Hyperspectral aerial images of crops, taken using technology developed by Cloud Agronomics

Brown student startup seeks to revolutionize crop disease prevention

Cloud Agronomics — a student and alumni venture launched with support from the Nelson Center for Entrepreneurship — uses hyperspectral imaging to detect crop-borne diseases that destabilize food supplies and cost farmers billions.

PROVIDENCE, R.I. [Brown University] — In the U.S., crop agriculture is billion-dollar industry with billion-dollar losses. While the country’s crop production is valued at roughly $200 billion annually, the industry loses as much as 25% of its crops each year, reducing both revenue and food supply.

Among the biggest culprits is disease. The vast majority of U.S. farmers still rely on the human eye to detect diseases in their plants — even though, by the time a disease becomes visible on a plant’s surface, a lot of damage has been done, said Jim Kellner, an associate professor of ecology and evolutionary biology and environment and society at Brown.

That limitation presents an opportunity for innovation, he said: “Consider what would happen if you could see inside the plant, and then use that information to make changes at a time when it’s still possible to affect the outcome.”

A new Brown startup has begun to do just this. Founded in 2018 by three Brown undergraduates, Cloud Agronomics flies airplanes carrying hyperspectral imaging technology — originally developed to help astrophysicists discern the composition of planets and stars — over crops to see what is happening in the plants on a molecular level.

“We think of it as a CT scan for crops,” said Jack Roswell, one of the company’s student cofounders and its chief operating officer. “It allows you to see beyond what the human eye can see — to really see into plant leaves and soil.”

And because the company captures aerial imagery, it is able to scan miles of fields in minutes.

“Up until now, using this imaging on plants has been the sole realm of lab research,” said Mark Tracy, chief executive officer of Cloud Agronomics and a Brown Class of 1995 graduate. “No one thought of deploying it at such a scale that it would be able to impact entire agricultural fields at a time — that rather than driving up to a plant with an instrument, you could fly over an entire field and learn its composition.”

The cofounders’ focus on solving a real-world problem — and their ongoing engagement with faculty and alumni mentors — has led Cloud Agronomics to the brink of a revolutionary agricultural innovation, said Jonas Clark, associate director of the Nelson Center for Entrepreneurship at Brown.

“Often entrepreneurs have a great solution and they run around looking for a problem,” Clark said. “What’s impressed me most about this team is how it’s always been about trying to solve this problem and doing the work to get there.”

From failure to launch

When the founders of Cloud Agronomics met in an engineering class during their first year at Brown, all three were itching to create something. They just weren’t sure what yet.

“The three of us were all taking apart computers and TVs in high school,” said Alex Zhuk, company president and cofounder. “And so we came together because of our sheer curiosity to build something.”

Over the course of that spring semester and the following summer, the pair — joined by Brown peers David Schurman, who would become Cloud Agronomics’s third cofounder and chief technology officer, and Julian Vallyeason — set out to build a drone that could run on solar energy alone. “Our only intention was to break the world record, which was 80 or so hours, or about four days in flight,” Zhuk said.

With the drone finally built, the moment of truth came — only to quickly pass: “We spent eight months building it and it crashed in about 8 seconds,” Roswell said.

But an idea arose from the proverbial ashes the next spring. With funding from the Brown Venture Prize — an award given annually by the Nelson Center for Entrepreneurship to sponsor student ventures with outstanding promise — Roswell and Zhuk bought their first spectrometer, a piece of equipment that reveals the composition of matter by separating and measuring layers of light within it. 

They brought the new equipment with them to Florida, where they traveled to research a crop disease that had been devastating the citrus industry.

“We just called professors at different land-grant universities and said, ‘We know you have this problem with your citrus crops. Would you mind us visiting your fields, talking to you and maybe even collecting some data?’” Zhuk said. “And they were all extremely open to us working with them.”

Up until now, using this imaging on plants has been the sole realm of lab research. No one had thought of deploying it at such a scale that it would be able to impact entire agricultural fields at a time — that rather than driving up to a plant with an instrument, you could fly over an entire field and learn its composition.

Mark Tracy CEO of Cloud Agronomics, Class of 1995
Tweet this quote

The pair drove from farm to farm, talking to farmers, cutting leaves from their plants, and using the spectrometer to collect measurements that could identify the beginning stages of disease weeks before the signs would become visible.

As Roswell and Zhuk gathered data by hand in the Florida citrus fields, they were reminded of the drone technology they had dabbled with the year before. “We realized then that, if we could deliver these types of insights and analytics on a square-meter basis, it would have much larger implications,” Roswell said.

Upon returning from Florida, Roswell and Zhuk reconnected with Schurman. Together, they began experimenting with collecting and calibrating this spectral technology first from drones, then from planes.

By using aircraft to provide molecular-level detail on an enormous scale, Cloud Agronomics is now poised to revolutionize the way crops are managed, Kellner said.

