Published on 06/02/04

Computer software program grows cyber crops

By Sharon Omahen
University of Georgia

A team of scientists has created a computer program that can model an entire crop cycle, from planting to harvesting, in just seconds.

The software is called Decision Support System for Agrotechnology Transfer, or DSSAT. It was created by a team of researchers from the universities of Georgia, Florida, Hawaii, Guelph and Iowa State and the International Center for Soil Fertility and Agricultural Development.

Planting cyber fields

DSSAT allows the user to simulate a crop's growth, yield, water and nutrient requirements and the environment's impact on agricultural production.

The program wasn't developed overnight. In fact, the software's fourth version was released earlier this year. About 50 researchers and graduate students from across the globe met on the UGA campus in Griffin, Ga., May 17-26 to try out the latest DSSAT software.

"This software program is by no means meant to be a substitute for actual experimentation," said Gerrit Hoogenboom, a DSSAT developer and an agricultural engineer with the UGA College of Agricultural and Environmental Sciences. "Experimental data is still needed to establish credibility for models like DSSAT."

Not a research replacement

Hoogenboom also said crop modeling software, like DSSAT, is not a substitute for critical thought.

"The results you obtain from the software are not ultimate truths, and they're not meant to replace real experiments, real data or critical thinking," he said. "Anytime you use a computer model you should question the results."

Though not a substitute for the real thing, the computer model can have great value to researchers, educators, extension agents and consultants.

"Computer models can provide an easy and very fast comparison of many different crop management scenarios and the interaction with local weather and soil conditions," Hoogenboom said.

Sharing information with farmers

DSSAT simulates the growth of crops like peanuts, sunflowers, sugarcane, wheat, soybeans, rice, tomatoes, sorghum, millet, barley, potatoes, corn, blackeyed-peas and dry beans. The next version of DSSAT will be of particular interest to researchers in the southeastern U.S. as cotton will be added to the simulated crop list. This version is expected to be released in two years.

The crop-simulation information gained through the DSSAT software will be shared with farmers.

"Our goal is to educate the people who talk to farmers directly," said Ken Boote, a DSSAT developer and University of Florida agronomist. "Consultants, ag industry representatives and extension agents have the potential to spread the word to farmers. Those farmers with interest in this technology would also benefit from actually using the software themselves."

Boote says the way the software presents the data is an essential part of the success of DSSAT.

"You can't give numbers that no one can understand," he said. "Our program calculates crop growth and development in a mathematical sense and then presents it through graphics."

DSSAT has also been used as an effective tool after a crop has been harvested to identify the source of production management problems.

"It's a way to see the whole picture and what is limiting the crop," Boote said. "The software actually works better this way."

In the early stages, the software was tested using several years of real-crop data from Florida and Georgia farms.

Applications continue to grow

DSSAT has been used on food security projects in Africa and other developing countries, too, and to study the impact climate change has on food production.

"It's been used in Arkansas to help with early-season soybean plantings, in Kentucky for determining planting dates, in Georgia for predicting agricultural water usage and in Africa to diagnose yield loss of peanut crops from disease," Boote said. "The list of applications is never-ending."

Two UGA agricultural economics students are using the program to evaluate crop insurance. They hope to show the actual risks of failure that farmers face. Two University of Florida students are using the software to predict the amount of hay a farmer's field will produce when planted with bahia or bermuda grass.

There are currently more than 1,500 registered users from more than 90 countries using the software.

"DSSAT users share their work and their data via a computer listserver and a Web site," Hoogenboom said. "In this way, the software contributes to the whole scientific community."

Sharon Omahen is a news editor with the University of Georgia College of Agricultural and Environmental Sciences.