Vertical farms nailed tiny salads. Now they need to feed the world

Gartenfeld Island, in Berlin’s western suburb of Spandau, was once the bellows of Germany’s industrial revolution. It hosted Europe’s first high-rise factory and, until World War II, helped make Berlin, behind London and New York, the third-largest city on Earth.

Today’s Berlin is still a shell of its former self (there are over a hundred cities more populous), and the browbeaten brick buildings that now occupy Gartenfeld Island offer little in the way of grandeur. Flapping in the gloom of a grey November morning in 2020 is a sign which reads, in German, “The Last Days of Humanity”.

Yet inside one of these buildings is a company perched at agriculture’s avant-garde, part of the startup scene dragging Berlin back to its pioneering roots. In under eight years, Infarm has become a leader in vertical farming, an industry proponents say could help feed the world and address some of the environmental issues associated with traditional agriculture. Its staff wear not the plaid or twill of the field but the black, baggy uniform of the city’s hipsters.

Infarm has shipped over a thousand of its “farms” to shops and chefs across Europe (and a few in the US). These units, which look like jumbo vending machines, grow fresh greens and herbs in rows of trays fed by nutrient-rich water and lit by banks of tiny LEDs, each of which is more than ten times brighter than the regular bulb you’d find in your dining room. Shoppers pick the plants straight from the shelf where they’re growing.

Ériver Hijano

Gartenfeld Island, however, is home to something more spectacular. Here, in a former Siemens washing machine factory, stand four white, 18-metre-high “grow chambers”, controlled by software and served by robots. These are the company’s next generation of vertical farms: fully-automated, modular high-rises it hopes will scale the business to the next level. According to Infarm, each one of these new units uses 95 per cent less water, 99 per cent less space and 75 per cent less fertilizer than conventional land-based farming. This means higher yields, fresher produce and a smaller carbon footprint.

Agriculture is a £6 trillion global industry that has altered the face and lungs of the Earth for 12,000 years. But, unless we change our food systems, we’ll be in trouble. By 2050, the global population will be 9.7 billion, two billion more than today. Fifty-six per cent of us live in cities; by 2050 it will be 70 per cent. If the prosperity of megastates like India and China continues to soar, and our diets remain the same, we will need to double food production without razing the Amazon to do it. That sign on Gartenfeld Island might not be so alarmist.

Vertical farmers believe they are a part of the solution. Connected, precision systems have grown crops at hundreds of times the efficiency of soil-based agriculture. Located in or close to urban centres, they slash farm-to-table time and eliminate logistics. New tech is allowing growers to tamper with light spectra and manipulate plant biology. Critics, however, question the role of vertical farms in our food future. They are towering lunchboxes for late capitalism, they argue – producing garnishes for the rich when it is the plates of the poor we must fill. Vertical farms already make money, and heavyweights including Amazon and SoftBank are investing in various companies in the hopes of cornering a market expected to be worth almost £10 billion in the next five to ten years. Infarm is leading that race in Europe. It has partnered with European retailers including Aldi, Carrefour and Marks & Spencer. In 2019 it penned a deal with Kroger, America’s largest supermarket chain. Venture capitalists have handed the firm a total of £228 million.

Not bad for a hare-brained experiment that started in a Berlin apartment.

Ériver Hijano

In 2011, a year before he moved to Berlin, Erez Galonska went off-grid. He grew up in a village in his native Israel, but the young nation was growing too, and farms made way for buildings. Soon the village was a town, and its inhabitants ever more disconnected from their natural surroundings.

Galonska’s father had studied agriculture, and the son had dreamed of recovering a connection with nature he felt he had lost. The search took him to the mountains of the Canary Islands, where he found a plot of land and got to work. He drank water from springs, drew energy from solar panels and spent long hours farming produce he then sold or bartered at local markets.

When he met his now-wife Osnat Michaeli, “I traded it for love,” he says. “Love is stronger than anything.” In 2012, the couple, alongside Galonska’s brother Guy, who had studied Chinese medicine, moved to Berlin to work on a friend’s social media project. But the hunger for self-sufficiency remained. It was “a personal quest,” Michaeli says. “How we can be self-sufficient, live off the grid. Food is a big part of that journey.”

