Information Possibly Outdated
The information presented on this page was originally released on February 22, 1999. It may not be outdated, but please search our site for more current information. If you plan to quote or reference this information in a publication, please check with the Extension specialist or author before proceeding.
Simulation Can Help Predict Crop Yields
MISSISSIPPI STATE -- Knowing what the weather will be like is about the only variable keeping Mississippi State University researchers from being able to predict some cotton and soybean yields.
Dr. Harry Hodges, crop physiology and production specialist with the Mississippi Agricultural and Forestry Experiment Station, said computer programs have been developed to simulate crop growth. The goal is to know how plants will respond to environmental variables.
"We have spent a lot of effort to develop these equations and then put them together in combinations that reasonably reflect what plants do in the real world," Hodges said. "Crop responses to weather and soils appear to be unique each year because there are so many variables impacting the plant, but these can be simplified and made understandable and predictable."
Hodges works with the Glycim Commax and Gossym Commax computer programs developed as a collaborative effort between MAFES, the U.S. Department of Agriculture and Clemson University. The programs factor in plant responses to individual variables and combinations of variables.
"Our approach is relatively unique in that we are detecting the environmental variables and the plants' response to them and trying to develop a mechanistic model," Hodges said. "Others have developed relatively simpler models that have been shown to simulate the crop in general, but are not sensitive to unique weather experiences."
If an unusual condition exists, the more general programs will not be able to simulate the unique response to it, Hodges said.
The program works when the producer inputs information on soil types, seed variety, management practices, and daily information on weather, insect pressure and more. The program analyzes the data and provides material with which farmers can make management decisions.
Program developers are still trying to simplifying the model results so farmers can use the information more readily.
"The model is ahead of the ability of the user to understand and use the information to help with decisions," Hodges said.
The cotton crop simulation program, Gossym Commax, has about 300 users in Mississippi and across the Cotton Belt. The soybean counterpart, Glycim Commax, is in more limited use as it has not been released to the public. Hodges said an attempt is underway to distribute this program to producers, but will depend on technical support being available to program users.
Dr. Frank Whisler, MAFES researcher, began field testing Glycim in 1997 with 12 farmers in Mississippi, Louisiana, Alabama, Arkansas, Missouri and Tennessee. He also tested another soybean model, PC Yield, a product of the University of Florida, which is similar to Glycim but does not have as much detail in physiology and soil water movement.
"These farmers had experience using the cotton model, so we put the soybean programs on their computers," Whisler said. "They have access to weather stations in their area, and are using these models for scheduling irrigations."
In winter, farmers use Glycim to evaluate different varieties, row spacing and plant populations on different soils. This information can help them plan for the next growing season.
The 12 farmers given the program in 1997 requested it again in 1998. Evaluations from the 1998 season are not in yet, but Whisler said growers appear satisfied with the results.
While both Glycim and PC Yield can be used for planning irrigation schedules, only Glycim can aid farmers in row spacing and plant population decisions, Whisler said.
Contact: Dr. Harry Hodges, (601) 325-2725 or Dr. Frank Whisler, (601) 325-2311