Assessment of an empirical spatial prediction model of vine water status for irrigation management in a grapevine field

This study proposes and evaluates an empirically derived spatial model to extrapolate midday stem water potential (MSWP) measurements over a furrow irrigated grapevine field from a single reference site. The methodology used to build the model has previously been used successfully under non-irrigated conditions in France with pre-dawn leaf water potential. It has not previously been applied on irrigated vineyards with moderate water restriction and using MSWP. The precision of the model was calibrated and validated using a database of MSWP measurements collected from a commercial Cabernet Sauvignon (Vitis vinifera L.) vineyard located in the Maule Region, Chile, at various times during the 2009–2010, 2010–2011 and 2011–2012 growing seasons. The proposed spatial model was able to predict the spatial variability of MSWP with an RMSE<0.12MPa. Also, the model significantly improved the prediction of MSWP (r2=0.76) compared to the conventional monitoring carried out by winegrowers (r2≤0.48) under conditions of absent to severe water restriction (>−0.5 to −1.3MPa). The choice of the reference site for vine water status monitoring is important regardless of the method used. Results also showed that irrigation practices may impose a specific soil moisture regime in parts of the field; thus the selection of a reference site that is representative of the field conditions is very important for good model performance under irrigated conditions. Variability in soil and ground cover properties, rather than vine vigour, appeared to be the best information for assisting in the correct location of reference sites.