Agricultural Water Management 163 (2016) 90–99 Contents lists available at ScienceDirect Agricultural Water Management journal homepage: www.elsevier.com/locate/agwat Coupling and testing a new soil water module in DSSAT CERES-Maize model for maize production under semi-arid condition Hamze Dokoohaki a,1 , Mahdi Gheysari a,∗ , Sayed-Farhad Mousavi b , Shahrokh Zand-Parsa c , Fernando E. Miguez d , Sotirios V. Archontoulis d , Gerrit Hoogenboom e a Water Engineering Department, College of Agriculture, Isfahan University of Technology, Isfahan 84156-83111, Iran b Civil Engineering Department, Semnan University, Semnan 35131-19111, Iran c Water Engineering Department, College of Agriculture, Shiraz University, Shiraz, Iran d Agronomy Department, Iowa State University, Ames, IA, USA e AgWeatherNet Program, Washington State University, Prosser, WA 99350, USA a r t i c l e i n f o Article history: Received 27 December 2014 Received in revised form 1 September 2015 Accepted 2 September 2015 Keywords: DSSAT SWAP Hybrid model Maize CERES-Maize-hbased a b s t r a c t Process-oriented crop simulation models are valuable tools for representing our understanding of the current and future states of a cropping system. The main objective of this research was to couple the Crop- ping System Model-(CSM)-Crop-Environment Resource Synthesis (CERES)-Maize (CSM-CERES-Maize) with the Soil, Water, Atmosphere, and Plant (SWAP) model in order to benefit from the advantages of both models. A new model was developed by replacing a simplified version of the SWAP with WatBal and SPAM modules of the Decision Support System for Agrotechnology Transfer (DSSAT) version 4.0. In this hybrid model, the CERES-Maize supplied the SWAP model with plant growth variables. Meanwhile, the SWAP model provided the CERES-Maize model with daily evapotranspiration, soil water content, and root water uptake. The model was then validated with a dataset including four irrigation levels (two deficit levels along with one full and one over-irrigation level), and three nitrogen levels (0, 150, and 200 kg/ha nitrogen) obtained from a field experiment in 2003 and 2004. The root mean square errors (RMSE) across all treatments in the simulation of final biomass were, respectively, 1175 and 2148 kg/ha in the first year and 1274 and 1514 kg/ha in the second year for the hybrid and original version of CERES- Maize model. Average RMSE for two non-water stress treatments was 1.29 and 1.35 cm in the simulation of soil water content for hybrid and original models, respectively. In general, our findings indicated that the new hybrid model was fairly successful in biomass simulation, which was due to better soil water simulations of all four irrigation levels except severe deficit irrigation. © 2015 Elsevier B.V. All rights reserved. 1. Introduction Crop models facilitate the clarification and evaluation of multi- dimensional relationships between factors affecting crop growth, development, and yield. These factors include planting date, cul- tivar selection, seeding rates, soil type, fertilizer and irrigation strategies, and seasonal weather patterns (Yang, 2008). For the past several decades, researchers have applied crop models to under- stand, organize, and develop new ideas and to analyze different ∗ Corresponding author. Fax: +98 31 33912254. E-mail addresses: hamzed@iastate.edu (H. Dokoohaki), gheysari@cc.iut.ac.ir (M. Gheysari), mousavi sf@yahoo.com (S.-F. Mousavi), zandparsa@yahoo.com (S. Zand-Parsa), femiguez@iastate.edu (F.E. Miguez), sarchont@iastate.edu (S.V. Archontoulis), gerrit.hoogenboom@wsu.edu (G. Hoogenboom). 1 Current address: Agronomy Department, Iowa State University, Ames, IA, USA. management practices (Jiang et al., 2011; Dogan et al., 2006). Due to their proven value in environmental and agricultural resource management and policy-making, these models currently play a critical role in agricultural systems (Ma et al., 2005). Crop models employ simple or complex approaches to simulate environmental processes based on their objectives and data availability. The Deci- sion Support System for Agrotechnology Transfer (DSSAT) (Jones et al., 2003; Hoogenboom et al., 2012), the Agricultural Production Systems sIMulator (APSIM) (McCown et al., 1995; Keating et al., 2003), the Root Zone Water Quality Model (RZWQM) (Ahuja et al., 2002), and the Soil, Water, Atmosphere, and Plant (SWAP) (Van Dam et al., 1997) are examples of popular simulation models world- wide. While SWAP model is an agrohydrological model mainly focusing on soil water (Van Dam and Feddes, 2000), the Crop Envi- ronment Resource Synthesis-Maize (CERES-Maize) model from the http://dx.doi.org/10.1016/j.agwat.2015.09.002 0378-3774/© 2015 Elsevier B.V. All rights reserved.