Quaternary integrated pest management concept for powdery mildew in sugar beet. I. Analysis of epidemic determinants to predict disease onset

Abstract

Epidemiological field studies on sugar beet leaf diseases under variation of site, year and cultivar (n = 76) were performed from 1993–2004 in sugar beetgrowing regions of Lower Saxonia and Bavaria, Germany. The records established Erysiphe betae to cause powdery mildew as a main foliar disease in sugar beet, in particular, when leaf blotching caused by Cercospora beticola is relatively missing and therefore leaf necrosis does not interfere with the obligate biotrophic mode of parasitism of powdery mildew. The epidemic onset of E. betae is widely spread from mid-July until mid-September. The epidemic onset in cultivars with a low susceptibility was delayed by two weeks and the amount of infection during a season measured as AUDPC (area under disease progress curve) was reduced by 20–60%. Under conditions favouring the disease development such as dry weather and high cultivar susceptibility, the epidemic showed typical abundance dynamics. After a first period of slow progression, disease severity (DS) of powdery mildew increased by 10–15% per week, followed by an exponential increase of conidia production. At the culmination point, the weekly increase of DS slowed down as powdery mildew tended to cover up to 50–60% of the leaf surface. This epidemic stage coincided with a peak of conidia production, which subsequently dropped down and sometimes was completely finished towards the end of the season (mid-September to October). As a consequence, the visible mycel layer disappeared, sometimes entirely. In view of disease prediction, temperature degree days (TDD) calculated as the sum of daily mean values over specific periods (alternatively 1/1, 5/1, 5/16, 6/1, 6/16 until symptom appearance) could not explain differing disease onset times. TDDs showed no advantage as the variance could not be reduced in relation to the corresponding number of days. Therefore only an empirical negative prognosis could be applied, where the epidemic onset is to be expected with increasing risk after certain thresholds have been exceeded. Using the parameter TDD the thresholds are defined with 1300 and 1400°C, or, oriented to the calendar, the thresholds indicating a beginning risk are mid-July or beginning of August, for highly susceptible or low susceptible cultivars, respectively.