We meet at a Jewish restaurant in Berlin’s historic Gropius Bau art museum. It is mid-morning, and Covid-19 has cleared the tables. But a row of Infarm units whirs away quietly along one wall, producing basil, mint, wasabi rocket (a type of rocket leaf with the punchy flavour of wasabi) and other, more exotic herbs. Such produce was a pipedream for the three Infarm co-founders eight years ago. Growing crops when living on a tropical island was one thing. Doing it in a small apartment, located in the tumbledown Berlin neighbourhood of Neukölln, was quite another. Soon after moving from the Canaries, Erez Galonska typed “can you grow without soil” into Google.

Japan had taken to indoor farming in the 1970s, and this bore some helpful information on its techniques. The same was true of illegal cannabis growers, who swapped tips about hydroponics – growing with nutrient-packed water rather than soil – across subreddits.
Several trips to a DIY store later, the trio had what resembled a hydroponic farm. It was a big, chaotic Rube Goldberg machine, and it leaked everywhere. Growing wasn’t simply a case of switching on the lights and waiting. Brightness, nutrients, humidity, temperature – every tweaked metric resulted in an entirely different plant. One experiment yielded a lettuce so fibrous it was like eating plastic. “We failed thousands of times,” Erez Galonska says.

Ériver Hijano

Eventually the team grew some tasty greens. They imagined future restaurant menus boasting of food grown “in-farm”, rather than simply made in-house, and founded Infarm in 2013. But there was a hitch: indoor-grown cannabis sells for around £1,000 per kilo. Lettuce for £1.20. Most of the early vertical farms required heaps of manual work and operated in the red. “It simply wasn’t a sustainable business model,” Erez Galonska says.

By 2014, they decided to roadshow their idea, and shipped a 1955 Airstream trailer – a brushed-aluminium American icon – to Berlin. The trailer belonged to a former FBI agent, but it was conspicuous in a city of Volkswagens, caravans and Plattenbau buildings. Michaeli and the Galonska brothers transformed it into a mobile vertical farm, then pitched up at an urban garden collective in Berlin’s trendy Kreuzberg district. There they proselytised indoor farming to urban planners, food activists, architects and hackers, handing out salads and running workshops. Fresh, local food – even if it cost a little more – would entice a growing number of foodies who were interested in where their meals came from. The trailer cost nothing but petrol money to move, and emissions from the growing process itself were almost nil.

When the designer of a swanky hotel across town came by the trailer, he asked if the team could install something similar in his restaurant. “That was really the trigger,” says Guy Galonska. “We rented a workshop and we got to develop a system for them.”

When they installed their first “farm” in a Berlin supermarket, VCs took notice and visited Infarm’s young founders at their Kreuzberg office-cum-kitchen, where they hosted dinner parties featuring Infarm crops. But a return on investment still seemed distant: some investors thought the farms were an art project. Maintaining locations manually was exhausting, and the team almost went bankrupt “two or three times,” Guy Galonska says. “I think all of us got a lot of white hair during that time,” he adds. “It was a very challenging thing to do.”

A €2 million grant from the European Union in 2016 helped. With it came deals to place Infarm units in supermarkets and restaurants across Germany. Managing them all would require something precise, connected and efficient. To become a sustainable business, Infarm would have to behave less like a farm, and more like a tech startup.

Ériver Hijano

For around 2,500 years after King Nebuchadnezzar II of Babylon gifted his wife some hanging gardens, little changed in the world of hydroponic farming. Asian farmers grew rice on giant, terraced paddies, and Aztecs built “chinampa” rigs that floated along the swamps of southern Mexico.

Life magazine published a drawing of stacked homes, each growing its own produce, in 1909, and the term “vertical farming” appeared six years later. The US Air Force fed hydroponically-grown veggies to its troops during World War II, and Nasa explored the tech as a solution for life off-planet. But vertical farming didn’t really capture public imagination until 1999, when Dickson Despommier, a Columbia University professor, devised a 30-storey skyscraper filled with farms. In 2010, Despommier published The Vertical Farm: Feeding the World in the 21st Century, which has become the industry’s utopian testament.

“I had no expectations whatsoever that this would turn commercial,” Despommier says. “We just thought it was a good idea, because we didn’t see any other way out of stopping deforestation in favour of farming, and keeping the carbon dioxide content of the atmosphere at a reasonable level. It turned out to be a crazy idea whose time has come.”

