Water management of soybeans (Glycine max L.) in clay soils in the Canadian Prairies

Abstract

This study aimed to (i) assess the effects of subirrigation and controlled drainage on soybean yield and yield components, actual crop evapotranspiration, yield response factor (Ky), harvest index (HI), soil salinity, and water table depth, (ii) develop a Drainmod model of the field to assess the benefits of controlled drainage and subirrigation on soybean from 2010 to 2024 in heavy clay soils, and (iii) develop Hydrus-2D model of the field to simulate soil water redistribution under three scenarios, i.e., pulsed subirrigation with rainfall, pulsed and continuous subirrigation under no rainfall. Field experiments were conducted in the 2023, 2024, and 2025 growing seasons in the Prairies East Sustainable Agriculture Initiative (PESAI) Research Farm at Arborg, Manitoba, Canada. In 2023 and 2025, subirrigation was applied to mitigate drought conditions, while in 2024, being a wet year, controlled drainage was implemented to manage wet and dry conditions. Soybean yield, above and below ground biomass, number of pods and seeds per pod, nodule counts, and height were significantly higher (p < 0.05) in the subirrigated treatments compared to the rainfed control. Soybean seed yield on-the-tile treatment was significantly higher at 2820.9 kg ha-1 compared to 2216.9 kg ka-1 in the no drain control treatment in 2023. In the 2024 growing season, there was no significant difference in yield between the treatments. In the 2025 growing season, soybean seed yield was significantly higher at 1503.2 kg ka-1 on-the-tile treatment compared to 1240.9 kg ka-1 obtained in the no drain treatment. Seasonal ETa calculated using the soil-water-balance method averaged 313.8, 339.4, and 176.0 mm on the drainage system in 2023, 2024, and 2025, respectively, while it was 204.7, 335.4, and 127.7 mm in the no-tile treatment in the respective years. Results showed the Drainmod as a reliable model in predicting water table variation under controlled drainage and subirrigation management practices with an overall coefficient of determination (R2) of 0.96, index of agreement (D) of 0.98, root mean square error (RMSE) of 0.11 m, and mean bias error (MBE) of -0.03 m. On average, controlled drainage in wetter growing seasons resulted in 10.7% higher relative yield compared to no tile, while the relative yield under subirrigation was 20.5% higher than the one on no tile treatment. Hydrus-2D model was reliable in simulating soil water dynamics at different depths in soybean production under subirrigation through tile drains in clay soils. Average soil water content at 20 cm depth across the tiles increased from 32.4 to 35.5% from 17 to 36 days after emergence (DAE) in pulsed subirrigation with inadequate precipitation to meet the crop water demand. However, under no precipitation conditions, when subirrigation is done as continuous or pulsed, the average soil water content decreased from 32.1 to 31.0% across the field from 17 to 36 DAE at 20 cm depth showing the importance of antecedent soil moisture conditions in redistribution of water towards midway between tiles. Under no rainfall conditions, results indicated pulsed subirrigation to be the best alternative for water application as it showed more soil water content uniformity across the tile drains. Overall, water management strategies i.e. controlled drainage in wetter growing seasons and subirrigation in drier periods increased soybean yield.