Development and validation of phenological models for eight varieties of sweet cherry (Prunus avium L.) growing under Mediterranean climate condition

Abstract

The projected increase in temperatures caused by climate change will modify the phenological stages of sweet cherry in Mediterranean-type climate regions. This study aimed to develop and validate phenology models for eight sweet cherry varieties (Sam, Lapins, Rainier, Royal D, Santina, Regina, Kordia, Sweet Heart) in the Maule Region, Chile. The models were based on growing degree days (GDD), and the stochastic monomolecular model was employed to account for the randomness caused by measurement errors affecting the prediction of phenological states. A dataset of phenological stages was compiled over four growing seasons: 2017-2018, 2018-2019, 2019-2020, and 2020-2021. The results revealed a significant correlation (with R2 values between 0.96 and 0.98) the modified phenological (PS) scale of Biologische Bundesanstalt, Bundessortenamt, and Chemical Industry (BBCH) and GDD for all eight varieties. The validation of the model demonstrated root mean square error (RMSE), mean absolute error (MAE), and model efficiency (FE) values ranging from 1 to 1.3, 0.79 to 1.1, and 0.82 to 0.92, respectively. Finally, monomolecular model estimated the modified scale of BBCH with errors less than 2 %.