The vertical farming concept is simple: growing produce on vertically-stacked levels, rather than side by side in a field. Instead of the Sun, the vertical farm uses artificial light, and where there is ordinarily soil, growers use nutritious water or, in the case of “aeroponic” farms, an evenly-dispersed mist.

Vertical farms take up a vanishing amount of land compared to their conventional cousins. They use almost no water, don’t flush contaminating pesticides into the ecosystem, and can be built where people actually live. But, by and large, they have not functioned as businesses. Only the black-market margins of weed, and Japan’s high-income, high-import food ecosystem, have catered to profit. It costs hundreds of thousands of pounds to erect a mid-sized vertical farm, and energy use is prohibitively high.

Advances in technology are changing this. By bolting automation, machine learning and cloud-connected software on to vertical farms, firms can trim physical labour, increase capacity and maintain a dizzying range of cultivation variables. Infarm staff at a separate office to the new Berlin farm, located some 23km southeast of Spandau in the Tempelhof district, keep track of “plant recipe” settings at any one of the startup’s 1,220 in-store units, including CO2 levels, pH and growth cycles, via the company’s Farm Control Cloud Platform, a bit like a giant CCTV room. Machine learning finesses recipes, and keeps each plant as uniform as possible.

Ériver Hijano

Gartenfeld Island’s employees – mechanical and electrical engineers, software developers, crop scientists and biologists – get closer to the produce, but only just. They monitor via an iPad and feed crops into the building’s four massive grow chambers, or farms, each one about the height and width of two London buses, with ventilation systems that whoosh like a subdued turbine hall.

From there on in, robots do the hard work. Inside the farms, a robotic “plant retrieval system” – basically a tricked-out teddy picker – scoots up and down a perpendicular beam, plucking trays of plants in various stages of growth and shuffling them closer, or further, from LED lights at the summit. The firm claims this reduces service time by 88 per cent. A sliver of window is the only way to see the device in-person: everything is hermetically sealed to keep out pests. “With automation, you invest once and then that price goes down over time,” says Orie Sofer, Infarm’s hardware lab lead. “With human labour, unfortunately, over time the price goes up.”

The number of crop plants varies depending on the produce, but there are usually just under 300 in a “farm” at any one time. Each farm yields the equivalent of 10,000 square metres of land and uses just five litres of water per kilo of food (traditional vegetable farming uses around 322 litres per kg).

Infarm is not alone in this revolution. AeroFarms, a Newark, New Jersey-based startup, feeds an aeroponic mist to roots that are separated from their leaves by a cloth. Its most recent funding round was led by Ingka Group, the parent of Swedish furniture giant IKEA. New York’s Bowery Farming, like Infarm, focuses on automation and a proprietary dashboard called BoweryOS that, among other things, takes photos of crops in real time for analysis. Its £123 million in backing comes from investors including Singapore’s sovereign fund Temasek. Bowery CEO and founder Irving Fain believes his addressable market “is about a hundred billion dollars a year, just in the US, of crops that we think are good candidates for us to grow.”

Leading the vertical farming VC race is Plenty, a San Francisco-headquartered brand that has raised almost half a billion pounds in capital since it was founded in 2013, including a 2020 Series D round led by Masayoshi Son’s $100 billion SoftBank Vision Fund. Plenty feeds its greens with water that trickles down six-metre-tall poles; infrared sensors pour data into an algorithm that nudges the plant’s growth recipe accordingly.

Plenty co-founder and chief science officer Nate Storey, who works at the company’s test farm in Wyoming, likens these deep-tech solutions to the tools that powered agriculture’s most recent revolution: “The tractor allowed farmers to be freed from constraints. Half of their land was dedicated to raising draft animals, and the tractor came along and freed them from a life where they were basically managing animals just so they could plough their land.”

For them, he says, automation is similar. “It allows us to get rid of the hardest work – the work that is unpleasant, the work [growers] don’t like to do – and focus on the work that really matters.”

Ériver Hijano

Infarm differs from the competition on two fronts. The first is its focus on modular design: each component is compatible and scalable, like a giant, noisy LEGO set. Modularity makes it possible to install Infarm units anywhere in the world in a matter of weeks, no matter the size. That enables the company’s second USP: its business model. Infarm has no stores, selling produce instead via its remote units.

