Precision agriculture: From real-time farming data to meta analysis
Susan Kuchinskas looks at how analysis and merging of data from farming equipment and other sources could improve the precision of agricultural practices – and maybe end world hunger.
John Reifsteck is a new kind of farmer. Unlike the stereotypical guy scanning the skies in search of clouds and rain, the corn and soybean grower from Illinois knows exactly where that next thunderstorm is in relation to his tractor and those of his crew.
"One of the worst things that can happen is to be planting and then have a heavy thunderstorm that packs the soil,” he says. “You can decide, I'd rather not plant and save the seed.”
For Reifsteck, things began to change in the 1990s with a simple mobile phone, which let him stay in contact via voice and e-mail while he was out in the fields. And things took off from there.
Today, smartphones and tablets in his pickup add up to a virtual office that lets Reifsteck do way more than check real-time weather maps.
He uses Evernote to keep track of documents and transactions with suppliers. He puts spreadsheets to use to track his inventory of grain. And he keeps track of real-time data from commodity markets so that he can make a quick sale if he sees soybean prices spiking – no matter where he is.
"The mobile phone was probably the most revolutionary device since the modern plow," Reifsteck says. "People in farming are pretty isolated and not connected except by a landline that might be hours away. The cell phone changed the way we farmed."
Fields of data
Using connected technologies to gather farming data from the field has become much more prevalent in recent years.
Most modern farm equipment is today enabled with sensors and monitors that gather data as they move across the fields, logging things like how many bushels of corn are being harvested from an acre.
And the number of progressive farmers like Reifsteck is also growing. But most still need to manually remove data from the machine via an SD card or USB stick and carry it back to the office, to be uploaded into a proprietary software system.
But mobile is beginning to change that and with it what data is becoming available and how the data is being used.
Connecting John Deere
In 2011, John Deere began shipping tractors with embedded cellular modems and introduced JDLink, a telematics product which gives remote access to fleet location, fleet utilization and a range of diagnostic data.
In 2012, the company went a step further by introducing MyJohnDeere, an online information platform letting farmers access, view, archive, manage and share a wide variety of business information, including equipment information, production data and farm operations.
With MyJohnDeere supported by Wireless Data Transfer, which was added last August, customers are now able to view and manage this information from smartphones, tablets and computers when and where they need it.
Originally, the business case for wireless in the farm machinery was to provide customers with basic machine-monitoring services while giving the company a means to provide remote software updates and analyze machines' performance, says Charles J. Schleusner, product line marketing manager, intelligent solutions group, Deere & Company.
Now it also allows farmers to make better decisions faster. "Now they can get a daily feedback loop and make adjustments in close to real time,” Schleusner says.
Finally, late last year, John Deere announced it was beginning to collaborate with other businesses to deliver new applications and services that help customers more quickly turn the gathered data into management decisions in areas like improved productivity, efficiency and yield management.
"Our goal is to provide a preferred, safe and secure environment for these applications, while building valuable features to connect customers with trusted advisors," said Cory Reed, senior vice president, intelligent solutions group, Deere & Company, in a Nov. 14 press release. "Customers will be able to make choices about the use and flow of the data."
A recent integrators' conference drew close to 100 companies looking to build on the MyJohnDeere platform.
Harvesting the data
Monsanto's new FieldScripts service is another example of how much more can be done with farming data.
In essence, FieldScripts are semi-automatically generated planting "prescriptions" that are transmitted wirelessly to a tractor running a planter, which makes it possible to automatically plant the optimum quantity of seed in different parts of the field.
Here is how the service works:
The farmer provides a specially trained Monsanto seed dealer with yield data, usually via a thumb drive, as well as soil reports.
The seed dealer uses his or her expertise to add value to the discussion; for example, if she knows last year was unusually dry, she may suggest that they use yield data from a previous year.
And he then transmits this information to Monsanto, where the company applies algorithms that include Monsanto’s understanding of how its seeds perform.
Finally, Monsanto generates a customized FieldScript to the farmer's tractor, which has been retrofitted with Monsanto’s 20/20 RowFlow, which controls the flow of seed out of the planter; 20/20 SeedSense, a monitor that sits in the tractor cab that runs the planter; and FieldView Plus, a Cloud-based service with an iPad app that creates integrated and shareable maps of planting and harvests.
The iPad in the cab acts as the switching station between data in the Cloud and the mechanical operations of planting or harvesting equipment.
FieldScripts for corn became commercially available in January 2014 in Illinois, Iowa, Indiana and Minnesota, with rollouts for the rest of the United States expected later this year. Initial field tests last year with 150 farmers found that they averaged five bushels per acre more using the service.
One Big Data silo
FieldScripts illustrates the benefits of combining multiple data sources when it comes to improving yields. But researchers at Purdue University would like to see all the farming data that is now becoming available used even more broadly.
Sylvie Brouder, a Purdue professor of agronomy, would like to see academia get together with industry, farmers and the government to use farming data not only to improve productivity but also to improve the environment.
The goal, she says, is "making data streams understandable and interoperable." The problem is, she notes, that even among academic researchers, data is inconsistent and incompatible.
"You have lots of different types of researchers, and some have been at it for decades,” she says. “We don't have a formal culture of how we collect and store data, and we have never shared it. The product of communication has been the research paper."
Reifsteck agrees that more broadly shared data could be valuable. If every farmer would share information on the performance of a particular variety of seed corn, he says, "We could look at how it yielded and make predictions for the future."
Feeding the world
Turning mounds of data into useful, widely accessible information could have another huge payoff: agriculture that can, through the combination of Big Data and small efficiencies on a large number of fields, do a lot to counteract climate change – and better feed the world, according Brouder.
Rabobank, a global financial services institution that offers agricultural lending, said in a December 2013 report that, if farmers could optimize the use of fertilizer and irrigation by using data, they could help conserve the world's water supply while increasing yields sustainably.
Reifsteck agrees. "If you can understand how changes in soil affect farming practices, you'll make a big impact on agriculture,” he says. “I'm a lot more optimistic because I think our ability to adapt with information and genetics gives us a lot of potential."
Susan Kuchinskas is a regular contributor to TU.
For all the latest telematics trends, check out Telematics for Fleet Management Europe 2014 on March 12-13 in Amsterdam, The Netherlands, Content and Apps for Automotive Europe 2014 on April 8-9 in Munich, Germany, Insurance Telematics Europe 2014 on May 6-7 in London, Telematics India and South Asia 2014 on May 28-29 in Bangalore, India, Insurance Telematics Canada 2014 on May 28-29 in Toronto, Telematics Detroit 2014 on June 4-5 in Novi, Michigan, Advanced Automotive Safety USA 2014 on July 8-9 in Novi, Michigan, and Telematics Munich 2014 on Nov. 10-11 in Munich, Germany.
For exclusive telematics business analysis and insight, check out TU’s reports: Telematics Connectivity Strategies Report 2013, The Automotive HMI Report 2013, Insurance Telematics Report 2013 and Fleet & Asset Management Report 2012.