Weather-based predictive models for Diaporthe helianthi ascospore release in Uruguay

Abstract

Stem canker (SC), caused by Diaporthe helianthi, is the most serious sunflower disease in Uruguay. Yield losses have been estimated up to 75%. Chemical control is one of the strategies used to manage this disease, but fungicide application should be done before symptoms are visible. Ascospores are the primary source of inoculum, they are produced in perithecia which develop in infected stubble and are dispersed by wind to infect plants. As in other monocyclic diseases, quantifying primary inoculum is essential to predict an epidemic. In this study, ascospores were trapped on microscope slides with solid petroleum jelly which were placed on top of flat open cages filled with natural infected stubble. Cages were placed outdoors, slides where replaced twice a week and stained ascospores were counted under the microscope. Our objective was to develop weather-based models to predict ascospore release levels of D. helianthi from infected stubble. Explanatory weather variables were calculated during the seven-day periods prior to each field weekly ascospore count using daily weather station data from La Estanzuela, Uruguay. Then, logistic models were fit to estimate probabilities of having severe or moderate to light levels of ascospore counts. The best models included variables associated to the precipitation and dew-induced wetness frequency, combinated with the simultaneous occurrence of high relative humidity or low thermal amplitude records. Estimating the evolution of ascospore release through the weather-based models might help to guide preventive fungicide applications to control stem canker in Uruguay.