In VISCA, we have recently published three reports. The first 'seasonal forecasts quality assessment report (D2.1)' assesses the quality of two seasonal forecast systems (European Centre for Middlerange Weather Forecasting (ECMWF) System 4 and SEAS5), two bias correction approaches (calibration and simple bias correction) and a downscaling strategy (calibration). The results obtained show that there is some degree of predictability in the three demo-sites in different variables that can provide value beyond the customary use of climatology. Moreover, the verification / bias-correction / downscaling workflow developed in this task provides the basics of all future refinements that we will conduct during the remaining two years of the project, e. g. through the exploration of new seasonal prediction systems and/or downscaling techniques.
The second 'weather forecast extreme events (D2.2)' presents a detailed description of the weather models and processes applied to develop the weather forecast products for the project. Two kinds of products are delivered: short-term weather forecasts and mid-term weather forecast. While having in two days in advance the weather forecast product is useful for near future in-field activities, having weather forecast information ten days in advance could be useful to wine producers to minimize risks related to coming extreme events. For instance, if a heat wave is forecasted in 6 days, wine producers could act in advance irrigating the field more than usual before the event. Meteosim (MET) has developed and supplied weather forecasts services, which consisted in deliver the best prediction of high impact weather variables at forecast time scales from hours up to 10 days (240 hours). For the short-term forecast (up to 48 hours), a regional weather forecast model is used. On the other hand, a probabilistic model is used for mid-term forecast (up to 10 days). Finally, all weather forecast products are sent to the VDI (VISCA Data Interface), where other modellers and end-users can download these information for their use in the VISCA platform.
The third 'Report on the performance of phenological model (D2.3)' describes the principles of bud break and berry models for the seasonal forecast of the key grapevine phenological stages. The document also includes the description of the LEAF model. The document presents the calibration and validation analysis performed to test the availability of the models to predict the different phenological events. Although in its first version the validation analysis shows a good performance of the models, some efforts are needed to improve seasonal predictions. The advances in model performance would be possible thanks to the experiments specifically designed in the pilot plots.