A water stress index based on water balance modelling for discrimination of grapevine quality and yield

摘要

Aims: A water stress index based on a water balance model was tested as a tool for classifying the water stress paths experienced by grapevines in various French Mediterranean vineyards. The relations between the index value and grapevine yield and berry quality (sugars, organic acids, anthocyanins) at harvest were investigated.Methods and results: A data set of 102 situations, each combining one location, one variety, one vintage and one water regime (irrigation or, most often, no irrigation), was collected for the study. The Fraction of Transpirable Soil Water (FTSW) was simulated by a unique-soil-reservoir water balance model at a daily time step. Five classes of water deficit were delimited from specific decreasing thresholds of FTSW over four periods between flowering and harvest. These thresholds were derived from predawn leaf water potential values because over decades, grapegrowers and researchers have shared references and built expertise by using this variable throughout the Mediterranean region. A water stress index resulting from the levels of water deficit reached at each of the four periods of the cycle was calculated. This index was correlated with yield per vine, berry weight, and berry sugar and organic acid contents but not with berry anthocyanin content.Conclusion: A simple water stress index, based on the water balance model, exhibited significant correlations with yield and berry quality for various cultivars and pedo-climatic conditions in Mediterranean vineyards.Significance and impact of the study: This water stress index is a valuable tool for explaining the variations in grape yield and quality among various locations and years because it reflects the vineyard water stress history in relation to rainfall regime and soil conditions. Improvement would come from the simulation of FTSW during winter, notably for soils of high Total Transpirable Soil Water. One potential application is the quantification of water stress change brought by irrigation in Mediterranean vineyards, and its relation to grapevine production. IntroductionFor many red grapevine cultivars, berry quality relies upon a progressive reduction of water availability after flowering to reach moderate water deficit during the ripening phase (Matthews et al., 1987; Matthews and Anderson, 1988; Ojeda et al., 2001, 2002). The relationships between water availability and berry yield and quality, together with the identification of relevant variables to diagnose the level of water deficit, have been investigated in numerous studies during the last decades.The grapevine water status can be assessed by soil water measurements, including soil water content or soil water potential measurements (Pellegrino et al., 2004). However, these measurements may be difficult in many vineyards, due to deep plant rooting and soil stoniness (Van Leeuwen et al., 2009). Under such conditions, leaf water status measurements are an alternative to soil measurements for the diagnosis of the level of water deficit experienced by the plant. Leaf water potential at predawn (Ψpd), at midday (Ψmin) or on non-transpiring leaves (Ψstem) have been widely used as indicators of water stress in vineyards (Van Leeuwen et al., 2009, 2010). Thresholds of water deficit levels with regard to grapevine production have been validated from the use of Ψpd (Ojeda et al., 2001). A major constraint in the use of leaf water potential-based indicators, specifically Ψpd, is the high measurement frequency required over the cropping seasons, due to their short validity after a rainfall event. To overcome the limits described above, attempts have been made to assess plant or soil water status from modelling approaches. A climate-based model was shown to be reliable to predict Ψpd in non-irrigated Mediterranean vineyards (Taylor et al., 2012). However, a local calibration of the model was necessary. Grapevine Ψpd was also shown to be tightly correlated with the Fraction of Transpirable Soil Water (FTSW), which can be simulated with soil water balance models (Lebon et al., 2003; Pellegrino et al., 2004).Substantial progress in vineyard soil water balance modelling was made over the last decades. Riou et al. (1989) parameterized a simple model of light interception. This model was used in a simple water balance model, based on a unique soil reservoir to simulate soil water content at a daily time step. This latter water balance model was improved to describe the transpiration response to soil water availability (Riou and Lebon, 2000; Lebon et al., 2003; Pellegrino et al., 2004). Pellegrino et al. (2006) have used this modelling approach to build a diagnosis tool simulating the pattern of water stress experienced by grapevines under Mediterranean conditions. Later developments have been the adaptation of the model to intercropped vineyards (Celette et al., 2010), with a soil water reservoir split into two compartments in order to consider the soil volume explored by both the interc