Adaptitude: India and climate change

The Economist. Hello and welcome to The Intelligence from The Economist. I'm Jason Palmer.

And I'm Rosie Bloor. Every weekday, we provide a fresh perspective on the events shaping your world.

Given Donald Trump's With a convincing win, Republican control of Congress, a stacked Supreme Court, you might reckon he can do anything he likes. Not quite. As part of our series, The World Ahead, we separate the probable from the unlikely from the forget about it.

And robots have been used to make a lot of things more efficient. On assembly lines, to clean floors, even to perform medical tests. Now a new robot is being tried out. To pick grapes.

First up, though—

This is Sukanya Ashin. She's a young mother living in Wayanad, a region in southern India. And recently, as we sat in a side room at her workplace, she told me of the terrible mudslide that almost took her whole family and did kill 17 of her neighbors in 2019. The mudslide came very quickly.

Robert Guest is a deputy editor at The Economist.

So Kanya realized in the middle of the night that the back door had been blocked with shifting mud and they couldn't get out. So she and her husband, they wrapped their 2-year-old in a blanket and they waded out through the front door instead. They were lucky. As they left, their wardrobe floated off. And shortly afterwards, their whole house slipped down the hill. But they found safety on higher ground, and her husband rushed off to help other people. There was only so much he could do, though. Sukanya's family are among the many in the world's most populous nation whose lives are being disrupted by climate change and who are having to adapt.

And it's clear that adaptation in India comes with specific challenges.

Well, the country is hotter than the global average, and it's getting hotter still. It's poorer than the global average, so a lot of its people don't have much of a cushion when things go wrong. And there are 1.4 billion Indians, so it really matters that heat, drought, and floods are all getting worse and will continue to for years to come. Spending on adaptation is already high. It rose from an estimated just under 4% of GDP in 2015 to nearly 6% in 2022. And still a lot of experts say that's not enough.

So what does adaptation look like? What are Indians doing to adapt?

One response is to move. That's what Sukanya and her family did, and it's what the World Bank predicts between 10 and 40 million people in South Asia will have to do because of climate change by 2050. But for most people, how well they cope will depend on the spread of know-how, old and new.

What do you mean by that?

Well, for example, to avoid future tragedies, researchers in Wayanad are devising early warning systems for mudslides.

The Wayanad mountains cannot hold that much of water because they are high slope and the soil don't have that much of capacity to hold the excess water that is released through the high precipitation.

CK Vishnudas works for the Hume Centre for Ecology and Wildlife Biology.

We developed a community-based weather monitoring system after 2018 landslides and floods in Kerala. We identified local people for measuring rainfall, we trained them, and these local people are measuring rainfall data on a regular basis every day.

So it's about collecting microdata. You have farmers in landslide-prone areas recording how much rain has fallen on their land. and then they put all these measurements together in a WhatsApp group. And that lets the researchers at Hume give really accurate warnings about when the landscape is becoming unstable and therefore when people need to evacuate, and also crucially, where it's safe to move to.

Yet, as you say, landslides, mudslides are far from the only problem that Indians face.

Information's an essential tool to fight heat. Two, this is a big problem, especially in Indian cities. Poor neighborhoods tend to be much hotter than richer neighborhoods. I mean, the biggest slum in Mumbai is 5 degrees hotter than the neighborhood next door, which is a bit more prosperous. And a lot of that's to do with lack of shade. So if you're able to teach people, look, if you plant a small amount of trees on the wasteland near your homes, it's actually gonna reduce the temperature for everyone and give shade that, you know, maybe children can do their homework in rather than sweltering inside small cramped houses with tin roofs. Climate change is making this more urgent. So the worst heat waves in India will be 3 times as common if global warming is kept to 1.5 degrees and 5 times more if it hits 2 degrees by one estimate. A lot of the cities are trying to adapt with heat action plans, which involves not just planting trees for shade, but also issuing warnings when the, the temperature's too high and giving people drinking water in public places. But they really need to work on the architecture of cities rather than having lots of concrete and glass and relying on air conditioning, which frankly not everyone can afford.

I mean, it's a lot to tackle. What is the sort of single most urgent thing here?

The most urgent climate-related problem in India is probably water stress. So 17 of the 28 Indian states suffer either high or extremely high water stress by one estimate. And the hotter climate is gradually making it worse.

And by water stress, you mean just simply there's not enough of it to go around. I mean, how to fix that?

So almost all Indian cities rely on costly engineering to pipe fresh water in from distant sources. But generally speaking, there's not enough. Aramai Ravi, the director of the Indian Institute for Human Settlements, told me that water is still a huge problem where he lives.

