Composition and evaluation of a novel web-based decision support system for grape black rot control
D. Molitor, B. Augenstein, L. Mugnai, P. Rinaldi, J. Sofia, B. Hed, P. H. Dubuis, M. Jermini, E. Kührer, G. Bleyer, L. Hoffmann, and M. Beyer
European Journal of Plant Pathology, vol. 144, no. 4, pp. 785-798, 2015
Black rot on grapevine is a fungal disease caused by Phyllosticta ampelicida (syn. Guignardia bidwellii) affecting grape leaves as well as clusters. A novel black rot decision support system termed VitiMeteo Black rot was assembled based on existing sub-models and incorporated into the established VitiMeteo forecast and decision support platform. Based on local weather data and a 5-day weather forecast, VitiMeteo Black rot simulates the relative susceptibility of grape clusters, the occurrence and severity of infection events as well as the duration of incubation periods. Data sets obtained in extended international (14 case studies; eight monitoring locations; 11 cultivars; seven countries in Europe and North America) field monitoring campaigns in 2012 and 2013 were used to evaluate the model predictions of newly expressed symptoms on leaves. In the case of the Vitis vinifera cultivars, on average 26.3 disease assessments took place per season. On average, 9.9 predictions were classified as true positive, 8.0 as true negative, 5.2 as false positive and 3.2 as false negative. Model precision, sensitivity and accuracy were on average 64, 77 and 67 %. Potential reasons for false positive and false negative predictions are discussed. VitiMeteo Black rot is freely available for several locations in Germany, Luxembourg and Austria on the internet via the VitiMeteo platform and might be expanded to other regions in the future.