• Climate
  • climate

Climate Change Is Threatening Ketchup. AI Could Help Save It

7 minute read

Hold on to your Heinz. The latest looming food shortage is likely to include ketchup, coming hard on the heels of last year’s potato chip crisis and runs on mustard (in France, at least). Three summers’ worth of unprecedented high heat in the world’s key tomato-producing regions—Australia, Spain, and California’s central valley—have led to a precipitous decline in tomato paste stocks, the key ingredient for ketchup and other condiments. California, which produces a quarter of the world’s tomatoes, and 95% of the tomatoes used in U.S. canned goods, delivered nearly 5% less than the expected crop in 2021, and 10% less in 2022 due to the ongoing drought, according to the United States Department of Agriculture. Record-setting precipitation earlier this year helped with drought conditions, but it also flooded fields, forcing farmers to postpone planting, which could lead to reduced yields this year as well.

Other grocery-store staples are likely to follow suit, as climate change, driven by ever-increasing fossil fuel emissions, wreaks havoc on crops ranging from corn to canola oil. The impacts will ripple through the supply chain in unpredictable ways, leading to higher prices and shortages of not just weather-sensitive fruits and vegetables, but also items that seem about as far removed from nature as a Flamin’ Hot Cheeto is from a cornfield. While our staple crops might eventually adapt to a warmer climate, evolution takes place on a timescale wildly out of sync with market demands. If ketchup, cocktail sauce, and marinara for pizza are to keep up with demand, science is going to have to step in to help speed things up.

Tomatoes thrive in high heat, but like humans, they need cool nights to rest, especially when they are in bloom. If the hot nights of a heat wave last more than a few days, as they have started to do in major tomato-producing regions, the delicate yellow flowers wither on the vine, along with any hopes for juicy red fruit a few weeks later. Unlike, say, cereal companies that can switch suppliers when local crop shortages loom, most tomato-based product producers have vertically integrated supply chains: they provide their own seeds to contract farmers who grow to spec, and then transport the crop to processing facilities nearby that also belong to the producers. Most of those tomatoes are processed into paste, a shelf-stable ingredient that condiment companies depend upon to keep production running even when fresh ones are out of season. But three years on, those reserves are starting to run thin. “This year is going to be critical,” says Mariano Alvarez, an evolutionary biologist and chief scientist at Avalo, a North Carolina-based bioscience company. “If they aren’t able to create a healthy harvest it’s gonna be challenging for anybody that uses tomato paste in their products.”

More from TIME

Freakish weather has always been a challenge in agriculture, whether it’s the late-season cold snap that freezes fruit on the vine or an inopportune storm that flattens wheat just before harvest. But climate change is undermining some of the basic assumptions behind where, and when, we grow our crops. Age-old rivers are drying up. Hard winter freezes are increasingly rare, meaning pests get to live—and multiply—for another year. Heat waves are coming earlier, staying longer, and messing with fruiting cycles. Rain comes hard and fast, washing away newly planted seeds before they have a chance to take root. Our staple crops haven’t been able to keep up, let alone the luxuries we have come to think of as staples: bananas, coffee, chocolate, wine, olive oil, truffles.

We either have to adapt to life without the crops that make living worthwhile, or adapt those crops to our changing climate. Plant scientists like Alvarez are working on the latter. Upping stakes and moving to a better location isn’t always possible, says Alvarez. “For a lot of [tomato producers], the only option is to somehow change the biology of the plants themselves.” Alvarez is trying to do just that, using machine learning to come up with new crossbreeds better adapted to warming weather. He is part of a growing cohort of scientist-farmers that are drawing on innovation in robotics, chemistry, genome sequencing, genetic mining, and artificial intelligence to engineer plants for a climate changed future.

On May 16, another North Carolina-based bioengineering startup called Pairwise released the world’s first CRISPR-engineered salad greens on the U.S. market, using the relatively new gene-editing technology to take the bitterness out of nutrient-dense mustard greens. They are also developing a pitless cherry and a seedless blackberry. While their goal is to make healthy fruits and vegetables easier to eat and more palatable, the technology can be used to climate proof staples as well. Agriculture scientists in Australia have used genome sequencing to identify and amplify heat-tolerant traits in chickpeas, producing a protein-rich bean that doesn’t just survive, but thrives, when temperatures surpass 100°F. Another research team in Edinburgh, Scotland, is working on developing a “super cow” through gene editing that is more resilient to heat stress and disease, for sustaining populations living in the drought-stricken countries of the future. Scientists are now seeking sales approval for Australia’s first genetically modified fruit—a banana—that has been engineered to withstand a deadly fungus threatening worldwide banana production. (The U.S. already allows the sale of GMO apples, papayas, and pink pineapples.)

Alvarez’s version of assisted evolution relies on a simpler strategy. Instead of using genetic modification, which targets specific genes, his team uses machine learning models to seek out desirable traits—like drought or heat tolerance—in both cultivated plants and their wild cousins. Then an AI-enabled recommendation system suggests which crossbreeds might produce the best results for taste, ease of production, and resilience.

Wild strawberries, for example, are much more drought and heat tolerant than the ruby behemoths available in most supermarkets today, but they are small, and easier to bruise, which makes shipping long distances harder. The AI can suggest which strains of the wild strawberries should be crossed with domesticated breeds to make a larger, tastier, and climate-adapted successor. Horticulturalists have been doing that kind of cross breeding for centuries—it’s how we got the domesticated strawberry in the first place—but AI eliminates the trial and error, making the process faster. Once a potential cross breed is identified, the scientists go to the greenhouse to try it out, manually fertilizing the plants, planting the resulting seeds, and then waiting to see what comes up.

In addition to his strawberries, Alvarez has a couple of AI predicted crosses growing in Avalo’s North Carolina greenhouse: drought tolerant rice, broccoli with softer, more palatable leaves (to reduce food waste) and, of course, heat-resistant tomatoes. The hard part now is the waiting. While computers can speed up the selection process, growth is still on nature’s timeline. “We can’t make the plants flower and go through that development cycle any faster. The best thing we can do now is produce new varieties in a couple of years,” says Alvarez, instead of the 7-10 years it would take for the more conventional process. Computers may be able to accelerate evolution, but it’s still not fast enough to save ketchup for the summer BBQ season, or tomato sauce for next winter’s spaghetti and meatballs. “If people can hang on for the next few years, I think there’s a really good chance we can build something that is much more climate resilient,” says Alvarez. “In the meantime we just have to hope for good rains and cool summers.” Not just for securing the ketchup supply, but for the sake of all our condiments.

More Must-Reads from TIME

Contact us at letters@time.com