Although nitrogen fertilizer has been widely used in agriculture, inappropriate nitrogen management practices can adversely affect crop growth, yield, and the ecological environment. Excessive nitrogen application may lead to elevated nitrate concentrations in certain areas, resulting in non-point source pollution. Currently, optimal nitrogen management strategies for different crops remain unclear. This study conducted a long-term investigation into soil nitrogen dynamics and yield variations under current nitrogen application practices in maize and sunflower fields, and proposes suitable nitrogen management strategies for major crops in the Hetao Irrigation District of Inner Mongolia, China, to improve economic returns. A three-year field experiment was conducted from 2022 to 2024 in the Hetao Irrigation District of Inner Mongolia, China, to evaluate nitrogen dynamics and yield characteristics in these two crops. The experiment monitored nitrogen dynamics in maize and sunflower crops, the two major economic crops in the region. The DeNitrification-DeComposition (DNDC) model was employed to quantitatively analyze soil nitrogen dynamics and nitrogen balance for both crops. Scenario simulations evaluated crop yield, NOB--N uptake, NOB--N leaching, gas emissions and nitrogen partial factor productivity (NPFP) under varying nitrogen application rates and topdressing frequency conditions. The VIKOR (VlseKriterijuska Optimizacija I Kompromisno Resenje) model was used to identify optimal nitrogen-reduction strategies that maintain crop yields. Model performance was assessed using three evaluation metrics: R2, RMSE, and MRE. Results demonstrated that the DNDC model effectively captured soil nitrogen concentration (SNC) dynamics and corresponding yield outcomes across different growth stages for both maize and sunflower. NOB--N variations exhibited similar patterns between crops, though sunflower fields showed 40% higher residual NOB--N than maize fields. Maize NOB--N leaching accounted for 50.46%-62.20% of total nitrogen application, while sunflower fields exhibited 90.53 kg center dot ha-1 less NOB--N leaching compared to maize. Scenario simulations indicated that two topdressing applications could reduce nitrogen leaching by 16.93%-29.91% while maintaining maize yield, whereas increasing topdressing frequency showed no significant benefit for sunflower. The DNDC-VIKOR coupled optimization framework was used to determine the optimal nitrogen application schemes balancing crop productivity and ecological benefits: 230 kg center dot ha-1 with two topdressing applications for maize, and 192.5 kg center dot ha-1 with one topdressing application for sunflower. The DNDC-VIKOR coupled optimization framework developed in this study provides a scientific basis for formulating fertilization management strategies that balance economic efficiency and environmental protection. This method can be extended to other arid and semiarid irrigation districts and similar agricultural ecosystems for fertilization regime optimization research, offering reference for nitrogen management and sustainable production in irrigated agricultural regions worldwide.