you mentioned there.

Nick Petrić Howe

Yeah, no, that's the idea. So the fact that this only occurs in 10% of people kind of doesn't matter because ultimately, it affects everyone, because 90% of people have different genetic variants, they have the opposite. So by understanding this, we could better understand sort of brain development, as I said, in these sort of very-early embryonic stages, and that could help us understand like different disorders and things like that. But I think it's also just kind of cool, like, oh, this is why people are left-handed or right-handed, it's because their brains are just literally wired differently.

Benjamin Thompson

And it makes me think about people who are ambidextrous. And I'm right-handed, but when I used to do a bit of boxing I would box left-handed, and also my dominant foot is my left foot. But I ride the skateboard right-footed, so I'm not sure with folks who are maybe, you know, in this kind of area, what what the answer would be there.

Nick Petrić Howe

I mean, ambidextrousness, is an interesting thing but it's not really in the scope of this study. And it's something that I don't think we really know that much about, because we're still understanding what the sort of genetic basis is for left-handedness. So I feel like ambidextrous is a whole other puzzle that researchers who have yet to solve.

Benjamin Thompson

Well I will say, I'm certainly not ambidextrous. My writing with my left hand is absolutely a childish scrawl, and to be fair my writing with my right hand isn't that much better. But let's leave it there for this week's Briefing Chat. Nick, where can listeners find out more about these stories?

Nick Petrić Howe

Well, as always, those links will be in the show notes. We will have links to the stories that we discussed, and also a link of where you can sign up to the Nature Briefing to get even more like them delivered straight to your inbox.

Benjamin Thompson

Well, that's it for this week. But just before we go, a little bit of good news – one of our videos which is called ‘How would a starfish wear trousers? Science has an answer’ has been shortlisted for a Webby Award. Now two things: one, you should go and watch this video because it is great and two, you can help us win a People's Voice Award. It'll only take a couple of minutes so we'll put links on where you can do so in the show notes.

Nick Petrić Howe

In the meantime, you can reach out to us on X, we're @NaturePodcast, or via email on podcast@nature.com. I'm Nick Petrić Howe.

Benjamin Thompson

And I'm Benjamin Thompson. Thanks for listening.