Schurman with a hyperspectral imaging machine
Schurman with a hyperspectral imaging machine.

“Not all grain is created equal,” he said. “It’s graded based on its molecular composition, and the different grades are used for different things. If you can tell someone early in the growing season where additional investment in a crop is likely to have a the largest payoff, that would improve efficiency for farmers and reduce waste.”

The technology is also able to measure and track carbon content in soil on a scale large enough to impact the ways that companies and governments monitor carbon emissions, making it easier for them to develop plans to reduce their carbon footprints.

To date, Cloud Agronomics has taken on such high-profile partners as Microsoft AI for Earth and other global agricultural giants. In June, the company hit a fundraising milestone, raising seed capital to launch its next phase.

“This seed round will provide us significant runway to expand our team, expand our commercialization of our dynamic crop nutrient solution, and establish the first scalable and scientifically rigorous carbon index to verify carbon credits on farmland soils,” Roswell said.

Currently, the team is working to combine data collected from planes with satellite data: “With this step, we’re hoping to actually scale this analysis internationally and reach communities that could really benefit from this information,” Zhuk said.

As the group has expanded the scope of their spectral imaging to cover larger and larger areas of land, they have also narrowed the scope of the data that they choose to analyze.

“These images collect a lot of information,” Schurman said. “We’re really homing in on the details that people actually care about — isolating parts of our imagery that are important for particular physiological properties of plants and soil in agriculture and presenting those in a new way to farmers.”

With each iteration — from bringing the spectrometer to the fields to launching it in the air, and from aggregating every thread of spectral data to curating those most relevant to agricultural improvement — the technology behind Cloud Agronomics had increased in complexity, creating more opportunities for trial, error and growth.

“There are thousands of moving parts that have to fit together perfectly,” Schurman said. “And if one of those goes wrong, the entire train stops. But the most powerful thing about failure is finding creative solutions around a problem, and expanding your mind to think about all of the ways that you can recombine the parts you already have to make a working system again.”

“Whether looking at our first drone, our classes at Brown or the many new experiences we were introduced to in college — learning from these sorts of challenges and figuring out how to navigate them has been very powerful.”

A Brown-based breakthrough

Since its inception, Cloud Agronomics has gained key collaborators from within the Brown community, with alumni and faculty supporters becoming key investors, mentors and team members.

Advice and financial support offered by alumni — many who learned about Cloud Agronomics through their Brown Venture Prize presentation — were critical during the company’s early stages, Zhuk said.

“A lot of alumni reached out to us simply because they believed in the idea and in us before we ourselves even had arrived at our mission and vision for the company. We are so grateful to the Nelson Center and to the extremely supportive alumni base that the center helped us reach.”

One of those alumni is Don Stanford. A longtime advisor to Brown undergraduate entrepreneurs, Stanford — who also teaches computer science at the University — became a mentor and strategic investor to Cloud Agronomics in its early stages after hearing about it through the Nelson Center.

“I was immediately taken by the overall vision of the project,” Stanford said. 

Tracy was similarly impressed with the trio and their project when he was first introduced to them via the Nelson Center in 2018.

“I’ve spent a lot of time around Brown students — as an alumni interviewer and a mentor for student entrepreneurs — and it’s hard for me to overstate how extraordinary David, Alex and Jack are, particularly because of their self-awareness” he said. “When you’re trying to do something so disruptive and so hard, it’s important to understand what you know and also what you don’t — to try and fail, and to not have hubris get in the way of your vision.”

Watching this happen in real time — them doing what they are doing with world-class science, technology and machine learning — I think this is destined to be one of the great breakthrough projects for the Nelson Center and for Brown.

Don Stanford Computer Science Professor, Class of 1972 (AB) and 1977 (MS)
Tweet this quote

The cofounders of Cloud Agronomics have also sought advice from faculty experts like Kellner, who first met the trio when they approached him to discuss how remote sensing technology might be applied to agricultural challenges.

For him, what most distinguished this from other student ventures was the team’s commitment to identifying and solving an important problem.

“When people learn to use new technology, it’s like they have a hammer and every problem they see looks like a nail,” said Kellner, who will be joining Cloud Agronomics as its chief scientist. “But some nails don’t need to be hit. They were different because they were invested in finding a real problem that hyperspectral imaging technology could solve.”

By applying large-scale hyperspectral imaging to the agriculture industry, Kellner believes, Cloud Agronomics has found the right problem.

“Agriculture has always operated with limited information on the composition and health of crops,” he said. “With more information, that system can be both more productive and more efficient at producing food. Cloud Agronomics generates that information.”

Stanford agrees: “Watching this happen in real time — them doing what they are doing with world-class science, technology and machine learning — I think this is destined to be one of the great breakthrough projects for the Nelson Center and for Brown.”

“They are a great example of what can happen when you harness student creativity with supports from the Brown community.”