Skip to Content
Artificial intelligence

Farming a war zone

November 16, 2022
Stephanie Arnett / MITTR

Shortages of everything from seeds to fertilizer might accelerate the adoption of technologies that can help supplies go further in war-torn Ukraine.

We meet: 

  • Roman Tarasevich, Farmer, Ukraine
  • Morten Schmidt, Chief Executive Officer, OneSoil
  • Inbal Reshef, Program Director, NASA Harvest
  • Olekssi Misiura, Head of Research and Development, IMC

Credits:

This episode was reported and produced by Jennifer Strong, Emma Cillekens and Anthony Green. It was edited by Mat Honan and contains original music from Garret Lang and Jacob Gorski. Our mix engineer is Garret Lang. We had field production help in Ukraine from Orysia Khimiak. Special thanks this week to Max Furman, Ty Walrod, Antonio Regalado and Megan Zaroda Mullenioux. Our artwork is by Stephanie Arnett.

Full transcript:

[TR ID]

[sounds of dogs, footsteps, wind, door slams] 

[DUB]: Yes, yes. I'm just going from place to place now, looking for, for internet.

Да да, да, слышал, слышал. Я сейчас с места на место просто перейду, чтобы нам было понятно с Интернетом. 

Jennifer: Roman Tarasevich runs a 5-thousand acre farm in the Ukrainian region of Zaporizhzhya. 

[DUB]: I have three children. My eldest son, it's from his phone that I talk. He's already married, and I'm thankfully, a grandfather. So, we sent them all to Poland. 

у меня трое детей старший сын Кирилл. Это с его телефона я разговариваю. Он уже женат, и я уже, к счастью, стал дедушкой. То есть мы их отправили всех в Польшу

Jennifer: He and his son stayed behind and ran the farm after the invasion, until late June… when his family came back home.

[DUB]: Then Zaporizhzhya came under massive attack of S-300 missiles. This is generally an anti-aircraft weapon… And they were blowing up residential neighborhoods. Here in our neighborhood, in a nine-story building, one entryway was completely demolished… It was there. Then it was gone.

Потом Запорожье подверглось массированным атакам ракет С-300. Это вообще это противовоздушная. Оружие, которое земля воздух стреляет.  И разносили жилые кварталы. Вот у нас на районе в девять этажном доме полностью снесло там один подъезд, там его не стало. 

Jennifer: And last week, one landed nearby.

[DUB]: Thank God, everyone is in one piece, everyone is healthy. Next to the warehouse, the equipment was banged up, the warehouse was damaged. Well, what to do? 

You know, we are already adapting to these rails and we live. 

вот у моего коллеги на той неделе прилетела С-300. Слава богу, все целы, все здоровы. Рядом со складом перебивал технику побило, склад побило. Ну ну а что делать? Ну, уже, знаете, перестраивается на эти рельсы и живем.

Jennifer: And through it all, he continues to farm… as he’s done the past twenty years… though he says it’s getting much harder to work in the way he knows he should to be most effective… as supplies decline… expenses mount…and exports are at a standstill.

And yet, this year he managed to grow wheat, barley, corn… sunflower and canola.

[DUB]: Rockets break shrapnel in our fields. We have repeatedly, unfortunately, burst tires on combines, because it cuts, punctures everything. The threat is always there, but, well? What to do? I have 60 people working at my company all the time. And on top of that, 500 shareholders from whom I rent land. 

If I to take my family and go to Europe or somewhere else. What will these people do?

То есть ракеты, которые разбиваются осколки у нас в полях, мы уже неоднократно, к сожалению, глотали шины на комбайнах, потому что оно режет, пробивает все. Угроза всегда есть, но. Ну а что делать? У меня на предприятии постоянно работающих 60 человек людей. И помимо этого где то в районе 500 пайщиков это у тех, у кого я арендует землю. Если я могу себе позволить забрать семью и уехать в Европу и дальше куда нибудь. Ну а что эти люди будут делать?  

Jennifer: Much of the agricultural land in his region is occupied… and he believes many of those farms’ harvests have been looted. 

