In

Over the course of 5 years, different maize residue treatments were conducted on 14 zero tillage on-farm sites in Switzerland to evaluate their effect on the development of Fusarium head blight (FHB) and the contamination with the mycotoxin deoxynivalenol (DON) in winter wheat grains and wheat straw following grain maize. Two experimental series with three and five different treatments were carried out, respectively. Fusarium graminearum (Schwabe) was the predominant FHB-causing species with an overall incidence of 15% infected wheat grains. A significant correlation between symptoms in the field, F. graminearum incidence and DON content in wheat grains and wheat straw was observed. The average DON content in both wheat grains and wheat straw was approximately 5,000 μg/kg and thus several times higher than the European maximum limit of 1,250 μg/kg for unprocessed small-grain cereals for human consumption. Of all grain samples, 74% were above the maximum limit. Pooled over both experimental series, the average reduction of DON in grains through treatments of the maize residue compared with a control treatment ranged between 21 and 38%. The effect of various other factors, including the year, the wheat variety, the site, the maize hybrid and the production system was evaluated as well. The year and the wheat variety were the most important FHB influencing factors. Over all treatments, the variety Levis showed a fivefold higher average DON content compared with the variety Titlis. From different categories of maize residue particles, intact pieces of 5-15 cm length were strongly correlated with F. graminearum incidence and DON content in grains. During the time course of this study, the recommendation from a preliminary version of the internet-based DON forecasting system FusaProg to apply or to omit a fungicide treatment was correct in 32 out of 42 cases. The results are currently being used to optimise the FusaProg models. This study has shown that in a grain maize/winter wheat rotation, the DON content in wheat grains frequently exceeded the European maximum limit, even with a thorough treatment of maize residues and less susceptible wheat varieties. Hence, in order to reduce the contamination risk in a zero tillage system, the crop rotation needs to be modified.

Fusarium head blight is one of the most serious cereal diseases of the world. Epidemics of Fusarium head blight can lead to a decline in grain quality and yield. In addition, grains often become contaminated with mycotoxins, which are harmful to humans and animals. In a field survey of winter wheat in Switzerland, Fusarium graminearum proved to be the most prevalent species responsible for head blight and deoxynivalenol the most common mycotoxin. To elucidate and quantify single or combined effects of cropping factors on F. graminearum infestation and to reduce the risk of mycotoxin contamination of wheat under conservation tillage, we developed the decision support system FusaProg. Our model takes into account the effects of cropping factors, previous crops, soil and straw management, as well as the F. graminearum susceptibility of the planted variety. These factors are used as driving variables and are combined with the prevailing weather conditions and growth stage in order to predict the deoxynivalenol content of a specific wheat plot before harvest. To use FusaProg as a threshold‐based tool to control F. graminearum with optimized timing of fungicide applications, forecasts of deoxynivalenol contents are conducted during the flowering period. FusaProg is an Internet‐based decision support system which not only provides information about the local and regional F. graminearum infection risks; it also forecasts plot‐specific deoxynivalenol contamination of winter wheat. In 2006, the system was evaluated by Swiss cantonal plant protection officers, and in 2007, the system will be made available to Swiss wheat producers.