Parameter estimation of the dynamic model for chill accumulation in temperate fruits
Accurate quantification of the chill requirements to release dormancy in temperate fruit trees is crucial to predict bud burst or flowering dates, to plan the addition of chemicals for dormancy-break or to predict possible consequences in quality or production due to the expected temperature increase caused by the climate change. Among the existing models to this purpose, the dynamic model by Fishman et al. (1987a, b) seems to be the most accurate and its use is advised by several authors. It considers that a dynamic process occurs at a biochemical level that determines the dormancy breaking through the accumulation of the so-called portions. This model contains some parameters, which describe reaction rates, whose values were initially estimated under several assumptions. These assumptions were done based on observations in peach trees. However, the model has been used for other species using the same values of the parameters proposed by the original authors. Since these assumptions might not be completely valid for other species than peach (or even for other varieties), the values of the parameters which make the model optimally describe the chill accumulation for different species could vary. Taking advantage of the huge amount of phenological data collected by the different breeding groups/entities, we propose here a data-based calibration of the dynamic model to obtain optimal values for the model parameters for different species or even varieties. Using field data, the optimization problems consists in finding the model parameters which minimize the dispersion of the chill requirements to break dormancy among years, subject to a set of constraints. The application of this approach over a set of data of apricot cultivars reveals that the dispersion observed among different years can be globally reduced.
Egea, J.A. and Ruiz, D. (2021). Parameter estimation of the dynamic model for chill accumulation in temperate fruits. Acta Hortic. 1311, 303-310
dormancy, chill requirements, dynamic model, parameter estimation