This App Lets Kenya’s Farmers Access Satellite Data to Monitor Crops

Climate change is the most horrific threat our species has ever known: No matter how powerful you are or how much money you have, our transforming planet is a reckoning for every one of us. But there are degrees to this misery. If you’re perched in a Manhattan penthouse, the effects might not be immediately apparent (because you don’t care or aren’t paying attention, or both). If you’re a subsistence farmer in Kenya, the situation is already much more dire.

There’s an equalizer, though, that is helping small farmers adapt to a changing planet: smartphones. More specifically, an app called PlantVillage—point your phone at a diseased plant, and artificial intelligence will analyze the leaves and tell you exactly what’s gone awry, so you can appropriately treat the problem. And now, in addition to that detective work, it’s gaining new powers thanks to eyes in the sky. Using free satellite data from the UN, PlantVillage can monitor biomass on a plot of land, giving small-scale farmers insight into how their crops are developing.

These days, that data isn’t just valuable—it may well become essential as our climate descends into chaos. No longer will rich companies in rich countries get exclusive access to fancy satellite data, while those hit hardest scramble to subsist.

Researchers have been testing the new system in western Kenya, handing out phones to “lead” farmers, who travel around the community bringing insight to their neighbors. You can see what users would see in the screenshot below. Notice that the data comes in every 10 days, so the farmers can track growth over the course of a whole growing season.

plant village

“If it’s not going well, then you have to do something very different from what you’ve done before, either plant a different crop or start irrigating,” says Penn State biologist David Hughes, who developed PlantVillage. The satellite data also reaches years into the past, so the farmers can see how well their plants are performing historically. On top of that, they can compare their performance with their neighbors’. Meaning, maybe it’s not a rainfall problem, but a pest problem that’s holding their crops back.

This is far more granular than what has been available before—continental-scale maps of Africa showing good and bad areas for agriculture. But for a farmer in an area running low on rainwater, they can’t just pack up and leave. So what they need is information on how they might adapt to a changing climate. That’s easy enough for industrial-scale farms with money to spend, but has been unobtainable for small-scale farmers. With this new service from PlantVillage, that’s beginning to change—all it takes is one smartphone per community for essential data to flow to farmers.

“We’re able to tell them they have a problem, and with the AI assistant go into the field and see if that problem is a disease that’s causing a decrease in the yield, or it’s actually drought,” says Hughes. This is particularly useful because the drought problem and the pest problem are intimately linked: Warm weather conditions lead some insects to breed more prolifically.

In western Kenya, farmers have historically relied on predictable rains. So every January, farmers would begin preparing their fields, then sow their seeds in February, then expect rains in March.

“Of course, this was information communicated over history,” says plant pathologist John Chelal, who grew up not far from where he and Hughes are testing the new system. “The common response I hear is, we don’t know anymore. The rains have become quite unpredictable.”

With this satellite data, though, they can react before it’s too late. “Maybe think about putting up some irrigation facility, or organizing how they can even manually apply water,” says Chelal. “This is a really objective value. They can’t see it with their eyes, but they can get a figure.”

That value comes from satellites that are looking at a couple of things. For one, they’re using near infrared sensors to measure land surface temperature, which helps calculate soil moisture. This helps form what’s known as the normalized difference vegetation index. “It can be used to see how healthy a crop is, and it can tell you what part is water and what part is bare soil, and so on,” says Jippe Hoogeveen, senior land and water officer with the UN’s Food and Agriculture Organization, who’s working with PlantVillage.

At the moment this data doesn’t come at a particularly high resolution—pixels of 100 meters square. “But at the same time, the good thing about satellite data is you have consistent data over time and space,” says Livia Peiser, FAO technical land and water officer. “So you can very easily compare the trend of your pixel in the last 10 years.” Plus, that resolution will only improve: In the near future, Peiser and Hoogeveen expect new satellites to sharpen the resolution to 10 meters square, giving farmers ever-better granularity.

Another issue at the moment is that it’s hard for the satellites to parse the subtleties of this kind of small farming. So for instance, the two staple crops in western Kenya are maize and cassava, which are often planted together. Farmers might harvest the maize first, thus changing the biomass signature of the field, while leaving the cassava to continue growing. They might also leave plant residue on the ground after harvesting the maize, further complicating the signature.

“It’s not going to be a perfect science,” says Hughes. “But at least it will be a science that is now involving the farmers, as opposed to what it’s been traditionally, which is just a bunch of academics providing these tools for broader-scale policymakers to understand.”

Satellite data won’t get us out of the mess we’ve made for ourselves. But it may well help the most vulnerable adapt to a climate gone mad.

More Great WIRED Stories

social experiment by Livio Acerbo #greengroundit #wired