Research on Farmland Drainage Pollutant Reduction Techniques in Hetao Irrigation District Under Ecological Drainage Ditch Conditions

To alleviate the eutrophication in the Wuliangsuhai watershed and evaluate the pollutant reduction performance of ecological drainage ditches in the Hetao Irrigation District, a controlled field simulation experiment was conducted using synthetic agricultural return-flow water formulated from long-term monitoring data. Three leguminous plant treatments, two microbial substrate treatments, and one control were established to compare the migration and transformation of total nitrogen (TN), total phosphorus (TP), and chemical oxygen demand (COD) in overlying water, sediment, and plants under different hydraulic retention time intervals (0–6 h, 6–12 h, and 12–18 h). The results showed that plant treatments generally improved conventional water quality indicators, with increased pH and dissolved oxygen (DO) and decreased electrical conductivity, salinity, and total dissolved solids, whereas microbial substrate treatments tended to reduce DO. Pollutant reduction performance differed among treatments. Medicago sativa showed the strongest TN removal from overlying water, Microbial biological rope exhibited the best TP removal from overlying water, and Melilotus suaveolens performed best in COD reduction. Among all plant treatments, Astragalus laxmannii exhibited the most stable overall performance and a relatively strong integrated capacity for nitrogen and phosphorus retention. Most TN and TP reduction in overlying water and sediment occurred during the initial hydraulic retention time interval of 0–6 h, whereas TN plant uptake became more evident during 12–18 h. These findings suggest that ecological drainage ditches vegetated with locally adapted leguminous species have potential to mitigate agricultural non-point source pollution in arid irrigation districts. In particular, Astragalus laxmannii appears to be a promising candidate for ecological ditch design in the Hetao Irrigation District. However, this study was conducted under controlled synthetic return-flow conditions rather than with actual field drainage water, and no tracer-based hydrodynamic verification was performed; therefore, the reported hydraulic retention time effects and treatment efficiencies should be interpreted cautiously. Further field-scale validations under real drainage, seasonal variation, and long-term operation conditions are still needed.