His land isn’t occupied… and for now his harvest is safe… but, beyond what he sells locally, it’s been impossible to move his crops.

[DUB]: There’s no point in hauling anything with the cost of diesel… of everything… we just cannot. So we have it all lying in warehouses now. That is, the harvest for 2021 and 2022, it’s almost all in warehouses.

нет смысла никакого везти. То есть стоимость солярки всего всего всего. Мы не можем. То есть у нас это все сейчас лежит на складах. То есть урожай за два года урожай 21-го года и 22 года это все практически лежит на складах.

Jennifer: I’m Jennifer Strong, and this episode, we explore how the war in Ukraine is changing what it means to farm in one of the world’s key breadbaskets. Where shortages of everything from seeds to fertilizer - might accelerate the adoption of technologies that can help those supplies go further… and we look at AI powered maps – created with years of Ukrainian data -- that are helping the sector plan and estimate yields.

[DUB]: I have to be here. That's all.

This is my land, my home, everything. Everything is here. 

Ну, я. Ну, я тут должен быть и все.

Это моя земля, мой дом, тут все, все все.  

[TITLES]

Inbal Reshef: Many people probably think about NASA and they think about space exploration. They think about astronauts. They probably don't usually think about agriculture. But actually the earliest satellites were designed with agricultural monitoring as one of the key applications. My name is Inbox Becker Reshef. I'm the director of NASA Harvest, which is NASA's food security and agriculture program.

Jennifer: It uses satellite data to better understand the world’s agricultural fields—like the impact a drought might have on nearby crops.  

Inbal Reshef: We have a lot of satellites that are passing over us all the time and collecting a lot of information about the earth's surface, about the air above it in wavelengths, both that we can see the visible wavelengths as well as many other wavelengths that our human eye can't see, but actually provide a lot of information.

Jennifer: Fun fact? They receive data from the entire surface of the Earth… including every agricultural field on the planet… each and every day.

Inbal Reshef: And that helps us to monitor both how a specific agricultural season might be developing and how a crop is growing and changing through that season. But it also helps us to look at how things have changed between one season to the next, how land cover and land use have changed.

Jennifer: It can help them study the impact of global warming. It can also help them understand how crops are doing in a war zone.

Inbal Reshef: Ukraine is a major bread basket of the world. It is a major producer and exporter of wheat. Around 10% of global exports prior to the war came out of Ukraine. Countries like Lebanon, over 80% of their wheat came from Ukraine prior to the war. The World Food Program I think around 40% of their wheat that goes into food aid comes out of Ukraine.

Jennifer: The country also exports about 40-percent of the world’s sunflower oil, and is a major producer of corn. In other words, Ukraine is really important to the global food supply.

Inbal Reshef: And so, long before the war, we'd actually worked a lot on monitoring agriculture in Ukraine and using satellites to map the crop plants, right? So where is wheat growing versus sunflowers? And to also be able to forecast what the ultimate yields would be. And so we had quite established relationships with different entities in Ukraine. And when the war broke out, we were in contact with their Ministry of Agriculture and forged a partnership with them to, to help, to support their analysis. And particularly what they asked us to focus on was the occupied territories where they couldn't get ground information and ground data. And it was really important to be able to understand what's happening in those territories.

Jennifer: This year’s wheat crop was planted last fall (before the war started)… and they were trying to figure out how much of that wheat would be able to be harvested, especially inside the occupied regions. 

Inbal Reshef: And so our first maps we released in, in April, and one of the first things we could do is say, well, Russia actually was occupying around 23% of Ukraine's total crop lands and around 29% of the planted area to wheat. And there were a lot of different estimates ranging anywhere between 30% to 50% in the occupied territories that would not be able to be harvested because farmers abandoned those fields… because they've been shelled and bombed… because their machinery has been destroyed. 

Jennifer: She continued to follow these crops on through harvest time… and we’ll hear more from her in just a bit. Meanwhile, ag tech companies working in Ukraine are also using satellites to watch these fields.  

