Western Farmer-Stockman Logo

Satellogic has been at work for eight years and hopes to image the entire plant weekly by 2020.

Willie Vogt

February 18, 2019

4 Min Read
satellite hyperspectral image of farms
GETTING LOCAL: This image is a 30-meter hyperspectral image of farms near Alison, Iowa, in the northeast part of the state.

There’s a kind of “space race” going on for satellite imagery. A wide range of startups is launching platforms with an eye on providing a range of industries a better look at the Earth. Agriculture is a prime target market for these firms, as evidenced by Planet’s work with Corteva’s Encirca platform and with Farmer’s Edge. And these firms are just getting started.

Satellogic is an 8-year-old startup that currently has eight birds in orbit, with the long-term goal of imaging the entire plant once a week at 1-meter resolution. To cut through the gobbledygook, at that resolution you could do a stand count from the image, or “see” weed infestations.

“The teams that are doing this are building the satellite constellation to get prices down,” says Marco Bressan, chief data scientist, Satellogic. “It will be feasible to launch hundreds of satellites.”

In fact, Bressan said that to fully image the Earth, the company would need 70 satellites at work, and they plan to launch 90 for redundancy. Those should be up by 2020.

“The goal of Satellogic is to focus on 1-meter resolution on a recurrent, weekly, basis,” he says. “In order to do that, we have to pull down the cost of satellites.”

The satellites, about the size of a small refrigerator, contain both hyperspectral and multispectral cameras on board. The hyperspectral cameras see more visible bands of light and provide increased insight into pollution or disease impacting an area. A third camera on board provides thermal imagery.

The company builds its own satellites and reception antennas as a vertically integrated business. Once that constellation is going, Satellogic will work with ag companies, farmers, forestry firms and others to provide imagery on a regular basis. And the service is new enough that the final cost of the service is not yet known.

Linking to others
Bressan knows that Satellogic won’t just be out there as a stand-alone image provider, but must work with other players to put images in the hands of farmers. “If farmers want images put into an existing geo-analytics system they might already have, we can provide that,” Bressan says. “We are finding out things farmers don’t want. They don’t want to build a cloud infrastructure, and they don’t want to get into big data.”

On a per-farmer basis that’s true, but Bressan acknowledges that there are other players out there providing those cloud services where this high-resolution imagery would have value.

Bressan sees the value of the imagery for a number of uses, from judging plant health to measuring crop yield. He sees the imagery, which will include hyperspectral data, helping predict soil nutrient needs. “We’re already seeing the power of 1 meter resolution when we enrich images from some other projects,” he says.

Bressan looks beyond the benefits of these images to individual farmers. Beyond the United States, he sees an opportunity for a company to help farmers get better on a regional basis: “You could monitor operations on a large scale and help those operations in a way that would be impossible with other technology. Even if you had hundreds of providers with drones, that would be basically impossible,” he says.

He points to coffee growers, noting that a small percentage of coffee grown goes to the espresso market. “If you had a group of farmers working with the same company, they have a common interest in providing a quality product for the end user,” Bressan says. The same could be true for other crops in the U.S., from food-grade corn to the potential for hemp.

Artificial intelligence
Bressan does note that while Satellogic is launching its own constellation, the company also works with a range of satellite imagery providers. It also works with scouts on the ground-capturing imagery. It’s all part of an effort to capture imagery to push it into a machine learning platform.

With machine learning, there’s a balance of capturing information, and then “ground-truthing” that information to verify decision-making tools for the future.

He notes that satellite imagery can also work to send alerts, noting if something is out of kilter in a field, which would then be used by scouts on the ground to direct drone use. “A satellite image can provide an alert, but not the resolution to make a judgment,” he says.

He does see a time in the future when an overarching system might link satellite images, drone imagery and even harvest systems together to provide cropping decisions. There’s a lot happening in this field, and it keeps evolving. Satellogic is another player in the mix to watch. Learn more at satellogic.com.

About the Author(s)

Willie Vogt

Willie Vogt has been covering agricultural technology for more than 40 years, with most of that time as editorial director for Farm Progress. He is passionate about helping farmers better understand how technology can help them succeed, when appropriately applied.

Subscribe to receive top agriculture news
Be informed daily with these free e-newsletters

You May Also Like