Calibration and validation of the SALTMED model for simulating wheat response to saline irrigation in arid Northwestern China

Introduction

Roughly 20% of the world's irrigated land suffers from secondary salinisation-salt accumulation caused by irrigation with water that carries dissolved salts into the root zone faster than rainfall or drainage can flush them out [1] . In northwestern China's Hexi Corridor, groundwater salinity ranges from 1 to 15 dS m⁻¹, and over 340,000 hectares of irrigated farmland show visible salt damage [2] . Farmers have no alternative water source; the question is not whether to irrigate with saline water but how to manage it. Process-based models that simulate coupled water flow, solute transport, and crop growth under saline conditions are essential for answering that question [3] . SALTMED, developed at Cranfield University, integrates the FAO dual crop coefficient approach for evapotranspiration, a one-dimensional Richards's equation for soil water flow, convection-dispersion for solute transport, and a salinity-stress function for yield reduction [4] . It has been validated for tomato, maize, and quinoa in Mediterranean and arid environments but not for winter wheat under the extreme continental conditions of the Hexi Corridor, where winter soil freezing adds a complication absent from previous test sites [5,6] . This research calibrated and validated SALTMED for that purpose.

Theoretical background

SALTMED couples four sub-models. The water balance module partitions precipitation and irrigation into runoff, infiltration, evaporation, transpiration, and deep percolation using the FAO-56 dual crop coefficient method [7] . Soil water redistribution follows the Richards equation solved by finite differences on a 5 cm vertical grid. Solute transport is modelled by the convection-dispersion equation with a retardation factor for cation exchange. The crop response module uses the Maas-Hoffman threshold-slope function, where relative yield Y/Ymax = 1 for ECe below a threshold (ECt) and declines linearly at slope b above it [8] . For wheat, ECt = 6.0 dS m⁻¹ and b = 7.1% per dS m⁻¹.

Simulation parameters

Soil hydraulic parameters were measured by pressure-plate extraction (van Genuchten model): saturated hydraulic conductivity Ks = 24.6 cm day⁻¹, residual water content θr = 0.078, saturated water content θs = 0.43, and shape parameters α = 0.036 cm⁻¹ and n = 1.56 for the 0-40 cm layer. The model domain extended to 120 cm depth with three layers. Crop parameters included maximum root depth (100 cm), maximum LAI (5.2), base temperature (0 °C), and Kcb mid = 1.10. Initial soil salinity was set from pre-season ECe profiles. Time step was daily; spatial discretisation was 5 cm. Simulations ran from sowing (late September) to harvest (late June) for both seasons.

Material and methods

Material

The field trial was conducted at Shiyanghe Experimental Station, Wuwei, Gansu Province (37°52′N, 102°50′E, and 1,367 m a.s.l.). The climate is arid continental (BWk) with mean annual temperature 7.8 °C, precipitation 164 mm, and pan evaporation 2,021 mm. Soils are silt loam Aridisols. Winter wheat cv. Ningchun 4 was sown in plots of 4 × 5 m under six irrigation salinity levels: 0.8 (control), 2.0, 4.0, 6.0, 8.0, and 10.0 dS m⁻¹, prepared by dissolving NaCl and CaCl₂ (2:1 molar ratio) in canal water.

Methods

Each salinity level had four replicate plots in a randomised complete block design. Irrigation was applied by border flooding at 75% depletion of available water (monitored by TDR at 20, 40, 60, and 80 cm). Soil salinity (ECe) was measured from saturated paste extracts at 0-20, 20-40, and 40-60 cm at 15-day intervals. Grain yield was measured from the central 2 × 3 m at maturity. SALTMED was calibrated by adjusting Ks and dispersivity against the 2019-2020 salinity profiles, then validated without parameter adjustment against 2020-2021 data. Model performance was assessed by RMSE, normalised RMSE (nRMSE), R², and index of agreement (d).

Results

Table 1). The validation RMSE of 0.51 dS m⁻¹ represents less than 9% of the observed range, and the index of agreement (0.95) indicates excellent correspondence between simulated and observed temporal patterns.

Performance evaluation

A sensitivity analysis showed that the model was most sensitive to saturated hydraulic conductivity (Ks), which controls drainage rate and hence salt leaching. A 20% increase in Ks reduced simulated root-zone ECe by 11.4% and increased predicted yield by 6.8%. Dispersivity had a smaller effect (3.2% yield change per 20% parameter shift). The model slightly underestimated yield at the highest salinity, possibly because the linear Maas-Hoffman function does not capture osmotic adjustment mechanisms in salt-tolerant wheat genotypes [9] .

