WebGro: A Web-Based Soybean Management Decision Support System

than variety selection. For example, numerous studies have been published from the upper midwestern United WebGro is a web-based soybean [Glycine max (L.) Merr.] decision States on the effect of crop rotation, row spacing, and support system built on the CROPGRO-Soybean model. The purpose of WebGro is to help soybean producers in the midwestern United plant population on soybean yield (Pedersen and Lauer, States understand how different stresses interact to limit soybean 2002, 2003a, 2003b). yield in their fields. Stresses include water, soybean cyst nematode In the midwestern United States, water stress is often (Heterodera glycines), herbicide injury, Rhizoctonia root rot disease viewed as the biggest yield-limiting factor. However, (Rhizoctonia solani), and hail damage. The user can set up a field other factors such as diseases, insects, and hail damage scenario by selecting cultivar, planting date, plant population, soil can also cause significant yield loss. Although herbicide type, and the nearest weather station using a web form. Different stress injury has been shown to not significantly affect yield, levels can then be entered, and the model can be run interactively by it can contribute to yield loss if other stresses are present simulating the effects of one or more stresses at a time. A summary (Browde et al., 1994). Producers can use management table of model simulations is generated and presented onscreen to alter the effect of interactive stresses. Variety selecallowing users to examine individual and combined effects of plant stresses. The results are stored on the server, thus allowing the users tion can be used to reduce the impacts of diseases while to view these results in another session. WebGro also offers interactive irrigation can reduce the effects of water stress. Howgraphical output that lets users view daily changes in soybean biomass ever, it is often difficult to understand these complex weight for a specific field or scenario. WebGro is available at http:// interactions and to determine how much each factor webgro.ae.iastate.edu. may affect soybean yield because of the magnitude of compensatory growth and alterations in plant development (Pedersen and Lauer, 2004). S yield is the result of complex interactions Crop models such as CROPGRO (Hoogenboom et among genetics, environment, and management deal., 1994) have been used as a research tool to study cisions and is correlated with the number of seeds and different interactions and to estimate the effects of difseed size (Salado-Navarro et al., 1986). Sinclair (2004) ferent factors on soybean yield (Batchelor et al., 1993; theorized that the maximum soybean grain yield was Paz et al., 2001; Pinnschmidt et al., 1995). CROPGRO is probably no greater than 7.3 to 8.0 Mg ha 1. This genetic a process-oriented soybean model that simulates growth yield potential is obtained only when environmental and development on a daily basis using C, N, and water conditions are perfect, but such conditions rarely exist. balance principles. The model computes daily photosynIn a field situation, soil and climate provide the major thesis based on temperature, water stress, and light inportion of the environmental influence on soybean deterception. Daily C is then partitioned to leaf, stem, root, velopment and yield (Pedersen and Lauer, 2003b). The pod, and seed pools based on the crop growth stage combined environmental constraints such as temperaand C supply–demand principles. Development rate is ture, solar radiation, and available water define rigorous simulated each day based on temperature and daylength. limits to crop biomass accumulation and consequently Soil water content is computed in different soil layers grain yield. each day using a water balance approach, and water A specific variety may vary in grain yield from year stress is computed based on the limitation of potential to year, even within the same field, indicating that some root water uptake and evaporative demand. Batchelor varieties are better adapted to a specific environment et al. (1993) added modules to incorporate the effect of than others. It is not unusual for one variety to outyield many different types of pest damages on different state another variety by 1 to 1.5 Mg ha 1 or more in the same and rate variables in the model. field. Producers will therefore select varieties that are Researchers have used crop models for many years yield stable within a region across locations and years but have had limited success in packaging these complex to get more accurate information on specific variety models in a framework that makes them easy for properformance and stability. Producers can also help to ducers to use. Intensive input requirements and the optimize yield by proven managerial decisions other problem of data entry and selection have prompted researchers to develop decision support systems (DSS) J.O. Paz and W.D. Batchelor, Dep. of Agric. and Biosyst. Eng., Davidand smarter, more intuitive user interfaces to different son Hall, and P. Pedersen, Dep. of Agron., Agronomy Hall, Iowa crop growth models. Decision support systems provide State Univ., Ames, IA 50010. Received 25 Mar. 2004. *Corresponding author ([email protected]). Abbreviations: API, application programming interface; ASP, active server pages; DSS, decision support system(s); GUI, graphical user Published in Agron. J. 96:1771–1779 (2004). © American Society of Agronomy interface; ODBC, open database connectivity; SCN, soybean cyst nematode; VBscript, Visual Basic script. 677 S. Segoe Rd., Madison, WI 53711 USA