A New Vision for Variety TestingJune 16, 2020
If you’re like me, you don’t often stop to think about where your food comes from or the effort it takes to get it onto your dinner plate. For many of us, agriculture is a parallel world to the one we live in. It’s a massive industry that goes unnoticed, even if we pass by grain storage facilities on our daily commute.
Three University of Idaho experts are working to convey the importance of this unseen world with a seemingly innocuous project—a database to store crop variety testing information. This database will become a key tool for both growers making planting decisions and researchers who need detailed agricultural information.
Gathering Scattered Data
The project began when University of Idaho Extension Crop Management Specialist and 'Potlatch Joe Anderson Cereal Agronomy Endowed Professor' Juliet Marshall was approached by College of Agricultural & Life Sciences (CALS) statistician Julia Piaskowski. Marshall studies diseases associated with varieties of crops like wheat and barley and directs the Extension Variety Trials for the eastern and southern part of the state. She has helped gather years of small grain variety performance data, which are stored in a disjointed manner online. With her background in agriculture, Piaskowski saw the importance of Marshall’s “goldmine” of variety testing data and asked to help reorganize it. “Our idea was to mine it for information, use it to inform future farming needs, and give it a format so people could explore it in an interactive fashion,” she says. After seeing a presentation by Northwest Knowledge Network, Piaskowski reached out to Research Applications Architect Jennifer Hinds for help creating a database. Marshall says finding a group that understood the value of her research was a rare and phenomenal opportunity.
The variety testing data that scientists like Marshall gather is generally accessible in written format, but is not manipulatable, available to everyone, or stored all in one place. Piaskowski says this information has traditionally been printed off and handed to growers. While that format may be adequate, it doesn’t allow for the fullest use of the data, and it doesn’t work at all for researchers. Even while being deeply engaged in this project, Piaskowski has struggled to locate and match up past data, “And thinking long-term, maintaining the integrity of the data is important because it’s an incredible resource.”
The team’s vision involves gathering the scattered data and making it more available and easier to use. Hinds says, “It’s an attempt to aggregate a rich collection of information, standardize it, and allow it to be dissectible depending on the user’s need.” Improved storage methods will result in better preserving data integrity, and graphical tools and interfaces will allow users to easily navigate the data. Programs similar to Marshall’s are being run statewide as part of the Idaho Agricultural Experiment Station in locations like Moscow (part of the Palouse Research, Extension and Education Center) and at the Parma Research and Extension Center by cropping systems agronomists Kurtis Schroeder and Olga Walsh, respectively. Piaskowski says, “I want to see all those results collated together and organized so researchers can access them, and growers can make comparisons and have a more extensive understanding of what’s going on with their crops.”
On the Foundation of Wheat
In its current state, you can search the database by year, location, crop, or variety. About 7,000 records have been imported from Marshall’s Southern Idaho Variety Testing Program—a limited sample—and more will be added as the database is developed. Though still in the initial stages, there are clear opportunities to integrate features like interactive maps and charts. This site will be unique because it is designed to display all the data from a trial instead of a summarized version. Piaskowski says, “No other variety testing program is maintaining this data in a public fashion. So, it’s very unique and extremely useful.”
The two main audiences of this database are agricultural researchers and growers. Their needs vary greatly, meaning that adaptability of the site is crucial. Piaskowski says that for growers, “Having summary functions and graphical user interfaces is extremely important. But for the other end user—researchers like myself—we have enabled the ability to search and download data.” There has already been a positive response from a researcher using the test site, which the team is excited by.
Available data is currently limited to wheat in Southern Idaho. The team chose to build the database on the foundation of wheat data because of its importance to the area. The team has high hopes for expansion, though it could be slow. They have already faced many roadblocks while constructing the database. Eventually, they would like to combine data from Idaho, Washington, and Oregon into one place—because of their similar ecosystems, comparing their data would be beneficial for both end user groups.
Getting People to Listen
Marshall, Piaskowski, and Hinds have been met with an array of challenges. Preparing data for the site is laborious and requires a substantial amount of knowledge. Maintaining communication and collaboration between disciplines is essential. Finding a common denominator for the data is key. Hinds says, “It’s a struggle to design a structure that can be robust but forces everyone to play into it in a consistent fashion. You need to have constant communication so you can design a database that’s going to be the best for users.” But through all these challenges, the hardest part might be just getting people to listen.
“Trying to present a vision to people that really don’t understand the potential long-term value has been difficult,” says Marshall. “The three of us on one side can really see the long-term application, and yet trying to convey the importance of the data to get funding has been unsuccessful.” Through all the doubt, the team remains determined though. Acknowledging there may be more iterations before they are successful, Piaskowski says they have a solid foundation. “What we have is pretty incredible and a great proof of concept. Once we have a web app, that will be a great step to help people see the value.”
The practical applications of the database go beyond simply storing information. Hinds says one of the primary benefits of the system is its ability to aid growers through decision support. Marshall calls it a risk management tool. “This database can help growers choose the best variety for their growing conditions.” Piaskowski says that growers want as much crop information as possible to try and avoid the cycles of catastrophic agronomic issues that occur. From an analytical perspective, the team also has goals to routinely integrate spatial statistics into variety testing, and to re-analyze past data using newer methods. The versatile nature of the database means that a wider audience will be able to adapt it to their individual needs.
Working From Both Sides
The immediate next steps for this project involve building a graphic web interface and hiring a team member to help speed up the flow of data onto the site. Piaskowski says that the most challenging part of building a web app is not actually the building but deciding what to build. With so many options available, choosing a suitable format is key. She acknowledges that “It will be an iterative process to produce something that’s useful. We need to give information and then get feedback.”
Eventually, the team hopes to expand the database to cover more area and crops. And as an extension of this project, they have plans to streamline data into the database to further enhance its features. Marshall says, “Once we understand how this data structure functions, then we can start building tools that funnel the data, so it more easily fits into how we want it organized.” Hinds believes working from both sides of the database is central to maximizing its abilities. “It’s one thing to have data and pull from it, but the idea of being on the other side and pushing data into it is even better.”
Speaking about the importance of this project, Piaskowski says, “It sometimes keeps me up at night, thinking about all of the variety testing data in the United States and what extremely valuable and reliable information it is, worrying that it might be lost. It’s important to capture that information because it has so much utility now and in the future.”
The variety testing database is currently supported with funds from Marshall’s endowed position which was established by the Idaho Wheat Commission on behalf of all Idaho wheat growers.
Article by Katy Riendeau
IBEST Design & Marketing Coordinator