OENO One (Jun 2014)
Spiderδ: an empirical method to extrapolate grapevine (<em>Vitis vinifera</em> L.) water status at the whole denomination scale using δ<sup>13</sup>C as ancillary data
Abstract
Aims: The aim of this study is to test a method to extrapolate vine water status (estimated by the water potential; Ψ) over a whole appellation (protected geographical indication). The spatial extrapolation is based on an empirical approach that combines a reference site (baseline measurements) and carbon isotope discrimination (δ13C) values as ancillary data (AD). Methods and results: Experiments were conducted on the whole Tavel appellation (Gard, France). The study focused on the dominant variety: Grenache. Ψ was measured as predawn leaf water potential and was monitored over three consecutive years, 2008, 2009 and 2010, on 10, 24 and 24 sites, respectively. δ13C measurements were made at harvest in 2010 on the 24 sites. The spatial model (SPIDERδ) was calibrated using Ydata from 2009 and 2010 and δ13C data from 2010. The quality of prediction was tested on the 2008 data, considered as an independent data set. The results show that SPIDERδ was relevant in estimating Ψ at the whole appellation scale. The extrapolation model significantly improves the prediction (R² = 0.88) compared to a conventional method based on Ψ averages across the appellation (R² = 0.66). Conclusion: Based on a single measurement taken at time «t» on a reference site, SPIDERδ makes it possible to estimate Ψ on all sites where a δ13C value is available. The use of AD like δ13C makes it possible to consider the spatial extrapolation of Ψ with higher spatial resolution than when only direct measurements are used to calibrate the model. Significance and impact of the study: This work demonstrates the value of using an AD like δ13C to assess Ψ at a scale larger than the single field. This significant result opens the door to the practical use of spatial extrapolation models with higher spatial resolution.
Keywords
