Application of the MaryBlytTM version 4.3 to the prediction of fire blight in raspberry

Abstract

Fire blight of apple (Malus ×domestica) and pear (Pyrus communis), caused by Erwinia amylovora, is a well-known disease and has been extensively researched. Fire blight of raspberry (Rubus idaeus), caused by a distinct strain of E. amylovora is less well known but appears to be similar in its behavior. The purpose of this study was to determine if the MaryBlytTM version 4.3 forecasting system for predicting fire blight in apples and pears could be applied to the prediction of fire blight in raspberry. Seven years of weather data and field observations from a naturally infected plot of Kentville raspberry selection ‘K81-6’ were entered into the MaryBlyt version 4.3 forecasting system. This system predicted flower infection events and the subsequent appearance of symptoms within 0-3 days of the observed symptoms for 2000-2006. In 4 of the 7 years, the appearance of systemically infected raspberry cane laterals was within ≤3 days of the MaryBlyt version 4.3 prediction for canker blight symptoms in apple when a modified degree-day threshold specifically for raspberry was used. Also, changing the degree-day base temperature from 12.7 °C to 4.0 °C reduced the error of prediction, but the error for three of the years was still unacceptably high. This study also showed that insect-vectored infections of primocanes cannot be successfully modeled by adaptations of the MaryBlyt version 4.3 program. However, a simple degree-day model independent of MaryBlyt was developed to predict flowering.