Fire4CAST – a new integrative epidemiological forecasting model for the accurate prediction of infection risk and effective control of fire blight in Pyrus orchards

Abstract

AbstractFire blight disease, caused by Erwinia amylovora is present worldwide and affects over 40 countries in Europe where it is listed as a quarantine or regulated pest often due to ineffective control strategies maladapted to the respective production systems. In Portugal, the disease was confirmed in 2010 and the occurrence of disease outbreaks in new production areas has risen over the years. The disease affects the national production of apple and pear fruits, with greater impact on the national pear variety ‘Rocha’, widely exported to European countries and Brazil. The mild temperatures and high relative humidity promote the progression of the disease during winter, revealing the potential activity of the bacterium in the latency period (LP) of the host. Infection alert risk using the established predictive models Maryblight TM, Cougarblight and BIS98 was put in place in 2013 by Centro Operativo e Tecnológico Hortofrutícola Nacional (COTHN). These attempts to control the spread of this disease, showed low accuracy for the Portuguese epidemiological reality. Within the framework of project Fire4CAST we developed a new epidemiological model to predict fire blight disease using a systems biology approach integrating microbiological, cytological and genomic pathogen data with phenological host development and climatic variables. The presence of E. amylovora was monitored in orchards with fire blight history using standard laboratory tests. Simultaneously, the implementation of immune-flow cytometry (IFCM) highlighted the viability of E. amylovora populations prevailing during winter and early spring, long before bloom risk period. The integration of the whole data set allowed the development of the Fire4CAST predictive model, able to monitor the expected infection date (EID) and to define accurate outbreak alarms. Fire4CAST model enabled the detection of outbreak risk during winter based on rules that consider climatic data variables, which were validated by effective presence of live and active E. amylovora populations and real-time quantitative PCR (qPCR) data, accomplishing a precision rate of 83%. Field application of Fire4CAST can hopefully guide the implementation of successful control strategies, leading to more sustainable pome chain production areas.