303 Disagreement between Tree Size and Crop Coefficient in ‘Conference’ Pear: Comparing Measurements by a Weighing Lysimeter and Prediction by CropSyst J. Marsal, J. Casadesus, G. Lopez and J. Girona Irrigation Technology IRTA, Rovira Roure 191, 25198 Lleida Spain C.O. Stöckle Biol. Systems Engineering Dept. Washington State University Pullman, Washington USA Keywords: simulation models, transpiration, leaf area, ground cover Abstract Accurate prediction of a crop coefficient (Kc) is necessary for proper irrigation management. We explored the use of a computer-based model, CropSyst, for determining irrigation requirements of ‘Conference’ pear trees and we assessed the accuracy of Kc predictions. Values of Kc were compared to those obtained, over 2002- 2010, from lysimeter-grown trees. Lysimeter data over the years indicated weak Kc response to the increase of tree size with age. CropSyst predicted irrigation require- ments using tree light interception and water uptake sub-model components. The most important parameters in CropSyst to determine Kc are: full canopy Kc (Kc,fc), and canopy porosity. Parameters of the model were adjusted using data obtained from the lysimeter in 2010. Using the parameterization of 2010, we found poor agreement between simulated and measured Kc over different seasons (2002-2009). The main reason for poor agreement was attributed to yearly changes in the Kc,fc parameter. When Kc,fc was yearly adjusted so that differences between observed and simulated Kc were minimized, it was observed that Kc,fc decreased with tree age. Therefore the lack of agreement between tree size and Kc was attributed to a decreased capacity for consuming water with tree age. In conclusion, irrigation scheduling methods to be applied to ‘Conference’ pear orchards shouldn’t only rely on canopy light inter- ception. INTRODUCTION Previous research carried out in pear grown in lysimeters (Girona et al., 2011) has shown that ‘Conference’ water consumption is difficult to assess following FAO’s water balance method (Allen et al., 1998). FAO method is based on calculating ETc = Kc × ETo, where ETc is crop evapotranspiration, ETo is the reference evapotranspiration and Kc is a crop-specific coefficient. Kc values are provided in the literature but site specific adjustment is needed. Kc is supposed to vary mainly according to the size of canopy cover, the phase of development within a year, soil cover conditions, and certain peculiarities related to each considered species. The canopy cover is perhaps the most important factor to take into account for accurately estimating Kc. The work of Ayars et al. (2003) showed encouraging results for Kc determination based on noon intercepted radiation for peach growing in a weighing lysimeter. Goodwin et al. (2006) showed disadvantages to this, and possible ways to overcome them, by extending measurements of intercepted radiation beyond noon. Tree height and estimation of effective canopy groundcover from a solid canopy have recently been included in the calculation of Kc (Allen and Pereira, 2009). Canopy porosity has also been highlighted as an additional factor that should be considered beyond the effective ground cover estimated from a solid canopy (Marsal et al., 2012). Nevertheless, the latter approach based on canopy light interception emphasizes a physical interpretation of tree water consumption. Girona et al. (2011) realized that only considering tree size and intercepted radiation would not suffice to predict ‘Conference’ pear Kc. In this study we would like to deepen the reasons for disagreement between crop intercepted radiation and observed Kc by considering a reinterpretation of Girona et al.’s Proc. VII th IS on Irrigation of Horticultural Crops Eds.: P. Braun et al. Acta Hort. 1038, ISHS 2014