Contents lists available at ScienceDirect Agricultural Systems journal homepage: www.elsevier.com/locate/agsy The role of soils in the analysis of potential agricultural production: A case study in Lebanon A. Bonfante a,⁎ , M.H. Sellami a , M.T. Abi Saab b , R. Albrizio a , A. Basile a , S. Fahed b , P. Giorio a , G. Langella a , E. Monaco a , J. Bouma c a Institute for Mediterranean Agricultural and Forest Systems - CNR-ISAFOM, Ercolano, Italy b Lebanese Agricultural Research Institute - LARI, Lebanon c Em. Prof. Soils Science, Wageningen University, The Netherlands ARTICLE INFO Keywords: Soil heterogeneity Supplemental irrigation Crop simulation models Durum wheat Barley Potential agricultural production SWAP ABSTRACT Maintaining cereal production in the Bekaa valley in Lebanon presents a serious challenge. Lack of water is the driving force of agricultural research which is mainly focused on introduction of drought resistant cultivars, application of conservation tillage and supplemental irrigation. In this context forty-eight experimental plots were laid out for three years in a statistical split plot design. The statistical analyses showed that aboveground biomass and yield were significantly affected by irrigation for barley but not for the yield of durum wheat. Effects of soil tillage practices and introduction of new cultivares were not significant. A soil survey indicated that the implicit assumption of soil homogeneity of the agronomic design was correct for surface soil but that two different soil types (Cambisols and Fluvisols) had to be distinguished considering subsoil conditions and corresponding rooting patterns. Therefore, the main objective of this paper was to de- termine the effects of different soil types on crop response and, in addition, to assess how physically-based modeling can predict future effects of climate change on crops and soils. Simulation model SWAP was validated for local conditions using measurements of soil water contents, aboveground biomass and yield of wheat. Considering two rather than one soil type for the experimental area resulted in different conclusions for both crops as to the effectivity of both conservation tillage and irrigation, demonstrating that a distinction of only one soil type results in misleading results. The validated model was applied to estimate yields considering climate change, focusing on the application of supplemental irrigation. Yields for “Mikii3” a durum wheat cultivar are expected to increase by appr. 14% in both soils due to climate change. More importantly, only 3 supplemental irrigations would be needed for the deep soil requiring 5% more water as compared with current climate trend, while the shallow soil needs 13 irrigations, corresponding with a need for 35% more water. This is highly significant from an economic point of view and supports the relevance of distinguishing two soil types. It was demonstrated the synergy of joint research by the agronomic and soil science community and the need for executing a soil survey in future when planning agronomic experiments, including a hydrological soil characterisation. 1. Introduction Soils play a crucial role in agricultural production systems by al- lowing water and nutrient uptake by roots, while providing physical support for plants to grow. Traditionally, crop growth on different soils has been characterized by correlating yield measurements, preferably made at different times during the growing season, with periodically measured soil characteristics in terms of water and nutrient contents. However, during the last two decades several operational simulation models have been developed for the soil-crop-water-nutrient system that allow a dynamic representation of crop growth as a function of environmental conditions, including soil processes and current and future climates (e.g. CERES, Ritchie and Otter, 1985; CropSyst, Stöckle et al., 2003; SWAP, Kroes et al., 2008). Even within farmer's fields, but certainly in a regional context, soil properties can vary considerably in space and this can be taken into account when running simulation models by using soil maps to dis- tinguish different soil types (e.g. Bouma and Bregt, 1989; Bonfante and Bouma, 2015). As will be discussed later in more detail, many agro- nomic experiments consider soils but not their heterogeneity and http://dx.doi.org/10.1016/j.agsy.2017.05.018 Received 12 August 2016; Received in revised form 22 March 2017; Accepted 27 May 2017 ⁎ Corresponding author. E-mail address: antonello.bonfante@cnr.it (A. Bonfante). Agricultural Systems 156 (2017) 67–75 0308-521X/ © 2017 Published by Elsevier Ltd. MARK