EFFECTS OF METEOROLOGICAL FACTORS ON CONIDIA INFECTION AND PRODUCTION OF ERYSIPHE NECATOR, PODOSPHAERA CLANDESTINA, AND PODOSPHAERA MACULARIS ON THE LEAVES OF GRAPE, CHERRY AND HOP
Grape, cherry and hop powdery mildew, caused by Erysiphe necator, Podosphaera clandestina, and P. macularis, are common problems in WA. Since conidia concentration is strongly related to disease intensity, it is an effective approach for forecasting powder mildews to explore the relationship between meteorological factors and conidia concentration. To assess effects of temperature (T) and relative humidity (RH) on the sporulation of P. clandestina, experiments were conducted using mature colonies of powdery mildew on the detached leaves. Study showed that sporulation were optimized at the T of 22°C and RH of 85%, and that 8 h incubation in dark with T of 10 °C and RH of 70-90% were favorable to sporulation. Similar studies were conducted to determine effects of RH on sporulation of P. macularis at 20°C. Results demonstrated that there was a significant negative linear relationship between conidia and RH at 80-97.5% during 48 h incubation. Under 24h fluctuating RH conditions, 8 h incubation at RH of 80% was favorable to sporulation. In the studies of ultraviolet (UV), conidia were trapped using Burkard volumetric sampler. Results showed that low levels of UV resulted in higher population of E. necator conidia, while high UV had the opposite effect. The thresholds of UV-B and UV-A were 0.5 W/m2 and 13 W/m2, respectively. The results of cross correlation analysis described negative relationship of UV with conidia concentration in current day and the previous two days. The cross correlations of conidia concentrations with UV-A were -0.306 in lag 0 and -0.182 in lag 2, and with UV-B were -0.311 in lag 0 and -0.235 in lag2. Modeling studies of conidia indicated that time series and multivariate techniques were appropriate methods on conidia forecasting. A dynamic prediction model of E. necator conidia was developed using AR (4) and PDLREG (7, 2). Predictors include previous 4 days conidia concentration, and previous 7 days average dewpoint and duration of temperature at polynomial degree of 2. Three dynamic prediction models of E. necator, P. clandestina, and P. macularis were developed using AR (4) and principle component analysis that simplified all related meteorological parameters to 5 components.