Clients tell Infarm which produce they want, and “create a schedule,” says Michaeli. “You buy the plants. Everything in the farm is controlled from Tempelhof. Everything that’s grown belongs to the client.” A chef may demand pesto that’s made from particular three-day-aged Greek and Italian basil, for example. Infarm can do that (Tim Raue, Berlin’s most famous chef, is a customer). “Everyone stops and asks about the farm,” one Berlin store manager says. “It’s great to have innovation here.”

Infarm has “two big advantages,” says Nicola Kerslake, founder of Contain Inc, a Nevada-based agtech financier. “One is that they’ve figured out how to do product onsite, which is really not very easy. And the other is that they have these great relationships with big purchasers like Marks & Spencer.”

“When you look at where the arms race is in this industry,” she continues, “it’s really been in two areas: How do I get hold of as much capital as possible, and how do I sign up the right partners? Having Marks & Spencer in your back pocket is really useful.”

It has helped encourage investors to open their chequebooks. Hiro Tamura, partner at London VC firm Atomico, first met Infarm’s founding trio in 2018. A year later he led its £75 million Series B round. “They could roll these things out,” he says. “They worked, and they didn’t need some industrial sized warehouse to do it. I didn’t lean in, I fell into the rabbit hole. And it was incredible. I was like, wow, these guys are thinking about time and speed to market modularity.”

Infarm ploughs a chunk of its revenue back into research. In a mezzanine-level lab sitting above the farms at Gartenfeld Island, a dozen white-coated analysts conduct tests on herbs to a soundtrack of Ariana Grande, measuring crop sugar levels, acidity, vitamins, toxicity, antioxidants and more. Via a process of phenotyping – the study of organisms’ characteristics relative to their environment – they hope to create more flavourful plants, or new tastes altogether.

“It’s not just about the hardware,” Kerslake explains. “It’s about how the hardware interacts with the rest of your farm system. And we’re starting to see a lot more sophistication on that front, because the AI programs these companies started three or four years ago are now starting to bear fruit.”

Infarm’s results are high-quality: juicy lettuce, wasabi rocket that kicks, and basil that’s far more fragrant than the budget variety. “The end goal with almost everything that we’re doing is developing some sort of playbook, some sort of modular and standardised system, that we can then copy-paste to wherever we go,” says Pavlos Kalaitzoglou, Infarm’s director of plant science. Across from the lab, tomatoes and shiitake mushrooms grow in wine cellar-size chambers. They are living proof of how the firm is looking to diversify from herbs and leafy greens, whose low energy and water requirements make them the staple crop of every vertical farming startup today.

Ériver Hijano

We are in danger of farming the planet to death. Agriculture already occupies 40 per cent of all liveable land on Earth, and food production causes a quarter of all greenhouse gases. An area the size of Scotland disappears from tropical rainforests, responsible for up to a quarter of land photosynthesis, each year. Clearing more trees to feed our spiralling population will not help.

“We need to go back to the drawing board and rethink which avenues we can environmentally afford to pursue,” says Nicola Cannon, a professor at the Royal Agricultural University in Cirencester. Nitrogen fertiliser is particularly harmful to the environment, Cannon adds, “and has led us to adopting systems which have grossly exceeded the planetary boundaries.”

Current food systems are wildly inefficient: waste accounts for 25 per cent of all calories. And yet, almost a billion people suffer from hunger worldwide. These are not issues vertical farming will solve, critics argue. Going local does little beyond satisfying consumers.

Energy is another tricky issue. Ninety per cent of Infarm’s electricity today is renewable, and it wants to reach zero emissions in the next few years. But this doesn’t factor in the environmental cost of building a steel-and-cement facility.

“Vertical farms are a round-off error to the round-off error in terms of contributing to the big levers out there,” Jonathan Foley, an environmental scientist based in Minneapolis, says. “Like most technologies that are getting a lot of venture capital and which come from Silicon Valley kind of thinking, it’s being massively overhyped at the cost of real solutions. There’s an opportunity cost to put all this technology, money and renewable energy – that could be used for other things that we need energy for – into growing arugula for rich people at $10 an ounce.”

More than half the world’s food energy comes from its three “mega-crops”: wheat, corn and rice. They require wind, seasons and micronutrients that vertical farms are unable to replicate today. These are the crops that can prevent famine in Somalia, Bangladesh or Bolivia – not lettuce. “Vertical farms are growing the edge of the plate, not the centre of the plate,” Foley says.

