Investigating the variability of basal crop coefficient across diverse production contexts in commercial vineyards

Abstract

Vine water use is a critical determinant of vineyard management practices, especially in the context of climate change. A common approach to estimate water use involves expressing it as a fraction of grass reference evapotranspiration using a crop coefficient (Kc), often divided into three components for plant transpiration (Kcb), cover crop transpiration (Kcc) and soil evaporation (Ke). A review of the literature reveals that Kcb and a fortiori Kc values obtained under specific experimental conditions often show significant variability and limited transferability to other contexts. Factors such as canopy management, cover crop, fertilization and irrigation practices introduce discrepancies, complicating the validation of Kcb modeling assumptions and underscoring the need for direct Kcb measurements to accurately parameterize water balance models.

Leveraging sap flow measurements collected during Fruition Sciences’ consulting activities, this study explores the variability of Kcb values across diverse production conditions. Data were drawn from 119 vineyard blocks with at least three years of measurements, ensuring that the two equipped plants per block exhibited comparable transpiration profiles to minimize plant-specific effects. In total, 844 block-vintage pairs from vineyards in France and California (US) were analyzed. Hourly Kcb time series were calculated using the FAO method and smoothed via a moving average. The variability of key parameters, such as the maximum Kcb value (Kcbmax), growing degree days (GDD, base temperature of 10°C) to reach Kcbmax and the variability of post-Kcbmax Kcb values, was studied across blocks and vintages.

Preliminary findings indicate that the first and third quartiles of Kcbmax across all blocks and years were 0.36 and 0.63, typically occurring between 570 and 1325 GDD. The block-specific and vintage-specific coefficients of variation of Kcbmax respectively ranged from 0.22 to 0.39 and from 0.13 to 0.38 (first and third quartiles). Post-Kcbmax block-specific CVs varied from 0.19 to 0.30. These results are intended to spark a discussion challenging our current practices for Kcbmax estimation and to generate leads for further exploration of this database.