Submitted to: Aflatoxin Elimination Workshop Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 10/26/2005
Publication Date: 10/24/2005
Citation: Dowd, P.F. 2005. Update on validation and distribution of a computer program for predicting mycotoxins in Midwest corn [abstract]. Proceedings of the Aflatoxin Elimination Workshop. p. 8. Interpretive Summary:
Technical Abstract: Since the last report, the predictive computer program has been further validated in 2003 and 2004. A correlation coefficient of 0.85, after one outlier was removed, was obtained for commercial field samples from 2000-2003 (which included a few dozen different fields and hybrids). Samples were predicted to have low levels of fumonsin in 2004 (less than 1 ppm) and this is what occurred in all fields sampled. No aflatoxin was predicted to occur in either 2003 or 2004, and none was found. In 2005, the program predicted a high probability that Aspergillus flavus inoculum would be present at silking, which was communicated to farmers present at a field day in late June. Continued predictions indicated low levels (from 30 to less than 20 ppb), depending on weather conditions at different locations. Insect levels were very low throughout most of the season. Ears in dry areas were monitored by both this researcher and representatives of a popcorn company and brought in to elevators for aflatoxin determinations by farmers. Kernels with BGYF, and in one case, visible A. flavus colonization, were encountered. A meeting with elevator operators indicated some samples were testing positive for aflatoxin at 20 ppb from several different areas, and scattered rejected loads occurred through late September. Samples were taken in fields under study, and results of analyses are pending. As indicated previously, cooperation between USDA and Illinois Central College resulted in a Windows version with an added help section, and a "custom" module that allows one to customize if outlier hybrids are encountered, provided prior year's data including mycotoxin levels are available. A "beta" version has been demonstrated via network conferences to different companies (contact Kate O'Hara, Technology information Officer, to arrange a demonstration: email@example.com). An additional economic decision-making module is partly written. Additional plans are to validate the program for popcorn, which is widely grown in Central Illinois (samples were taken in 2005) and other food grade corn as information becomes available. A finalized version of the initial program may be put on a website within the next year or two, but commercial interest may alter this schedule. Overall, hybrids and cultural management presently used by farmers, coupled with more recently identified factors, such as use of Bt versions of preferred hybrids, early planting to escape caterpillar damage in milk stage, scouting for damaging insects including use of lures and traps identified in project research, fungal monitoring using leaf axil material, and the predictive computer program indicating when mycotoxin-producing fungi occur, or specific mycotoxin levels that may occur without intervention by the farmer), could be rationally combined into a management plan.