In this study, we employed a constructed wetland-microbial fuel cell (CW-MFC) for the treatment of agricultural non-point source pollution, and the impacts of CW-MFC, constructed wetland (CW), and microbial fuel cell (MFC) on agricultural non-point source pollution processing performance, electrical performance, and changes in the microbial community within the system were compared and analyzed to explore the synergistic effect of CW-MFC coupling. The findings indicated that the coupling of CW and MFC effectively enhanced the processing performance, electrical performance, and enzyme activity of CW-MFC. The electric field’s promotion of the removal rate of chemical oxygen demand (COD) and total nitrogen (TN) was greater than that of plants, while the electric field’s promotion of the removal rate of total phosphorus (TP) was smaller than that of plants. The promoting effect on the removal rate of atrazine was greater than that of plants at a low concentration of atrazine, but smaller than that of plants at medium and high concentrations of atrazine. The promoting effect of the electric field on urease (UE) and catalase (CAT) activities was generally greater than that of plants, while the promoting effect on alkaline phosphatase (ALP) activity was greater in plants. Affected by the coupling of CW and MFC, the microbial community diversity of CW-MFC and the abundance of bacterial phyla, classes, and genera such as Acidobacteria, Deltaproteobacteria, Aquabacterium, Geobacter, and Subgroup_6 were detected. The abundance of terpenoids and polyketides increased to varying extents. The results of this study laid a solid theoretical foundation for the advancement of agricultural non-point source pollution control technology and provided strong support for the expansion of CW to new application fields. It has profound significance for effectively controlling agricultural non-point source pollution and ensuring the sustainable development of agriculture.

Cool-climate maize fields are characterized by low soil temperatures, strong seasonal hydrothermal fluctuations, and peat-influenced soil profiles, which may lead to patterns of nitrogen (N) cycling distinct from those in conventional agricultural soils. During maize growth, soils from three depths were characterized using physicochemical measurements, N-transformation and enzyme-activity assays, metagenomic sequencing, Mantel tests, variation partitioning analysis, and partial least squares path modeling (PLS-PM). Soil environmental factors varied significantly over time and with depth; soil organic matter (SOM) and total nitrogen (TN) increased with depth, while ammonium nitrogen (NH₄⁺-N) predominated early and nitrate nitrogen (NO₃⁻-N) predominated during the middle and late growth stages. The nitrogen fixation rate (NFR), nitrification rate (NitR), and denitrification rate (DNR) all peaked in August and showed a spatial pattern characterized by nitrogen fixation in the deepest layer and denitrification in the upper and middle layers. Bacterial communities varied less spatiotemporally than fungal communities. The genes nifK, hao, nirS/nirK, NR, nrfC, and hzsA/hzsC were identified as key nitrogen-cycling functional genes. Mantel tests and PLS-PM further characterized these relationships, with PLS-PM showing that soil physicochemical properties were positively associated with bacterial community composition (β = 0.87, p < 0.01), which, in turn, was negatively associated with N-cycling functional genes (β = −0.97, p < 0.001). Together, these pathways were associated with variation in N-cycling processes. Overall, this study advances an integrated understanding of N-cycling patterns and their potential controls in cool-climate maize fields and provides a scientific basis for optimizing N management strategies.