Abstract
Use of seed-applied fungicides has become commonplace in the United States soybean production systems. Although fungicides have the potential to protect seed/seedlings from critical early stage diseases such as damping-off and root/stem rots, results from previous studies are not consistent in terms of seed-applied fungicide's ability to mitigate yield losses. In the current study, the relationship between estimated soybean production losses due to seedling diseases and estimated seed-applied fungicide use was investigated using annual data from 28 soybean growing states in the U.S. over the period of 2006 to 2014. National, regional (northern and southern U.S.), state, and temporal scale trends were explored using mixed effects version of the regression analysis. Mixed modeling allowed computing generalized R2 values for conditional (R2GLMM(c); contains fixed and random effects) and marginal (R2GLMM(m); contains only fixed effects) models. Similar analyses were conducted to investigate how soybean production was related to fungicide use. National and regional scale modeling revealed that R2GLMM(c) values were significantly larger compared to R2GLMM(m) values, meaning fungicide use had limited utility in explaining the national/regional scale variation of yield loss and production. The state scale analysis revealed the usefulness of seed-applied fungicides to mitigate seedling diseases-associated soybean yield losses in Illinois, Indiana, North Carolina, and Ohio. Further, fungicide use positively influenced the soybean production and yield in Illinois and South Dakota. Taken together, use of seed-applied fungicide did not appear to be beneficial to many of the states. Our findings corroborate the observations made by a number of scientists through field scale seed-applied fungicide trials across the U.S and reiterate the importance of need base-use of seed-applied fungicides rather than being a routine practice in soybean production systems.