In a good city, you have water coming 24/7. The best parts of Bangalore will have water coming once in maybe 1 or 2 days. The worst, uh, would have water coming once every 7 days or so.

In March 2024, Bangalore came close to running out of water, and that matters not just for Bangalore but also for the surrounding countryside, which relies on a lot of the same water sources, the same river. Paramar has a partial solution to do with water recycling.

What you're going to see is an attempt to take a large amount of water from the city and make that available to agriculturalists in the periphery who are growing things, who actually would normally be growing one or two crops in a season and are significantly impacted by the monsoon and climate change.

By cleaning up wastewater from homes and factories, so it's not clean enough to drink, but it's good enough for irrigating crops, they're able to recharge the groundwater that farmers use.

You say that's a sort of partial solution to the problem at hand.

Yeah, well, if you want to fix the issue completely, you have to think about price. So right now in India, farmers pay nothing for the groundwater they use under their land or the water they take from rivers. So they don't have much of an incentive to conserve. And householders tend not to pay very much either. It's only really big business that pays a realistic price for water. And that's where you see a flurry of high-tech firms devising clever ways to use it more efficiently.

Because there's a shortage and competition for water, industries are buying today water almost like 10 times the price they were buying almost like 5 years back.

Ganesh Shankar is the boss of FluxGen.

FluxGen helps in digitization of the water infrastructure, make sure the data of water is actually captured and put into the cloud. And using artificial intelligence, we've been able to identify inefficiencies in the water, such as leakage, wastage, excessive usage, and also ensures the rainwater harvesting system is efficiently operated and the wastewater treatment produces enough water to augment the freshwater needs.

Ganesh says his technology can reduce freshwater consumption by 30%.

But as we're on the question of what things cost to change for the better, what about the rich-poor divide that you said is already clear in terms of adaptation in India?

Well, inevitably rich people will be able to adapt faster than poor people. I mean, that's what rich people can do. You see, the biggest, most sophisticated cities will adapt the quickest. But overall, water management's improving in most Indian states. But still, an official survey in 2021 found that about half the cities with more than a million inhabitants were doing nothing at all to recycle the water. There's a chance that the terrible air quality in big cities like Delhi, which is not directly related to climate change, might nonetheless wake politicians up to the need to do something about green politics. One expert I spoke to, Sunita Narain of the Centre for Science and Environment, was quite optimistic.

We are ahead in thinking.

Practice may not be so evolved, but I think with climate change, that practice is going to get scaled up.

She says that as climate change progresses and the need to adapt becomes more obvious, that innovation will speed up. She'd better be right.

Thanks very much for your time, Robert.

Thank you, Jason.

While running for reelection, Donald Trump was mostly back to his old ways of making promises without ever really putting forward a policy platform. When he won in November, he made a pretty broad pledge.

We have a country that needs help, and it needs help very badly. We're going to fix our borders. We're going to fix everything about our country.

Everything about America. Like I say, pretty broad. Concrete stuff was hard to come by, with the exception of the 20 headline policy priorities published during the campaign. Thing is, as much as America's is a powerful presidency and the Republican Party has a lock on Congress, Mr. Trump can't just do anything and everything. So what will he be able to pull off? Tricky question. So as part of our World Ahead series, I asked John Prado, a United States editor and host of Checks and Balance, our sister show on American politics.

It'd be hard to get inside Donald Trump's head. He changes his mind. He can be unpredictable, inconsistent. Another way to do it is to look and say, okay, given these are Donald Trump's stated policy goals, in which areas does he actually have the power to make these things happen? And so what I did is go through all of those pledges and assign them a score between 1 and 5, where 5 is a dead cert, and 1 seems highly unlikely.

Right. So like hurricanes, like DEF CON. Okay, let's start with the easy stuff then, John. Of Mr. Trump's 20 policy priorities, which are the ones most likely to happen?

So the pledges that are most likely to be fulfilled, I think, are things that have happened before Donald Trump became president. And so when he becomes president, he can just look at them and declare that it is so. One of them is record energy production, record fossil fuel production. That's already the case under the Biden administration. Another one also in that category is to make America's economy hugely successful, make it great again. America's economy is extraordinarily strong, has been for some time, is the envy of other rich world countries. So that's sort of done already as well. And also in this category is to make America's cities beautiful again. I would argue that with the possible exception of Houston, which has other things going for it, most big American cities are rather beautiful. They have spectacular architecture and they're generally great places to live. So I think that's a tick as well.

