ITK Vintel™: predicting midday leaf water potential from pedoclimatic data

Abstract

Vine water stress significantly affects wine production. Dynamics of soil water availability during the growing season is a major component of grape quality and wine typicity. In the context of climate change, efficient management of vine water stress is mandatory to produce wines in accordance to market demand. Vine water status is usually assessed in the field by occasional vine water potential measurements. The innovating Decision Support System (DSS) ITK Vintel™ allows following dynamics of vine water status during the growing season without any soil or physiological sensors. Based on a mechanistic model soil-vegetation-atmosphere-transfer (SVAT), ITK Vintel™ is able to compute daily predawn leaf water potential from mean vine and soil physical characteristics data and daily local weather data at plot level. According to wine production regions, leaf water potential is preferentially measured at midday than at predawn. Leaf water potential reflects soil water availability when measured at predawn whereas it is integrative of both soil availability and vine physiology when measured at midday. This study aims at predicting daily dynamics of midday leaf water potential during the growing season from ITK Vintel™ mechanistic model. Outputs from the ITK Vintel™ model and data collected in California and Oregon allowed computing a statistical model able to predict midday leaf water potential from soil water content, daily meteorological data and plant leaf area index. Results show correlation up to 70% between simulated and observed data and a high correspondence between observed and simulated midday leaf potential in terms of irrigation preconization. The equation still needs to be improved and validated at a larger scale.