A new ML-based system for early and accurate vineyard yield forecasting

Vineyard yield forecasting is a key issue for vintage scheduling and optimization of winemaking operations. High errors in yield forecasting can be found in the wine industry, mainly due to the high spatial variability in vineyards, strong dependency on historical yield data, insufficient use of agroclimatic data and inadequate sampling methods. Today, errors can reach values within the range of 20%-30% per block. Thus, improved methodologies for early and accurate vineyard yield forecasting are needed.

The work "New ML-based system for early and accurate vineyard yield forecasting" presented by José Cuevas-Valenzuela, innovation leader of the Center of Research and Innovation, Viña Concha y Toro, Chile, outlines a new system based on precision viticulture, which is cost-effective, fast and accurate for forecasting the yield in the vineyard before veraison is completed. With this system, the forecast error is reduced to below 10% per block, the trials carried out concerned high yield Cabernet Sauvignon (CS) vineyards located in Maule Valley (Chile), during seasons 2019 and 2020.

Video of the seminar held during ENOFORUM USA 2021 (4-5 May 2021)

Annual subscription to Infowine: The subscription, at a cost of € 60 (VAT included) entitles you to one year's access to all the documents published on the site, including the historical archive (click here).