Main content area

Including root architecture in a crop model improves predictions of spring wheat grain yield and above‐ground biomass under water limitations

Mboh, Cho Miltin, Srivastava, Amit Kumar, Gaiser, Thomas, Ewert, Frank
Journal of agronomy and crop science 2019 v.205 no.2 pp. 109-128
aboveground biomass, crop models, field crops, grain yield, prediction, root growth, root systems, spring wheat, water stress, Germany
Although the root length density (RLD) of crops depends on their root system architecture (RSA), the root growth modules of many 1D field crop models often ignored the RSA in the simulation of the RLD. In this study, two model set‐up scenarios were used to simulate the RLD, above‐ground biomass (AGB) and grain yield (GY) of water‐stressed spring wheat in Germany, aiming to investigate the impact of improved RLD on AGB and GY predictions. In scenario 1, SlimRoot, a root growth sub‐model that does not consider the RSA of the crop, was coupled to a Lintul5‐SlimNitrogen‐SoilCN‐Hillflow1D crop model combination. In scenario 2, SlimRoot was replaced with the Somma sub‐model which considered the RSA for simulating RLD. The simulated RLD, AGB and GY were compared with observations. Scenario 2 predicted the RLD, AGB and GY with an average root mean square error (RMSE) of 0.43 cm/cm³, 0.59 t/ha and 1.03 t/ha, respectively, against 1.03 cm/cm³, 1.20 t/ha and 2.64 t/ha for scenario 1. The lower RMSE under scenario 2 shows that, even under water‐stress conditions, predictions of GY and AGB can be improved by considering the RSA of the crop for simulating the RLD.