Chatting about Big Data in agriculture can be a mixed bag at times. There are companies and growers out there that harbor a Cold War era-esque suspicion that some faceless dark force is conspiring to steal and profit off their yearly data stores.
On the other side of the coin, there are many open and progressive groups that have no problem discussing their outlook and general philosophy on managing data. These companies are more than likely “Thinking Data First” in virtually every aspect of their day-to-day, and they made up a large proportion of the folks at this year’s InfoAg, which occurred in late July in St. Louis, MO.
While last year I focused on drone applications and other shiny new technologies such as Israeli firm Phytech’s “plant heartbeat” sensors, this year I wanted to go there with InfoAg attendees, to talk data with them, even if it made them as nervous as a long-tailed feline in a room full of rocking chairs.
Luckily for me (and this magazine), the folks at InfoAg were prepared to address my queries.
Take Randy Kasparbauer, technical product manager, John Deere, for instance. I’d met Kasparbauer back in November at Deere’s API-focused “Develop With Deere” conference in Kansas City, so I was already familiar with some of the work he and his colleagues had been engaged in ensuring Deere-branded equipment exchanges data with other systems. I wanted to get an update on the progress of Deere’s API initiative and I also found time to talk data in general with Kasparbauer, who indulged me as follows:
“I think we’re all in that same fight right now,” he said as we talked in the Deere booth. “Where we work as the API Enablement team is right with those software companies so if you’re working with a trusted advisor and you have a piece of software that uses precision ag data that you’re trying to make insights off of and it doesn’t connect to your MyJohnDeere account, we’re asking those software companies to get in touch with us, to get in touch with the API Enablement team and have us work through the integration and get them on board to leverage the connections that we provide.”
Aaron Hutchinson, president and CEO, iCropTrak, is another prior contact (I met Hutchinson in Phoenix, AZ, this past spring during Commodity Classic) who’s brain I decided to pick on what he feels it means to think data first.
“One thing you need to keep in mind is to not let anything fall on the ground, because what we have learned from our time doing military work and other types of businesses in the past, is that we really don’t know what data we will need from the past in order to help us make decisions,” Hutchinson explained. “A lot of people are using historical GEDs over a season or historical weather patterns over a season, to actually forecast what this year will look like in comparison to past years. So something that’s relatively simple like that, that you can do in Excel, will give you the ability to make some advanced predictions without really having to have a complex set of tools. And I think what’s coming down the pipeline from all of the data vendors here at InfoAg is we’re trying to help you reduce that data into something a little more manageable. So Big Data in a sense will become automatically entered data.”
Craig Houin, data management lead, Sunrise Cooperative (Fremont, OH), was one of the more interesting interviews from the week in St. Louis. Houin is the exact type of person we like to talk with for these types of articles — he’s in the field on a daily basis, working with growers on their data management efforts at the ground level.
“What we’ve done at Sunrise over the last few years since 2012 — especially with yield data — is we’ve been able to make applications for the grower’s P and K fertility program based off of yield data,” explained Houin. “One challenge we have in north central Ohio — especially when we are working in the Western Lake Erie Basin — is monitoring phosphorus and not over applying. So using just the straight yield data is not enough; we have to use soil test information coupled with the actual yield data and then we can develop a multi-input fertility recommendation based on that.
“We’re limited on what we can put phosphorus on; we cannot put any phosphorus in broadcast form on any field with anything over 40 ppm on the soil test reading, so we have to have that maximum level coupled with the yield data to make sure we put the right rate on in the right place. So using good quality yield data and good quality soil test information we are able to do that through the 4R Nutrient Stewardship program that we’ve implemented in Ohio due to the harmful algal blooms in Lake Erie.”
Another new contact I made at the show, WinField young gun Keaton Krueger, agriculture technology specialist, Northeast Iowa, talked about the planning aspect and how important it is to have a set “data game plan” going into the season and to stick with it.
“I’d say the key thing with managing data today is just planning,” recommended Krueger. “What I always say is, when we do the variable rate seeding for growers, the first time is always a challenge and everybody is always nervous. But every monitor that I’ve loaded with a prescription before March, I’ve never had a single problem. But sometimes when we’re in mid-March or early April and we’re rushing to get into the field and get everything planted on time, that time frame tends to be where we see the issues popping up. So planning is definitely a key aspect.
“The other thing is, as an industry we’re on the edge of the data storage systems changing a lot. The connections between everybody’s clouds are increasing. At WinField we have what we refer to as our ‘data silos’ which is basically where we’re going to store all of our information and also give our local cooperatives the opportunity to store a lot of their other information that wouldn’t be in our WinField tools, and then that way we can share to other equipment manufacturers and other programs so growers don’t have to go in manually and enter the same information for the same field in six different programs. It’s kind of like if you think of the way money moves between banks; I can go to Wells-Fargo and say send $100 to U.S. Bank and the next day it is there.”
It could be interesting to look back on this write-up next year during InfoAg 2016 to see how far we’ve progressed (or regressed) on the issues and challenges facing data management efforts in precision agriculture. Stay tuned to the “Think Data First” series in December for a grower-retailer profile on Minnesota grower Brad Hagen’s operation in Southeast Minnesota and his work with local cooperative Central Advantage GS on managing his data.