Evaluation of EPI forecasting model for grapevine infection with inclusion of uncertainty in input value and traceable calibration

Forecasting models are useful tools to predict and manage epidemiological grapevine infection such as downy mildew (Plasmopara viticola). These models have improved the quality of the output data, but none of them considered the quality of the input data in terms of evaluation of measurement uncertainty and traceability of the sensors of the weather conditions measuring system. Air temperature and relative humidity sensors used in this study have been calibrated by means of the transportable climatic chamber developed by INRiM, directly traceable to reference sensor, a Capsule type Standard Platinum Resistance Thermometer calibrated at the fixed points of the ITS_90 scale. The outcomes data collected during 2011 and 2012, with inclusion of measurement uncertainties, were used as a input values for the simulation by employing the EPI model. The simulations with inclusion of measurement uncertainty foresee with better accuracy the period of infection and the moment to make treatments in the vineyard respect the simulation without inclusion of uncertainty, up to 7 days. In this study the meteorological observations in field have been improved by applying a metrological approach and calibration methods for automatic weather stations in homogeneous agricultural sites.