Modelling the impact of future climate scenarios on yield variability of grapevine (Vitis vinifera L.).

A mechanistic growth model was used to evaluate the mean yield and yield variability of grapevine Vitis vinifera L. under current and future climates. The model used was previously validated using field experiment data. The effect of elevated CO2 on grapevine growth was also considered. Adaptation of 2 varieties (Sangiovese and Cabernet Sauvignon) to scenarios of increased CO2 and climate change, and potential changes in agricultural risk (i.e. inter-seasonal variability), were examined. Before testing the effect of climate scenarios, we analysed the sensitivity of modelled grapevine yield to arbitrary changes in the 3 driving variables (temperature, solar radiation and CO2). The results showed the model to be more sensitive to changes in CO2 concentration and temperature than to changes in radiation. Analyses made using transient GCM (general circulation model) scenarios (UKTR and GFDL) showed different changes in mean fruit dry matter for the different scenarios, whereas mean total dry matter, and fruit and total dry matter variability, were predicted to increase under almost all the scenarios. Predictions based on equilibrium scenarios (UKLO and UKHI) gave similar results. For Sangiovese, variety adaptation analysis suggested a better adaptation in terms of mean production, but a worse adaptation in terms of yield variability.