Submitted to: Agricultural Systems
Publication Type: Peer reviewed journal
Publication Acceptance Date: 2/15/2002
Publication Date: 6/20/2002
Citation: Timlin, D.J., Pachepsky, L., Whisler, F.D., Reddy, V. 2002. Experience with on-farm applications of glycicim/guics. Agricultural Systems. p.55-69. Interpretive Summary: A Crop simulation model is the representation of the growth and development of an agricultural plant as a function of environmental inputs, and runs on a desktop computer. An interface is a program that helps a user input data to the simulation model and view simulation results. Crop simulation models can be used on-farm to help growers make management decisions. Research is needed to develop improved crop simulation models and user-interfaces and understand how crop models are used on-farm. The purpose of this project was to field test the model GLYCIM and the interface GUICS (Graphical User Interface for Crop Models) with soybean growers. Growers in the Mississippi Valley have been using the soybean simulation model, GLYCIM over the past ten years. The growers use GLYCIM to choose soybean varieties appropriate for specific soil conditions and expected weather patterns, and schedule irrigation and harvest operations. This study found that grower input during development of the interface GUICS greatly influenced the final design and usability of the interface. The use of GLYCIM/GUICS has helped the growers know their crops better as they were more likely to visit their fields to compare actual plant growth with simulations. The results of this research benefit the agricultural community by developing more accurate simulation models in improving the usability of these models. The soybean growers using GLYCIM/GUICS attributed an increase in soybean yields of up to 29 percent and irrigation use efficiency of up to 400 percent to the use of GLYCIM.
Technical Abstract: GLYCIM/GUICS is a Decision Support System (DSS) used "on-farm" by growers in the Mississippi Valley to provide decision support information for farm operations. This DSS contains a mechanistic soybean simulation model (GLYCIM) that is linked to a graphical user interface, and rule based interpreter of the simulation output (GUICS). The DSS provides warnings of the need to provide irrigation or to harvest, and outputs a summary of the consequences of insufficient or no irrigation. The purpose of this project was to evaluate GLYCIM/GUICS with soybean growers in terms of simulation results and usability of the interface. The use of the model to plan specific operations evolved as the growers ran the model, became familiar with its operation, and requested enhancements. Many growers used the model to choose soybean varieties appropriate for specific soil conditions and expected weather patterns. The interface, GUICS, was developed to provide for better management of scenarios and provide a more intuitive point and click front end. GUICS was designed and modified using input from the growers during development. Growers were less interested in accuracy of the model than in how well the model predicted differences relative to a specific baseline. The growers have also reported that use of the model has helped them know their crop better as they are more likely to visit their fields to check actual plant growth stages. In a survey by Mississippi State University, the soybean growers using GLYCIM/GUICS attributed an increase in soybean yields of up to 29 percent and irrigation use efficiency of up to 400 percent to the use of GLYCIM.