With over 90 stacked, single trait and conventional cotton varieties available today, plus 80 or so experimental lines in waiting and a bunch of others discontinued every few years, it's no wonder cotton producers get edgy each spring when they pore over listings from university variety trials.
Selection anxiety was the driving force behind a Cotton Incorporated and University of Arkansas project in which producers can use the Web to summarize tons of varietal data at the click of a button.
Called COTVAR, the program is available on the Web at http://cotvar.uaex.edu.
The project was developed by Fred Bourland, director of the Northeast Research and Extension Center, Keiser, Ark., at the request of northeast Arkansas cotton producer David Wildy, “who asked us to put a program together to summarize these data.”
The idea sat on the back burner for a while, and it wasn't until several Mid-South states started working on a parallel project to determine yield stability across states, that data started coming together. “This provided the impetus for starting the project.”
Two programmers, Chalmers Davis and Becky Bridges with the Arkansas Cooperative Extension Service, Little Rock, designed the program after meeting with Bourland and Don Jones from Cotton Incorporated. “They did a super job,” Bourland said.
COTVAR is not intended to replace annual state variety trial publications, Bourland notes. Rather, it's a way to summarize the data without having to sort through page after page of trial results. Another advantage is that the program summarizes data across states. “We believe that the best observations of a variety is how the variety performs in a general area, not just at one test site in one state.”
COTVAR combines information from university trials in Louisiana, Mississippi, Missouri, Arkansas and Tennessee.
COTVAR is updated annually to provide the latest information, noted Bourland. “For example, each year, a list of varieties is sent to seed companies, that respond for each variety in terms of its availability and other characteristics.”
Keeping up with variety name changes in not a problem either. “With COTVAR, we draw the data from the experimental name, too. The producer may not have that information.”
To start the program, select the year you want to look at, then determine the status of the varieties you want to compare — commercially available, experimental lines or obsolete varieties. Then select conventional, multiple trait lines or single trait lines.
COTVAR will then ask you to select locations. “You can choose all locations, or cherry pick locations by states,” Bourland said. “Those are arranged north to south by soil type and whether or not they're irrigated.
“We encourage them to pick as many sites as they are comfortable with. A screen will show the varieties tested at those locations and the frequency. Then they can select up to five varieties to compare at a time.”
One output screen will show lint yield and fiber quality rank over all locations for the selected varieties. A second output screen will list how each of the varieties performed at each of the locations relative to the mean (or average) at that location. A third output screen lists other characteristics such as plant height and seed information across all locations.
Another feature of the program is the quality index, which weighs four quality measurements common to all varieties — length, length uniformity, micronaire and strength. The score was suggested by Arkansas cotton producer, and Cotton Board vice chairman Bob McGinnis, who wanted a way to evaluate varieties compared to the needs of the export market for U.S. cotton.
Weight was added to quality values based on what Cotton Incorporated suggested was needed for the export market. The score ranges from zero to 100, with a higher score reflecting better quality. “I tell our producers that once they find a few varieties that are their best yielders, then go in and choose the one that has the best quality score.”
The score does not necessarily establish an exact threshold for the export market, “but you don't have to worry about mike or length or all these different things you have to balance,” Bourland said. “You just look a the score. We're still tweaking and testing, but we feel pretty comfortable with it. I'm already using it in my breeding program. I can sort by quality score. It really helps me.”
Bourland says that U.S. varieties that meet current minimum loan requirements might score around 50. “Some of our high quality varieties will get up around 75. That doesn't mean that a variety that scores a 50 will get discounted in loan value. But during tough years, they'll be more vulnerable and simply may not meet our long-term market needs.
“Conversely, you should be in a premium range with a 75. Our market doesn't work too well on premiums, but the important thing is that we need to improve quality so we can still participate in the market.”
Bourland suggest that to get the most from COTVAR, “incorporate as many locations as possible to get the best indication of which variety will do well this year, rather than the best variety closest to you in last year's trials.
COTVAR should also identify over time the varieties that are broadly adapted, that do well regardless of soil type and irrigation.
Cotton Incorporated plans to continue its support of the project in hopes of making it a national project. “We are starting to put data files together,” Bourland said. “One of the big hurdles we have is defining different growing regions. But we're going down that road.”