Evaluation of saline water irrigation on cotton growth and yield using the AquaCrop crop simulation model

It is imperative to evaluate the practicality of brackish water irrigation for alleviating the conflict between freshwater resources shortages and food production. Four years experiments were carried out on cotton from 2012 to 2015 seasons with four salinity levels in irrigation water, including 1.3 dS m-1 (CK), 5.4 dS m-1 (S1), 8.8 dS m-1 (S2), and 12.4 dS m-1 (S3) in lowland plain of Hebei in the North China Plain (NCP). Parameters of the FAO AquaCrop model were calibrated with a dataset collected in the 2012 season, and the model was validated with independent data from the next three seasons. Under the condition of saline water irrigation, the simulation of canopy cover, soil water content, soil electric conductivity, biomass, and yield was accurate (the coefficient of determination (R2) > 0.65, the index of agreement (d) > 0.89). The scenario simulations with different salinity levels in irrigation water were run in growing degree days mode in AquaCrop with climatic data from 1958 to 2010. In the high (483 mm) and normal (430 mm) rainfall years, salt was drained out of the soil profile. Whereas, salt was accumulated in the soil profile under the scenario of S2 and S3, while was leaving out the root zone in the scenario of CK and S1 in the low (324 mm) rainfall years. With the salinity of 4.5 dS m-1, the difference between salt inside and outside of the soil profile tends to be zero. Based on the scenario simulations, the relationships between crop evapotranspiration (ETa) and cotton yield and between ETa and water productivity were fitted with a quadratic curve. It is concluded from the scenario simulations that there are real risks in soil salinization for cotton production in the lowland plain of Hebei in the NCP due to saline water irrigation.