ISHS Acta Horticulturae 785: International Symposium on Grape Production and Processing
A STATISTICAL MODEL FOR DESCRIBING THE EPIDEMIOLOGY OF INCIDENCE OF DOWNY MILDEW IN GRAPES |
| Authors: | R.D. Rawal, R. Venugopalan, A.K. Saxena |
| Keywords: | Anab-e-Shahi, downy mildew, Plasmopara viticola, statistical model |
| Abstract:
Statistical models were developed to optimize the role of ecological variables and simultaneously predict downy mildew incidence in grapes (cv. Anab-e-Shahi) at Indian Institute of Horticultural Research, Bangalore, India. Further, as a measure of goodness-of-fit, the coefficient of determination (R2) was used to evaluate the empirical models developed. The initial results indicated only 39% of the variability in percent disease incidence was reflected collectively by preceding week’s ecological variables viz., minimum temperature (X1), maximum temperature (X2), relative humidity at 07.30 h (X3) and at 13.30 h (X4), evaporation (X5), amount of rainfall (X7) and wind speed (X6). This may be due to the reason that percent disease incidence was in the range zero to as high as 90% during the period of study. To eliminate such a high variability, new models were developed separately for severity less than 10% and severity exceeding 40%. Correlation analysis showed that both maximum and minimum temperatures negatively influenced, whereas humidity levels both at morning and evening h were positively influenced towards the absence of disease incidence. The results indicated that if the disease severity was less than 10%, about 65% of the variability in percent disease incidence followed the equation of y = -12.29 – 0.065X1 – 0.04X2 + 0.01X3 + 0.156X4 – 0.53X5 – 0.31X6 + 0.03X7. Correlation analysis in case of high disease incidence showed that relative humidity, maximum temperatures, wind speed and number of rainy days jointly influenced the disease. In the case of high severity about 60.3% variability in percent disease incidence followed the equation of y = 88.23 – 4.17X1 + 3.43X2 – 1.77X3 + 0.83X4 + 3.06X5 + 1.36X6 – 0.41X7. This model would be highly useful in predicting the disease severity and thereby programming for the disease management could be worked out, which will further reduce the fungicide load on the vine. | |
Download Adobe Acrobat Reader (free software to read PDF files) | |
