Although China has achieved great advancements toward national food security, the country is still confronted with a range of challenges, including natural resource stress, imbalanced diets and environmental pollution. Optimized management of crop–livestock systems is the key measure to realize agricultural green transformation. However, optimized management of crop–livestock systems that use multi-objective zoning is lacking. This study employed a multi-objective zoning management approach to comprehensively analyze four indicators: ammonia volatilization, nitrogen surplus, soil carrying capacity and ecological red line area. With its significant ecological integrity and a strong emphasis on sustainability, the Baiyangdian Basin serves as a unique and suitable test case for conducting analyses on multi-objective nutrient optimization management, with the aim to facilitate the agricultural green transformation. This study finds that less than 8% of the area in the Baiyangdian Basin meet the acceptable environmental indicator standard, whereas around 50% of the area that had both nitrogen surplus and ammonia volatilization exceeded the threshold. Implementation of unified management, that is, the same management technique across the study areas, could result in an increase of areas meeting environmental indicator thresholds to 21.1%. This project developed a novel multi-indicator partition optimization method, in which distinct measures are tailored for different areas to satisfy multiple environmental indicators. Implementation of this method, could potentially bring more than 50% area below the threshold, and areas with ammonia emissions and nitrogen surplus could be reduced to 15.8%. The multi-indicators partition optimization method represents a more advanced and efficiency-oriented management approach when compared to unified management. This approach could be regarded as the best available option to help China achieve agricultural transformation to improve efficient production and reduce environmental pollution. It is recommended that current policies aimed at nutrient management toward sustainable agricultural development should shift toward the application of multi-indicators partition optimization.