Decision Trees to Forecast Risks of Strawberry Powdery Mildew Caused by Podosphaera aphanis

Powdery mildew (Podosphaera aphanis) is a major disease in day-neutral strawberry. Up to 30% yield losses have been observed Eastern Canada. Currently, management of powdery mostly based on fungicide applications without consideration risk. The objective this study use P. aphanis inoculum, host ontogenic resistance, and weather predictors forecast the risk strawberry using CART models (classification trees). data used build trees were collected 2006, 2007, 2008 at one experimental farm six commercial farms located two main strawberry-production areas, while external validation same 2015, 2016, 2018. Data proportion leaf area diseased (PLAD) grouped into four severity classes (1: PLAD = 0; 2: > 0 <5%; 3: >5% <15%; 4: 15%) for total 681 136 cases training validation, respectively. From initial 92 variables, 21 selected following clustering. tree with best balance between number highest accuracy was built with: airborne inoculum concentration susceptible leaves day sampling, mean relative humidity, daily hours temperature 18 30 °C, saturation vapor pressure 10 25 mmHg during previous 6 days. For training, internal datasets, sensitivity, specificity, ranged from 0.70 0.90, 0.87 0.98, 0.82 0.97, classification rules estimate can be easily implemented decision support systems treat only when necessary thus avoid preventable unnecessary treatments.