Morten Schmidt: My name is Morten Schmidt. I'm the CEO of OneSoil. So I make data and information about what is growing and how healthy is it and I do it by combining satellite imagery with machine learning and actual data from fields to create products that can predict or diagnose in real time what is going on in the fields.

Jennifer: It's a Swiss company that originated in Belarus and is now based in Poland… and its data products are used by farmers as well as industry players.

Morten Schmidt: Let me see if I can show you my screen here. Now you should see my screen?

Jennifer: He's showing me a map that visualizes this data. It recognizes what’s growing… and predicts what the yields might be. 

Morten Schmidt: And here we are detecting 13 crops. And if we just do a quick zoom here on uh this is Indiana, you can see this very nice colorful picture. This is of course a visualization - the real value behind here is the data. If I go to my Ukraine map here this is what grows in Ukraine now. So basically this is a detection from beginning of August.  If I zoom in here you can see the Ukraine fields, they are different color because there are different crops than in Indiana. Here is much more sunflower - you don’t have that there.

Jennifer: This tool can be used to look back at the last six years… and he says it has very good data about Ukraine because that’s basically what helped him build it. And he’s using that data to try to figure out the impact of the war on specific crops.

Morten Schmidt: And the map that I'm showing you right now is exactly that analysis. So here we took corn and sunflower, which are two of the largest agricultural industrial crops in Ukraine. And we looked at each region and we compared it with last year's acreage. And you can see in the, in these eastern regions, we are at, you know, a significant reduction in these crops, which are the core industrial crops, but also the summer crops.  

Jennifer: Usually when we have these conversations about precision farming… or using data and technology to maximize crops and minimize waste… we’re talking about how these practices reduce pollution and are good for the planet…as well as for farmers' bottom line… 

But here we’re talking about a war torn nation that’s also a global breadbasket, and since the invasion, the most basic of farming supplies have been hard to come by.

And… that scarcity… it might actually be leading to quicker uptake of these precision tools, fueled by data and AI.

Morten Schmidt: There is less fertilizer, there's less seeds, there's less chemicals. So that meant that these technologies that have been available and were used to some extent have been accelerated because they help them get the maximum output of the scarce resources they have. 

Jennifer: He shows me how it works.

Morten Schmidt: You can see the productivity map that we detected on the, on the right side and on the left side you can see the actual measured map from the combine. And that allows us to help the customer to make his planting map or his fertilizer map that he then loads into his tractor and allows him to do this variable planting and fertilization or chemical application of the field.

Jennifer: You can find links to our reporting in the show notes and you can support our journalism by going to Tech Review dot com slash subscribe.

We’ll be back… right after this. 

[MIDROLL]

Olekssi Misiura: Okay. My name is Oleksii Misiura. I'm head of R&D Department of I-M-C. IMC it's like a top 10 Ukrainian Agri Holdings. Uh, we are doing mostly cash crops - corn, winter wheat, sunflower. 

Jennifer: He’s been working with OneSoil for the last three years… providing data from their lands and testing their products, among many others.

Olekssi Misiura: I'm searching for new technologies in growing crops and I'm testing it. Perhaps I was planning to go more in production, in management, but to try something new, I think it's more fun. 

Jennifer: Of the projects he’s working on, he says he’s most excited about the maps we heard about just before the break… where crops can be monitored down to the individual seed. 

Olekssi Misiura: And you, you can see it online, you can, you have a iPad in a tractor connected to the monitor of the planter. And uh, basically you can see at home and you can see on your personal ipad all these problems. You can like manage it from, from your house. Basically.

Jennifer: Not every part of a field is going to have the same productivity, and this helps farmers direct limited supplies to the most productive parts.

Olekssi Misiura: Essentially these prescription maps which you use to differentiate your seeds, like to sew different rates of seeds in different parts of field, you need a prescription map. We used to create it, like, manually on, on our computer it took us, uh, from 30 to one hour for one field. Now it's like it takes, uh, one minute, maybe two. // including the, like, downloading this map to the monitor of a spreader.

