New publication - PhenoFlex phenology modeling framework

Predicting bloom dates of temperate fruit tree species, such as apples, cherries, peaches etc., is challenging. Trees fall dormant in winter, and they don’t seem to be doing much until they start growing again in spring. They are receptive to temperature signals during this time, however, and the date of budbreak depends on accumulation of so-called chill and heat. How exactly to quantify these two temperature cues has been under study for well over a century. Especially for chill, quite a few models have been proposed, but none of them have been fully convincing. The most promising model so far, the Dynamic Model, is so complicated that very few researchers have been able to adapt its parameters to specific cultivars. To predict bloom, chill models must be combined with a heat model – another challenging task, because the nature between the chill and heat accumulation phases remains poorly understood. At the University of Bonn, a team of researchers from the HortiBonn group and the Helmholtz Institut für Strahlen- und Kernphysik has now produced a modeling framework that combines state-of-the-art capabilities in chill and heat modeling to predict bloom dates. The PhenoFlex framework features a flexible transition between the major dormancy phases that can accommodate different concepts of how chill and heat may be related.How exactly the phase transition occurs and what parameters to use in the chill and heat sub-models can be decided by data. The framework can be fitted to observed bloom data for specific cultivars, with the fitting procedure producing suitable estimates of all model parameters.The PhenoFlex framework, which was developed by Prof. Dr. Eike Luedeling, Dr. Katja Schiffers, Till Fohrmann and Carsten Urbach, has already been tested with apple and pear phenology data from Campus Klein-Altendorf, the experimental station of the University of Bonn. Results from these tests are promising, even though the absence of any really warm winters at this location presented a challenge to the fitting algorithm.The PhenoFlex framework has been integrated into the chillR package for R, where it is accompanied by a detailed vignette.The article describing the model was published in Agricultural and Forest Meteorology. It can be accessed free of charge here.Reference: Luedeling E, Schiffers K, Fohrmann T, Urbach C, 2021. PhenoFlex - an integrated model to predict spring phenology in temperate fruit trees. Agricultural and Forest Meteorology 307, article 108491. doi: https://doi.org/10.1016/j.agrformet.2021.108491