Cue the letter writing from Houstonites. Okay, so let's keep moving through the categories. What's the next most likely bucket of things?

So category 4 is things that I think are almost certain to happen, things that are within the president's power and look very likely to be achieved. So one of those would be scrapping the EV mandate, which came in under the Biden administration to try and encourage Americans to buy electric cars. Donald Trump has said he'd scrap that. That's something he can do through executive action or executive order. Doesn't need Congress to do that. Another would be cutting regulation broadly. That's something that the Trump administration did a fair bit of in the first term. He famously now has Elon Musk and Vivek Ramaswamy working on this, That's likely to happen, and I don't think you have to cut a load of regulation to define that as a promise fulfilled. Another one is keeping the dollar as the reserve currency. It's really hard to see anything happening over the next 4 years that would change that. Another one in this bucket would be tax cuts.

Moving swiftly along, the neither likely nor unlikely squishy middle ground.

I'm going to put tariffs in category 3. I think tariffs will go up. But on the trail, Donald Trump said he would introduce a 10% universal tariff. He then increased it to a 20% universal tariff on all American imports. I'm not sure that's going to happen because I think that would be inflationary and would get unpopular pretty quickly. Also in category 3, I would put the pledge to stop trans girls competing in women's sports. This promise is a hard one to bring about. I expect to see an executive order from the Donald Trump White House that'll be tied to federal education funding, trying to use that as a lever to change policy on that. But the federal government doesn't have absolute power over what happens in American schools, doesn't have absolute power over all the many sports clubs throughout the country.

Moving swiftly along again to less likely still, bucket 2.

Bucket 2, I have record deportations. Donald Trump was very clear that he wants to increase the number of illegal or undocumented workers in America who are deported in his next 4 years. I think it's pretty likely he will do that. However, if you're looking at record deportations, then that's a pretty specific promise, and you have to go back and look at previous records. So the most recent record high was actually when Barack Obama was president, and then at the peak year, about 400,000 people were deported. In order to fulfill that pledge, to get a level higher than Barack Obama was at, there's lots of stuff you have to do. You have to build more migrant detention centres. You have to have even more funding for Customs and Border Patrol, which is very, very high already at a time when there's lots of competition for money in the budget. But going back even further, Jason, if you look under President Eisenhower's presidency, then there was a huge deportation programme. Most historians think that in the peak year of that, about a million people were deported. So I think there's a limit practically there. I think there's also a political limit. More than half of undocumented migrants in America have been there for more than 10 years. These are people who are really knitted into their communities. I would expect that were the federal government to uproot 10 million people, as Donald Trump has sometimes suggested he would, that would get unpopular fast. And Donald Trump doesn't actually like being unpopular. So I think that's a constraint too.

So that then brings us to bucket number 1, which I guess for you is just simply ain't going to happen.

In bucket 1, Jason, are make American students and universities patriotic again. I hope American students are as patriotic as the next American. However, telling them to be patriotic, the White House issuing orders ain't going to work. If anything, it'll have the reverse effect and students will react against it. So that's in category 1. Also in category 1 is peace in the Middle East. Several thousand years of history suggests that it'll be a stretch. And then also in category 1 is build an Iron Dome over America, like Israel's missile defense shield with American-made technology and weapons. The people I've talked to just suggested that is not practical in large part because America is bigger than Israel.

So some things then that Mr. Trump will find easy or already done, some things that are kind of in the middle, some things that are going to be functionally impossible. What is he gonna call a win out of all of this?

One of the things about Donald Trump is that he can define reality as he wishes it to be. I think on many measures, America is doing extraordinarily well at the moment. Its economy is strong, crime is falling. The country's moving in the right direction. Donald Trump will be able to come in and say, look at all the ways the country is flourishing and claim victory for that. And if you look at polling, such is the partisanship in the country and such is the degree to which people get their news from different sources, that won't seem strange to lots of people.

John, thanks very much for joining us.

Thanks, Jason.

The English countryside is increasingly full of vineyards.

Alex Hearn writes about artificial intelligence.

The Crouch Valley in Essex. Essex and northern Kent are surprisingly good areas for growing the sorts of grapes that in France get used to make champagne. Here they get used to make English quality sparkling wine, a very carefully legally protected term I discovered when I went out to Saffron Grange to one of these vineyards. Unlike champagne, obviously, this isn't a storied industry. The buildings and the human landscape of the English wine industry are new and they are purpose-built. You don't have the lovely tumble-down shacks of Burgundy or Champagne. You have big steel sheds and some nice glass tasting rooms. And at least when I visited Saffron Grange, there was another thing that was different as well. Two robots prepped for plucking grapes from the vines for the first time.

