Increasing use of manure nutrients in crop production and decreasing the input of chemical fertilizers, is a promising way to achieve a more circular economy and more sustainable food systems. The integrated effects of amended composting manure with additives and a 50% substitution of chemical fertilizer nitrogen (N) input by amended manure N input to cropland were proved to halve ammonia (NH3) emissions from an integrate manure-cropland system. However, the accompanied impacts of these measures on greenhouse gas (GHG) emissions from the integrate manure-cropland system have not been well quantified. Here, we examined the effects of improved manure storage and treatment and its usage at field on GHG emissions from a system perspective, based on a two-year experiment in the North China Plain (NCP). Manure storage and treatment included natural storage (NS) and composting with straw, with or without amendment by zeolite (Z) and superphosphate (SP). Manure products were applied as basal fertilizer for both crops. Results showed that CH4 was the dominant GHG source in the natural storage treatment, whereas N2O dominated in the manure composting treatments. Amending manure with straw, Z and SP stimulated 15–48 times N2O emission and decreased CH4 emissions by 82%–91% during storage and treatment. Replacement of 50% synthetic fertilizer N by manure N decreased total system GHG emissions by 25%–34%. Differences between composted manure treatments in total system GHG emissions were relatively small, although comparatively higher in amended manure treatments; however, they promoted greater soil organic carbon sink activity, resulting in no significant difference in net global warming potential (GWP) and greenhouse gas emission intensity (GHGI). Application of amended manure produced high nitrogen use efficiency (NUE) in the integrated crop–manure system. The NH3 mitigation approach of replacing 50% of synthetic fertilizer N by amended manure N synergistically decreased total GHG emissions, increased system NUE and also increased the net economic benefit by 5%–11% over the 2 years of the experiment.