Simulating the effects of subsurface drainage on winter wheat yield in coastal reclamation areas of Jiangsu, China

Abstract: Shallow groundwater and poor drainage condition often cause yield reduction of winter wheat due to waterlogging stress in the coastal reclamation area of Jiangsu Province, China. As the subsurface drainage construction has been advocated more in recent years to provide timely soil drainage and improve farming mechanization level for modern agricultural development, this study aims to examine the effect of subsurface drainage on waterlogging control and the subsequent effect on winter wheat production, the field hydrology model - DRAINMOD was used in combination with the crop growth model- AquaCrop to predict the effect of subsurface drainage system layout on winter wheat yield in the study area. The more accurate predictions of water table depth by DRAINMOD model was used to improve calculations of crop uptake from shallow groundwater in drained fields, which is currently simplified in the AquaCrop model. Both models were tested based on field observations of soil, weather, water table depth and winter wheat production in the Dongtai experimental station of the Jiangsu Academy of Water Conservancy Sciences in China. The model testing results showed that DRAINMOD accurately predicted groundwater depth in the study area, and the winter wheat yields predicted by AquaCrop was close to the measured values. Because winter wheat growth relied on groundwater uptake from the shallow water table in the study area, the accurate groundwater depth input to AquaCrop model is critical to crop yield predictions. After verification of both models, DRAINMOD was used to predict the daily water table depth over a long term period, considering the great variations in the water table depth over the different hydrological years. Different rainfall frequencies of 10%, 15%, 50%, and 85% during the wheat growing season were selected to predict drainage needs for building high standard farmland in the study area. The predicted groundwater depth under different subsurface drainage conditions were fed into the AquaCrop model for yield prediction. DRAINMOD simulation results showed that crop water stress (as expressed by <30 cm depth to water table in 3 days) increased with the drain spacing but decreased with the drain depth during the wheat growing season; subsurface drainage layout of 1.2 m deep and 20 m spacing, or 1.5 m deep and 30 m spacing may achieve water stress free in 85% years. The AquaCrop predicted wheat yield was generally higher in the dry and average years owing to the less water stress, the predicted crop yields varied over years, and the effect of subsurface drainage on yield increase was more obvious in the wet years. With shallower depth of subsurface drainage pipes (80-100 cm), DRAINMOD predicted groundwater depth was generally maintained at a high level, rainfall recharge rapidly raised the groundwater level and crop yields were affected by the waterlogging stress in 20% years. The wet stress in the wheat growing season can be reduced significantly with the increase of buried depth of subsurface drainage pipes, the predicted waterlogging stress occurred in less than 15% years when the drain depth was at 1.5 m and the drain spacing was less than 50 m; when drain spacing was reduced to less or equal to 20 m, the waterlogging stress on the crop yield can be reduced to in 2% years. In conclusion, the subsurface drainage layout of 1.5 m depth and 20 m spacing can achieve the ideal wheat production goal in 85% years in the study area based on simulation results with DRAINMOD and AquaCrop. Findings from this research may provide technical support for subsurface drainage system construction to improve agricultural production in the coastal reclamation area in eastern China and the similar regions.