Department of Plant Pathology, The Ohio State University, Ohio Agricultural Research and Development Center, Wooster 44691.
Mist chamber experiments were conducted to quantify and model the effects of post-anthesis moisture on Fusarium head blight (FHB) index (IND) and deoxynivalenol (DON). Four mist treatments, one daily and three intermittent, were applied during an 8-day window immediately after anthesis, plus an untreated check. All intermittent mist treatments received moisture on 4 of the 8 days, but the distribution of the supplemental moisture during the treatment window varied among the treatments. Separate sets of spikes in each treatment were either spray or point inoculated with a spore suspension of Fusarium graminearum. Based on results from linear mixed-model analyses, mist treatment had a significant effect on arcsine-square root-transformed IND (arcIND) and log-transformed DON (logDON) in spray-inoculated spikes but only a marginal effect on DON in point-inoculated spikes. The daily mist treatments (Mist1) consistently had the highest mean IND and DON but several of the 4-day intermittent mist treatments were not significantly different, particularly for point inoculations. Only Mist1 and one of the intermittent mist treatments (Mist2; 2 days of mist at the beginning and end of the treatment window) had significantly higher infection efficiency (estimated diseased spikelets per spore) than the check; however, none of the treatments increased the rate of disease spread within the spike (based on visual symptoms) relative to the check. For all treatments, there was a significant, positive linear relationship between IND and logDON, with estimated mean regression slopes (rates of logDON increase per unit increase in IND) of 0.04 and 0.02 logDON %−1 IND for the point- and spray-inoculated experiments, respectively. Mist treatment did not have a significant effect on the slope but had a significant effect on the intercept. The height of the regression line (logDON after adjusting for IND) was consistently higher for Mist2 than for Mist1 for both point- and spray-inoculated spikes. Estimated mean back-transformed DON at a fixed level of IND was 4.9 and 2.9 ppm higher for Mist2 than Mist1 in the spray- and point-inoculation experiments, respectively. Generalized linear mixed models were used to estimate the risk of IND and DON exceeding critical thresholds, showing similar results among treatments in terms of estimated probabilities. The estimated probabilities of IND ≥ 10% at 20 days after inoculation and DON ≥ 2, 5, and 10 ppm were not significantly different between Mist1 and Mist2. These results suggest that post-anthesis moisture patterns may play a role in DON exceeding critical thresholds even when FHB level are relatively low.
logistic regression model.