Evaluation of model for maturation prediction in apricot cultivars

Phenological predictions may help to obtain more stable crop yields and fruit quality through better crop management and may also improve market delivery of fruits. The aim of this study was to evaluate the performance of an easy to apply model for early prediction of maturation in apricot cultivars based on the responsiveness of phenological timing to the preceding thermal conditions (air temperature averaged over 30, 45 and 60-day periods starting from the date of full flowering). The model was initially developed using daily temperature data and phenological records for 20 apricot cultivars grown in the Belgrade region (Serbia) over eight seasons during the period 1995-2004 (two years were excluded because of severe spring frost damage). The data set for revalidation of the model performance consisted of phenological and temperature records from 2009 to 2014. The proposed model with newly included data proved to be capable of predicting the onset of maturation in apricot cultivars up to two months ahead with high accuracy. On average for all cultivars, the mean absolute differences between the observed and predicted maturation dates obtained by 30, 45 and 60-day models were 5.0, 4.2 and 2.7 days, respectively.