Simulating maize (Zea mays L.) growth and yield, soil nitrogen concentration, and soil water content for a long-term cropping experiment in Ontario, Canada

Abstract

Liu, S., Yang, J. Y., Drury, C. F., Liu, H. L. and Reynolds, W. D. 2014. Simulating maize (Zea mays L.) growth and yield, soil nitrogen concentration, and soil water content for a long-term cropping experiment in Ontario, Canada. Can. J. Soil Sci. 94: 435–452. A performance assessment of the Decision Support Systems for Agrotechnology Transfer (DSSAT) model (v4.5) including the CERES-Maize and CENTURY modules was conducted for continuous maize production under annual synthetic fertilization (CC-F) and no fertilization (CC-NF) using field data from a long-term (53-yr) cropping experiment in Ontario, Canada. The assessment was based on the accuracy with which DSSAT could simulate measured grain yield, above-ground biomass, leaf area index (LAI), soil inorganic nitrogen concentration, and soil water content. Model calibration for maize cultivar was achieved using grain yield measurements from CC-F between 2007 and 2012, and model evaluation was achieved using soil and crop measurements from both CC-F and CC-NF for the same 6-yr period. Good model–data agreement for CC-F grain yields was achieved for calibration (index of agreement, d=0.99), while moderate agreement for CC-NF grain yields was achieved for evaluation (d=0.79). Model–data agreement for above-ground biomass was good (d=0.83–1.00), but the model consistently underestimated for CC-F and overestimated for CC-NF. DSSAT achieved good model–data agreement for LAI in CC-F (d=0.82–0.99), but moderate to poor agreement in CC-NF (d=0.46–0.64). The CENTURY module of DSSAT simulated soil inorganic nitrogen concentrations with moderate to good model–data agreement in CC-F (d=0.74–0.88), but poor agreement in CC-NF (d=0.40–0.50). The model–data agreement for soil water content was moderate in 2007 and 2008 for both treatments (d=0.60–0.76), but poor in 2009 (d=0.46–0.53). It was concluded that the DSSAT cropping system model provided generally good to moderate simulations of continuous maize production (yield, biomass, LAI) for a long-term cropping experiment in Ontario, Canada, but generally moderate to poor simulations of soil inorganic nitrogen concentration and soil water content.

Long‐term Trends in Corn Yields and Soil Carbon under Diversified Crop Rotations

“…Results for CC (R 2 = 0.40 and EF = 0.35, Table 1) were better than previously reported for Woodslee by Liu et al (2010) (R 2 = 0.36 and EF = −0.70) with the Decision Support System for Agrotechnology Transfer (DSSAT)-Crop Environment Resource Synthesis (CERES)-Maize model. Simulations over shorter time periods (4-6 yr) have shown better agreement with observed yield for the Woodslee site (Liu et al, 2014;He et al, 2017). However, our RMSE values are slightly better than obtained with the crop model BioStar in Germany (2.1 Mg ha −1 ) for corn yield (Bauböck, 2014).…”

Comprehensively evaluating and modelling droplet diameters and kinetic energies of low‐pressure sprinklers

“…Figure 15 shows comparisons between the measured and simulated results obtained by the three specific power models for the three sprinkler types with three nozzle diameters and two operating pressures. Apparently, the fitting accuracy of Model 2 was the best among the three models, with RMSE and NRMSE values reaching 0.011 W m À2 and 29%, respectively, corresponding to an appropriate level of model accuracy (Liu et al, 2014). Consequently, Model 2 proposed in this study is relatively reliable and worth recommending for predicting the radial distribution of specific power under various operating pressures and nozzle sizes for lowpressure sprinklers.…”

Using the DSSAT Model to Support Decision Making Regarding Fertilizer Microdosing for Maize Production in the Sub-humid Region of Benin

“…The calibrated model was evaluated against the phenology, LAI, and above-ground biomass time series, final grain and biomass yield data from the remaining treatments in 2014 and all treatments in 2015. The accuracy of model simulations was assessed based on the predicted deviation (PD, difference between the predicted and observed values in %), root meansquare error (RMSE) (Willmott et al, 1985), relative RMSE (RRMSE), index of agreement (d), and coefficient of efficiency (E1) (Liu et al, 2014). For time series data, the performance indicators were calculated across all measurement dates.…”