Apple scab: numerical optimization of a new thermal time scale and application for modelling ascospore release in southern France

Abstract

Adequate protection of apple trees during the primary contamination period is a cornerstone for management of apple scab. Correct timing of spring treatments is fundamental and thus, much effort has been devoted to forecasting ascospore release by Venturia inaequalis. Most models rely on degree‐day accumulation starting from a biofix date established yearly on the basis of biological observations. Here, the potential of using a single calendar date as a biofix and new types of time scales were explored, with the help of numerical optimization with field‐collected data. Using data acquired between 1996 and 2008, the daily rate of development for V. inaequalis primary inoculum was assessed by fitting generic time scale functions, a method that requires the smallest number of assumptions about the effect of temperature on the biological phenomenon. An optimal calendar biofix was established for Provence and use of non‐linear functions relating pseudothecial development rate to temperature for accumulating thermal time was compared with the usual linear response in standard degree‐day models. A model was then constructed using four additional years of data for validation. The predictive value of the model was further improved by adjusting the time scale with ‘accelerating rules’ to take into account the positive influence of rainy days on pseudothecial maturation prior to ascospore release. However, ‘halting rules’ inserted in the time scale to account for dry days during the ascospore release period strongly reduced the predictive value of the model for southern France, suggesting the possible occurrence of strains adapted to dry conditions.