Development of a predictive model of phenology in grapevines cv. Cabernet Sauvignon under conditions of high spatial variability in Maule Valley

Phenology monitoring is indispensable for planning vine managements. This study aimed to develop a prediction model of phenology. This was applied to Cabernet Sauvignon variety, grown under high and low spatial variability field located in the Maule Valley (Chile). The predictive model was based on Mitscherlich monomolecular equation, where the dependent variable corresponded to the phenological stages and the independent variable corresponded to the degree-days (base 10°C). Model calibration was performed with data from 2 seasons (2009-10 and 2010-11), while model validation was performed using data from 2011-2012 season for the field under study, as well as a separate dataset coming from three Cabernet Sauvignon fields grown under low spatial variability, for which phenological information from 2000-2001 season was available. The results showed that the proposed model had a high degree of fit between estimated and observed phenological stages (R2 > 0.98 and standard deviation between 1.0 and 1.84 phenological units), suggesting that the model proposed can be used to predict phenology stages in Cabernet Sauvignon vineyards established under high and low spatial variability and different management systems in the Maule Valley.