Model for Spatial Extrapolation of Stomatal Conductance on a Young Drip-Irrigated 'Arbequina' Olive Orchard under Semi-Arid Condition

Stomatal conductance (g(s)) is a plant variable normally used in models that estimate plant water requirements, and furthermore can be used as an integrative tool to monitoring the level of plant water restriction. However, g(s) requires skills and heavy devices, which drastically limit the number of replications either in space or in time. Moreover, due to the significant spatial variability in olive orchards, extrapolation of one measurement on a whole field or on a whole orchard scale can hardly be done. The aim of this paper is to propose a model that uses only one measurement performed on a reference site in order to predict g(s) and its spatial variability over the whole field. An experiment to identify and test the model was carried out over a commercial drip-irrigated olive orchard (Olea europaea L. 'Arbequina') located in Pencahue valley, Maule Region, Chile (35 degrees 23'LS; 71 degrees 44'LW; WGS 84; 96 m a. s. l.). Measurements of g(s) were done in 29 dates using a porometer (PP Systems, PMR-5, Amesbury, Massachusetts, USA) during three growing seasons (2008-09, 2009-10, 2010-11). A regular grid of 30 measurement sites was distributed over the field. Results showed a significant within-field variability of g(s) over space and time, and the existence of significant linear relations among g(s) values. Also, the model was able to predict spatial variability of g(s) with a spatial and temporal mean error of 53.3 and 52.0 mmol m(-2) s(-1), respectively. Further studies will aim to confirm the relevancy of this approach for different cultivars and locations.