So Alex, robots wasn't where I was thinking you were going with that. Tell me about these robots.

This is a partnership between Queen Mary University London, a robotics company called Xtend Robotics, and the vineyard itself. The hope is to solve a few problems at once. When you send a human out to pick grapes, it's not like harvesting cereals, right? The reason why agricultural robotics works if you're growing wheat or corn is when it's harvest time, you rip it all up. You go top to bottom of the field, churn everything up, sort it out later, and that's why a combine harvester works. You don't do that with grapes. You don't want to uproot the vines for one thing, but you don't even want to pluck every single grape you see. Instead, you go down the same row multiple days, only picking the ripe grapes each time. So if you want to get a robot involved, it has to be able to tell whether or not a grape is ripe. It has to do that really using the same way a person would. And you do it with very rough heuristics, right? You look at a grape, you tell if it's roughly the right size, roughly the right colour, And you as a skilled agricultural labourer will know, maybe I should leave this till tomorrow, maybe I should pick it now. You'll even do things like assess whether or not there's a frost coming, and perhaps you should pick some grapes that are slightly less than perfect quality rather than risk them dying on the vine.

So how does a robot work out which grapes to pick?

The goal here is to use a technique called visual spectrometry. In the lab, this is Not easy, but fairly straightforward. If you or I look at a glass of water, it's see-through. If we look at a glass of ethanol, it's also see-through. An unpleasant mistake to make if you try and chug one of them. If you've got the right sensitive camera though, you can tell the difference. The wavelengths of light that pass through each of those transparent liquids are subtly different. And if you can pull up a spectrograph, the full chart of all of the wavelengths of light that pass through it, You can tell what water is and what ethanol is very easily. The trick is doing that not with glasses of water, but with grapes, and not in a lab, but on a vine, and not lit by a perfect white spotlight, but by the sun. That's somewhat harder, right? But the goal here is to build something that can ultimately look at a grape on the vine and tell from the light that passes through that grape exactly what the sugar level is, and from that sugar level to work out if it's at the optimum ripeness.

And if this works, what difference would it actually make to the winemaking business?

A pretty chunky one. Harvesting the grapes for wine is weird. You have to be incredibly precise. You will not find many soft fruits that have a sort of single-day window of ripeness. But if you want to make the best champagne, or in England's case, sparkling wine, you have to really get that window down. Your harvest will be about 2 weeks, but any individual grape on any individual vine there'll be a day, maybe two, that you would ideally like to harvest it. Do it too early, the sugar level will be too low. Do it too late and you run the risk of it going rotten, being eaten, being frostbitten, any of the other things that can cause your harvest to fail. Interestingly, one of the problems that you've had with climate change in England and France and around the world is that grapes are getting sweeter because their growing period is better because it's warmer. but that means that the resulting alcohol is getting stronger. So increasingly farmers, vineyard owners are having to move their harvests earlier for fear of making ever stronger, often rosés is the problem there.

Some wouldn't say that's a problem. So what you are saying is that having robots picking grapes theoretically saves you money in the picking process, but it also means you get better grapes, or rather at the right time. How likely is this actually to happen?

One of the really compelling things about this project is they're not trying to jump straight to full automation. The thing that makes FineWine uniquely suitable for this is even if you don't have an automated robot, you can still make some of this investment work out. There's the prospect, for instance, of setting up floodlights and letting workers in Australia, where the harvest happens at the other half of the year from the UK, dial in pilot these robots virtually and pick overnight in the UK. So you can compress a 2-week harvest to 1 week, make the most of that incredibly tight window of optimum ripeness, and then reciprocate when your labourers have nothing to pick in the British spring and Australia is in the middle of its harvest season. And then hopefully in 3 or 4 years, you end up with something that can become semi-automated. Can be guided but no longer need a human hand explicitly saying, "This grape and this grape, please." Fascinating stuff.

Thanks so much, Alex.

Thank you for having me.

That's all for this episode of The Intelligence. We'll see you back here tomorrow.

Health care can feel complicated. That's why Optum uses technology to connect the people and processes that make health making healthcare easier, more affordable, and more effective. We're making it clearer for you to know exactly what your benefits cover. And to help you better manage your health, we're coordinating care between your doctors and your technology. We believe better, simpler healthcare is always possible. That's healthy optimism. That's Optum. Visit Optum.com to learn more.