By Ben Potter and Nate Birt
Drones draw industry buzz, but limitations lurk
Did you know more drones buzz around U.S. airspace than airplanes? This new, albeit disruptive, technology continues to fill the skies.
While the vast majority of unmanned aerial vehicles are flown by hobby enthusiasts, more than a handful are showing up on farms and ranches. Some merely go up to count cattle faster or scout crop diseases more efficiently. Others are getting a heavier workload—taking hundreds of photos and creating complex NDVI or infrared maps that empower farmers to make management decisions on the fly.
Those with experience flying a drone on the farm don’t question the technology’s usefulness—depending on how it is used. For example, Illinois precision ag specialist Brad Beutke says most farmers don’t have the required amount of time or energy to gather images and analyze them over even a few hundred acres.
“A drone as a scouting tool and for research purposes is still probably a good idea,” he adds.
Is there a viable alternative for scanning larger areas of farmland? Several agricultural technology startups are providing satellite imagery and analysis to their customers as an alternative to drone-flown image gathering.
“We use satellites to look at every farm in all 3 million square kilometers in the Corn Belt every day, instead of just sampling a few farms,” says Mark Johnson, Descartes Labs cofounder and CEO. “Even if you added up all of the drones in the U.S., the percent coverage of fields would be minimal. By contrast, satellites have far greater breadth, providing coverage of the entire U.S. on a daily basis.”
Drones do pack a few advantages of their own. Flights are immediate and under the farmer’s control. As Johnson points out, one of the biggest advantages of drones is they fly below clouds.
Even so, Johnson says he can foresee a time when new “satellite constellations” could encroach on drone precision ag applications. Descartes Labs works with satellite manufacturer Planet Labs, which is launching a constellation that will be able to cover the entire Earth every day at a 3-meter resolution.
“That may be useful enough for many field monitoring applications and even better than drones because of the frequency of observations,” he says.
Even though satellite imagery lacks the “on demand” aspect of drones, Rich Kottmeyer, vice president of food and agriculture for custom technology company Luxoft, says most farmers will find images captured at three- or five-day intervals to be completely adequate.
With more virtual eyes circling the globe than ever, Kottmeyer predicts satellite use will spread much more widely during the next three years.
“Satellite is going to become the primary way in which people gather data on the farm,” he says.
Jesse Vollmar, CEO and co-founder of FarmLogs, notes another interesting advantage that satellite imagery has over drones—the images don’t have to be stitched together. Why is this significant? Aside from time involved, Vollmar says it cuts down on potential error.
“If you have various images of one field, there’s a chance that many of those images will look the same—especially if that field doesn’t have many distinguishing landmarks,” he says. “Putting them together would be like putting together a puzzle with pieces that all look the same.”
Of course, for farmers, collecting data is only the beginning, Kottmeyer notes.
“You can buy satellite imagery from anyone under the sun,” he says. “What you can’t buy is analysis.”
Because of this, he suggests farmers reach out to whatever partners are available to help them connect the dots, whether that’s agronomists, cooperatives or other third parties. In particular, look for those who partner with experienced satellite companies, particularly those who interpret satellite images for the military.
Kottmeyer points out good military-grade satellites cost $100 million or more to manufacture, and they have bigger payloads and come with better sensors for cloud-busting, hyper-spectra imagery and capabilities to pull data to a ground station. Compare this to the lesser capabilities many smaller satellite companies use, he says.
It might also help to think about satellite data analysis like a reverse paint-by-numbers coloring page, Kottmeyer says. Once a satellite expert has woven together this “digital quilt” of numbers, they can start comparing all of the numbers within images of an identical location taken at different points in time.
Did the colors change? If so, it could reflect crop disease, pest pressure or weather-related crop injury, Kottmeyer says.
To garner even more information, he says, if he was a farmer, he’d want to look at three to five years’ worth of images of all of the farms around his operation for comparison analysis of several variables.
“I’d want either my seed company or my agronomist to provide me with the baseline of new seeds, chemicals or whatever input it is,” he says. “How is it supposed to perform? I’m going to take my performance, my neighbor’s performance and the performance of any new input, and I’m going to triangulate between those three. I’m going to get decisions that are 20% to 40% more accurate.”
How Satellites Work
Numerous private companies are hard at work deploying small satellites that orbit the earth and collect information farmers and other businesses will pay to analyze. How exactly do those machines work once they reach outer space? Scott Herman, product development manager for Seattle-based BlackSky Global, explains the basics.
Step 1: Decide where to shoot. Satellites either act as a vacuum, continuously collecting photos and other data from a single location throughout time, or respond to specific tasks sent by mission control back on Earth. “Do you shoot because a customer told you to, or do you collect the world and hope someone wants to buy it?” Herman says.
Step 2: Program the satellite. Commands are uploaded to the spacecraft. In the case of BlackSky’s equipment, the constellation will be commanded to take a photo when one of the satellites passes over the target during its normal orbit of Earth.
Step 3: Download and process. Data is gathered and processed back on Earth. Analysts must know what information is being sought (for example: looking for drought damage or over-irrigated areas)and develop an answer that can be passed on to the customer.
Step 4: Rinse and repeat. A single satellite company can perform this process hundreds of times per day for many different customers, Herman explains. Satellite data will be prolific.
Is Higher Resolution the Smarter Choice?
When collecting images of your field, you have a choice between a satellite grabbing photos at a 5-meter-scale resolution or a drone that can zoom down to centimeter-scale resolution. Which would you pick?
If higher-resolution images sound like a no-brainer, Jesse Vollmar, CEO and co-founder of FarmLogs, says it’s not quite that simple.
“Higher resolution isn’t always better,” he says. “If you’re using drones to scout rotation crops, centimeter-scale resolution isn’t required to spot a portion of a field that might be affected by infestation or diseases.”
Vineyard and orchard owners can make a stronger case for getting the highest resolution possible, Vollmar says. That’s because making a close inspection of perennial crops could be a savvy strategy to prevent spread of disease. Plus, tree and vine crops typically have “far fewer acres” than the average row-crop farm, he adds.