Jennifer: It’s one way the war with Russia has accelerated change in Ukraine’s farming industry. He says another is the adoption of drones. 

Olekssi Misiura: If we take our problems with the war. We used to, we used helicopters for spreading chemistry on our crops because it is high and you can't use a normal sprayer without harming the plant. And now when the war came, we have fights on some of our land, then it was de-occupied, but still, aviation is forbidden there. So we started to use drones. It has less productivity than helicopters, but I think in future everything will, will be spreaded by drones. 

Jennifer: A drone takes more time to spray everything but also wastes a lot less water and chemicals.

Olekssi Misiura: Ordinary sprayer. You, you use like, a hundred to 200 liters per hectare. And, when you use a drone, you, you might use like 10, 20 liters per hector and even less five liters. It's good for our water usage.

Jennifer: It’s safe to say most farmers don’t have access to all these tools. Especially if they didn’t already have the right infrastructure and equipment on hand before the war started, and that includes the farmer we spoke to earlier.   

But there’s something farmers everywhere do have… and that’s a deeply held sense of purpose. Without farms, there’s no food. 

And thanks to satellite data, we now know how much of the crops farmers were able to harvest, despite being in the middle of a war.

Once again, this is Inbal Becker Reshef from NASA Harvest.

Inbal Reshef: As the season progressed, we could then look at how much was actually getting planted. And what we found was that by and large, most fields were actually getting planted, including in the occupied territories. And so we found that only about 11 or 12 -percent of Ukraine’s agricultural land went unplanted… and that those areas were largely in the occupied territories… and along the front line. 

Jennifer: It means more land was planted with summer crops than had been expected. The other big question was ‘how much wheat planted before the war could be harvested’. And watching from the sky? She could see it was much more than what was included in those early estimates.

Inbal Reshef: And in fact, we found that 89% of the wheat in the occupied territories was harvested. What of course we can't answer is who's harvesting that wheat, right? So satellite data can tell us that it's being cultivated, it's being planted, it’s being harvested. We can't say who is harvesting that wheat… Where is that wheat ending up being stored… Is it getting exported? All of that, of course, raises very important questions.

Jennifer: Estimates for Ukraine’s wheat harvest had been in the range of 20-million tons. But, final numbers suggest it’s almost 7-million tons higher. And she says more than 20-percent of the wheat harvest came from occupied territory… which conservatively… is worth about a billion dollars… assuming it can be exported. 

Jennifer:  And in the future, she thinks this type of analysis will only become more common… / but she says you need local partners to do it well… so that what’s done on the research side with data and machine learning makes the correct assumptions and fits with what's actually needed on the ground.

Inbal Reshef: Because today it's Ukraine, but tomorrow it can be any other

country, right? Whether it's due to conflict or whether that's due to

natural disasters or a drought or a flood… And so it’s really important information that satellite data can provide and I see us moving a lot more in that direction of being able to respond to policy needs and requests.

Jennifer: This episode was produced by me, Emma Cillekens and Anthony Green. It’s edited by Mat Honan and mixed by Garret Lang, with original music from Garret Lang and Jacob Gorski. 

We had field production help in Ukraine from Oryssia Khimiak… and special thanks this week to Max Furman, Ty Walrod, Antonio Regalado and Megan Zuh-roh-duh Mull - en - yo.

Thanks for listening.  I’m Jennifer Strong. 

[TR ID]

Deep Dive

Artificial intelligence

A Roomba recorded a woman on the toilet. How did screenshots end up on Facebook?

Robot vacuum companies say your images are safe, but a sprawling global supply chain for data from our devices creates risk.

The viral AI avatar app Lensa undressed me—without my consent

My avatars were cartoonishly pornified, while my male colleagues got to be astronauts, explorers, and inventors.

Roomba testers feel misled after intimate images ended up on Facebook

An MIT Technology Review investigation recently revealed how images of a minor and a tester on the toilet ended up on social media. iRobot said it had consent to collect this kind of data from inside homes—but participants say otherwise.

How to spot AI-generated text

The internet is increasingly awash with text written by AI software. We need new tools to detect it.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.