Many practices are associated with regenerative agriculture — anything from no-till practices to pesticide-free farming. What’s more, the concept means different things for different crops in different regions. What is considered regenerative in one location might not qualify for the same label under other agricultural conditions.
It’s clear the food and agriculture sector needs to start defining regenerative agriculture specifically and measuring it quantitatively — it’s essential for the concept to scale. Some practitioners and regenerative ag pioneers are piloting new technologies to help with that process. These new tools — under development or in the early phases of testing — are helping put numbers to the abstract concept of regenerative agriculture and helping measure metrics such as biodiversity, carbon sequestration and other soil health considerations.
Following is a list of five emerging options, focused on two primary concerns, measuring biodiversity on agricultural land and gauging soil health and carbon levels.
1. Quantifying insect diversity using AI
Ecdysis is building an insect database that will use artificial intelligence to identify insect species and extrapolate the population of each species on a farm. The nonprofit, based in South Dakota, got off the ground with a crowdfunding campaign and now pays the bills with a combination of donations, foundation money and competitive and corporate grants. The 11 team members, four of which have Ph.D.s, are working with General Mills’s regenerative agriculture pilot to build up and verify its library of insects.
"Insect diversity actually scales really well and is a good indicator of profits, of soil carbon, of soil, of water," said Jonathan Lundgren, director of Ecdysis. "We can use them as bioindicators because insects are a great responder, they’re so sensitive to what’s going on in a habitat. Just by taking the snapshot, you can tell an awful lot about the health of that environment."
Ecdysis asks farmers or people involved with regenerative agriculture projects to use a butterfly net and take 50 sweeps of the air near their wheat or oat fields to collect a sample of insects to send to Ecdysis as live samples or as photos. Then Ecdysis uses those photos to train its artificial intelligence system to identify the insects and model out the population. It then compares the insect population to other regenerative agriculture indicators taken on the farm, such as carbon soil levels and pest outbreaks.
Once the image recognition database is ready next year, Ecdysis says it will be ready to start predicting all the factors that affect a farm.
2. Measuring soil carbon levels with a handheld probe
Chris Tolles, CEO of Yard Stick, is working with Christine Morgan, Ph.D., a chief scientific officer at the Soil Health Institute, to create a handheld soil probe that measures carbon levels with VisNIR spectrometry and pressure sensors. The original probe was so large it had to be mounted on the back of a truck. Tolles’ company, Yard Stick, is miniaturizing the technology so it can be used with a simple handheld drill and creating a commercial business to support the product.
The tip of the probe is a small camera that uses wavelengths to sense the presence of organic carbon the way our eyes sense the presence of blue when looking at the sky. The resistance sensors on the probe calculated the density of the soil. With those two inputs, Yard Stick says it can calculate the amount of carbon sequestered in a particular area of soil.
These new tools are helping put numbers to the abstract concept of regenerative agriculture.
Right now, the company is working on verifying the accuracy of the soil probe by comparing its data to the traditional method of measuring carbon in soil, dry combustion. Using the latter technique, a sample of soil is burned to indicate the amount of carbon stored within it.
"That’s not scalable," Tolles said. "The incineration of things at thousands of degrees, people trudging through fields, scooping up soil and putting in the mail. [Yard Stick] can take samples way faster, we can take more samples per field, the cost is dramatically lower, there are no consumables, we’re not shipping anything so you’ll get a more accurate measurement of your carbon stock."
Yard Stick, based near Boston, plans to have a commercial product ready for sale by 2022 and is partnered with large industrial food companies to connect the probe to U.S. farmers for testing.
3. Mapping soil health with satellite data and remote sensors
Using satellite data publicly available from National Aeronautics and Space Administration and the European Space Agency, along with specialized algorithms, Dagan is monitoring the landscape to map adoption of conservation practices and soil health management.
Its platform can monitor tillage practice, cover crop planting and rotational crop-growing methods, and track soil residue dynamics — how the materials left on the surface decay. A model created at the University of New Hampshire predicts how those agriculture management practices map to greenhouse gas emissions and nutrient cycling.
Dagan, a startup based in New Hampshire, was spun off from Applied Geosolutions and supplies the agriculture sector with data services to recommend regenerative agriculture practices and track the results. It was able to create a system that can calculate the emissions and soil health without on-the-ground baseline reference data usually needed for projects such as these. And it is working on a way to calculate biomass.
"We can create not only maps where cover crops have been adopted, but actually information on cover crop performance," said William Salas, interim CEO of Dagan. "How well the cover crop establishes will influence the nutrient loss, sediment and nutrient uptake, as well as the amount of biomass to cover crops achieves because that is organic matter going back into the soil."
Dagan is working with The Nature Conservancy and the Ecosystem Service Market, whose members include food partners such as Mars and Nestle.
4. Evaluating bird diversity with microphones
Imagine standing in a field 24 hours a day, 365 days a year, pressing record every hour for a five-minute audio sample of the bird noise. In some ways, that’s the dream job of John Quinn, associate professor of biology at Furman University. But Wildlife Acoustics, the 18-year-old Massachusetts-based company, created the next best thing: a programmable, weatherproof recorder that does just that.
"Instead of me having to drive 9,000 miles to visit all these sites, multiple times, I can put the recorder out once and then program it to record and really quickly, we have scaled up to massive amounts of acoustic data that we can then go back and analyze," Quinn said.
Quinn is working with General Mills to categorize and identify each bird in the thousands of hours of recordings taken by WildLife Acoustics microphones. The goal is to compare the bird populations on farms practicing regenerative agriculture to control groups to see if there is a statistically significant difference in bird populations.
Last summer, he had recorders on 30 farms in Kansas and is working on analyzing the sounds. Databases of bird sounds such as Kaleidoscope, also created by Wildlife Acoustics, and BirdNet can group or identify bird sounds, but the songs are so complicated that it still takes a trained ornithologist to make the final call.
"The regional dialects that different birds have is so diverse," Quinn said. "A Carolina wren down here in South Carolina, that might sound different than one out in Kansas."
The hope is to one day be able to mail these recorders to farmers, have them program them and stick them in fields all over the world to get a clearer picture of bird populations and changes as regenerative practices are adopted. That’s something that would be cost-prohibitive before the invention of this easy-to-operate technology, Quinn said.
5. Identifying insects using lasers
Denmark-based FaunaPhotonics creates a sensor that gives farmers real-time information about the type, number and activity of insects flitting between their crops. The company, spearheaded by two Ph.D.s., hired a business expert to bring the sensor to market. The sensor uses LEDs and photodiodes to see and interpret the wing flutter patterns of insects that fly past the sensor. The machine-learning algorithm uses the wing flutter to identify the insect and create a report for the farmer.
"The sensor is like a one-stop-shop," said Kevin James Knagg, commercial director of FaunaPhotonics. "You can see how many bees, the number of different types of bees, this many moths, or break that down to say, you’ve had 1,628 honey bees been active past this center in the last four hours. Or if you're looking to see how you can generate more bee activity, or maybe you’re looking at biodiversity as a whole and want to see all the insects."