But Despommier says it’s too soon to criticise the young industry for not addressing issues such as crop diversity. “What you’re really seeing is a rush towards profitability to get their feet wet, and to get their ledgers in the black and to pay off their investors, before they start diversifying,” he says.

“In a world where you think that land is unlimited and that resources are unlimited, indoor farming would be nonsensical,” Plenty co-founder Storey says. “As crazy as it seems to replace the Sun with electricity, it makes sense today. And it really makes more and more sense as time goes on.”

Much of the hope vested in vertical farms rests on the light-emitting diode. This tiny bead of light is the industry’s packhorse: it is a farm’s biggest financial layout, and the nucleus of its most exciting advances. Modern LEDs are nothing like the ones that powered your childhood TV. They’ve progressed at such a rate, in fact, that they’ve developed their own law to adhere to: “Haitz’s Law”. Each decade, their cost drops by a factor of ten, while the light they generate leaps by a factor of 20.

That curve will eventually plateau, experts say. But not before LEDs improve enough to allow vertical farms to profit from food closer to the middle of the plate. Infarm’s current smart LED set-up is over 50 per cent more efficient than the one that lit its first farms. Haitz’s Law has helped some companies experiment in growing potatoes, which require far more energy and water than leafy greens. Turning profit from a crop that delivers the highest calories per acre would be momentous for the industry.

The cutting edge of LED technology today is smart sensors that can regulate the brightness and spectrum of light to replicate growing outdoors – or enhance it. Much of the planet’s first flora grew only in the ocean, which looks blue because it absorbs blue light least.
Photosynthesis, therefore, occurs best between the blue and red light spectra. By tailoring LEDs to emit only these colours, or by dimming at intervals meant to mirror a plant’s natural cycle, vertical farmers can further reduce their energy burden – like stripping a road car to its bare bones so it can drive faster.

Recent discoveries have been more surprising. Strawberries, for example, react particularly well to green light. Some spectra can increase the vitamin C in concentrated fruits like kiwis, while others extend shelf-lives by almost a week. In the future, says Fei Jia, of LED firm Heliospectra, growers “can get feedback from the lighting and the plants themselves on how the lighting should be applied… to further improve the consistency of the crop quality.”
“If you judge it from what you have today, you understand what [critics] are saying,” Guy Galonska says. “How can you grow rice and wheat and save the world? And they are right. But they can’t see ten years ahead: they can’t see all the different trends that are going to support that revolution.”

Other technological advances are helping agriculture in different ways. Drones and sensors help map and streamline growing. Drip irrigation dramatically reduces the burden on dwindling water supplies. Circular production – where waste products from one process contribute to fuelling another – is becoming more commonplace, especially in livestock farming. Cell-grown or insect-based meat (or vegetarianism) will reduce our reliance on livestock, which consume 45 per cent of the planet’s crops. Infarm, and the broader vertical farm cohort, may not be saving the world today. But it wants to build taller farms, place them in public buildings like schools, and teach people the value of fresh, healthy vegetables. If 70 per cent of us are to live in cities, then cities “can become these communities of growing,” says Erez Galonska.

Ultimately, Infarm wants to build a network of tens of thousands of automated farms, each one pumping streams of data back into a giant AI system in Berlin. This “brain”, as Galonska calls it, will pour that information into algorithms to generate better food at lower costs, each new yield shaving fractions from the water, energy and nutrients required. Then, Infarm could become something closer to the dream Galonska left behind in the Canaries: truly self-sufficient.

It’s a long way from the leaky, DIY gadget he and his co-founders built in their front room. “The way the world is going now, it’s very clear to everyone it’s running into the wrong direction,” Galonska says. “We definitely believe in the power of collaboration: bringing those outside-the-box thoughts to create a new system that will generate more food, better food, much more sustainably, and help to heal the planet – because that’s the main issue on the table.”

More great stories from WIRED

🌍 Bill Gates has a plan to save the world. Will the world listen?

🇦🇺 Facebook did the right thing. Here’s what its fight with Australia is really about

🖥️ Working from home? These are thebest computer monitors you can buy

🔊 Listen to The WIRED Podcast, the week in science, technology and culture, delivered every Friday

👉 Follow WIRED on Twitter, Instagram, Facebook and LinkedIn