Discussion

The validation statistics (RMSE 0.51 dS m⁻¹, R² 0.91, d 0.95) compare favourably with previous SALTMED applications for tomato in Egypt (RMSE 0.63, R² 0.89 [5] ) and quinoa in Morocco (RMSE 0.47, R² 0.92 [6] ). The strong yield-salinity linearity above 2.0 dS m⁻¹ implies that farmers could use a simple threshold rule: every additional dS m⁻¹ of irrigation water salinity costs roughly 0.58 t ha⁻¹ of wheat, or about 8% of the control yield. The practical implication for Hexi Corridor farmers is that groundwater up to 4 dS m⁻¹ can sustain wheat yields above 5 t ha⁻¹ with proper leaching fractions, but beyond 6 dS m⁻¹ the economic viability of wheat drops sharply [10] . SALTMED can be used to optimise the leaching fraction under different salinity scenarios, helping water managers allocate scarce freshwater resources where they have the greatest marginal return [4] .

Conclusion

SALTMED reproduced root-zone salinity dynamics (RMSE 0.51 dS m⁻¹) and wheat grain yield (deviation 1.8-5.6%) under saline irrigation in arid northwestern China. Yield declined linearly at 0.58 t ha⁻¹ per dS m⁻¹ above a threshold of 2.0 dS m⁻¹, reaching 70.6% loss at 10 dS m⁻¹. The model provides a reliable platform for optimising irrigation scheduling and leaching management with saline groundwater in the Hexi Corridor and similar arid environments.

References

1. Qadir M, Quillerou E, Nangia V, Murtaza G, Singh M, Thomas RJ. Economics of salt-induced land degradation and restoration. Nat Resour Forum. 2014;38(4):282-295.
2. Li X, Kang E, Zhang J. Hydrological alterations and integrated water resources management in the Heihe River Basin, northwest China. Hydrol Sci J. 2006;51(6):11361151.
3. Minhas PS, Ramos TB, Ben-Gal A, Pereira LS. Coping with salinity in irrigated agriculture: Crop evapotranspiration and water management issues. Agric Water Manag. 2020;227:105832.
4. Ragab R. A holistic generic integrated approach for irrigation, crop and field management: The SALTMED model. Environ Model Softw. 2002;17(4):345-361.
5. El-Gamal E, Ragab R, Hellal FA, El-Hady MA. Modelling the impact of different water management scenarios on saline soils in Egypt using SALTMED model. Agric Water Manag. 2021;244:106577.
6. Hirich A, Choukr-Allah R, Ragab R. Prediction of climate change impact on corn grown in the south of Morocco using the SALTMED model. Irrig Drain. 2014;63(1):91-101.
7. Allen RG, Pereira LS, Raes D, Smith M. Crop evapotranspiration: Guidelines for computing crop water requirements. FAO Irrigation and Drainage Paper No. 56. Rome: Food and Agriculture Organization of the United Nations; 1998.
8. Maas EV, Hoffman GJ. Crop salt tolerance: Current assessment. J Irrig Drain Div. 1977;103(2):115-134.
9. Munns R, Tester M. Mechanisms of salinity tolerance. Annu Rev Plant Biol. 2008;59:651-681.
10. Kang S, Zhang L, Liang Y, Hu X, Cai H, Gu B. Effects of limited irrigation on yield and water use efficiency of winter wheat in the Loess Plateau of China. Agric Water Manag. 2002;55(3):203-216.
11. Ayers RS, Westcot DW. Water quality for agriculture. FAO Irrigation and Drainage Paper No. 29. Rome: FAO; 1985.
12. Rhoades JD, Kandiah A, Mashali AM. The use of saline waters for crop production. FAO Irrigation and Drainage Paper No. 48. Rome: FAO; 1992.
13. Letey J, Hoffman GJ, Hopmans JW, Grattan SR, Suarez DL, Corwin DL. Evaluation of soil salinity leaching requirement guidelines. Agric Water Manag. 2011;98(4):502-506.
14. Skaggs TH, Van Genuchten MT, Shouse PJ, Poss JA. Macroscopic approaches to root water uptake as a function of water and salinity stress. Agric Water Manag. 2006;86(1):140-149.
15. Steppuhn H, Van Genuchten MT, Grieve CM. Root-zone salinity. II. Indices for tolerance in agricultural crops. Crop Sci. 2005;45(1):221-232.
16. Ragab R, Malash N, Abdel Gawad G, Arslan A, Ghaibeh A. A holistic generic integrated approach for irrigation, crop and field management: 2. The SALTMED model validation using field data of five growing seasons. Agric Water Manag. 2005;78(1-2):89-107.