Field-scale modeling of root water uptake and crop growth in a tropical scenario

Abstract

Context

Modeling root water uptake (RWU) coupled with crop growth processes is crucial for optimizing water management in field crops, especially in tropical scenarios with high water demand. However, field-scale assessments of process-based RWU models are limited, as most efforts have focused on the root system scale, leading to few practical applications.

Objective

To evaluate the performance of the SWAP agro-hydrological model in simulating and predicting crop growth, soil water balance components, and water transfer processes in the soil-plant-atmosphere continuum, using field observations and long-term simulation results of soybean and wheat cultivation in a tropical scenario.

Methods

Measurements of crop growth variables, soil water content, and evapotranspiration were used to calibrate parameters related to crop growth, soil hydraulic properties, and RWU, and to evaluate the performance of the SWAP model in replicating three irrigated field experiments with soybean and wheat conducted in Brazil. Model performance was assessed using Root Mean Square Error (RMSE) and Nash-Sutcliffe Efficiency (NSE) statistics. After validation, long-term simulations (1978–2022) were performed to evaluate crop yield, irrigation requirements, and water productivity under a rainfed scenario and two irrigated scenarios with different irrigation triggering criteria.

Results

The SWAP model, incorporating a process-based RWU function (MFlux), effectively simulated crop growth, water use, and specific RWU mechanisms on a daily time step. After model calibration and evaluation, the RMSE values were ≤ 421 kg ha⁻¹ for grain yield, ≤ 0.029 cm³ cm⁻³ for soil water content, and ≤ 1.07 mm d⁻¹ for evapotranspiration, with mostly positive NSE values for all variables. Seasonal analyses of water management scenarios indicated that irrigation amounts were reduced by an average of 41 mm for wheat, while water productivity of irrigation increased by an average of 61 % for soybean when irrigation was triggered based on relative evapotranspiration instead of total available water depletion.

Conclusion

The SWAP/MFlux model is suitable for simulating crop growth and soil-vegetation-atmosphere water transfer processes at the field scale in tropical soybean and wheat cultivation. Long-term analyses suggest that efficient water management, which maintains high crop yields while minimizing irrigation, depends on an irrigation criterion based on relative evapotranspiration.

Implications

This study presents a process-based approach for predicting RWU coupled with crop growth processes at the field scale. This model provides a robust framework for optimizing water management in field crops, enabling a comprehensive evaluation of management practices through detailed simulation of key mechanisms.