Returning crop residues to soil is an effective approach for sustaining organic matter concentrations and increasing nutrient availability in soils. A 2-year micro-plot field experiment was conducted in dryland to determine the green manure, straw, and fertilizer nitrogen (N)-15 uptake by wheat, their residual N in soil and losses; the effect of straw application on the fate of N from green manure and vice versa was also determined, as well as the effect of crop residue additions on the fate of fertilizer N. All the micro-plots were treated with the same amount of 15N-labeled or unlabeled fertilizer. The green manure N uptake by wheat, residual N, and N loss were 22.4, 51.7, and 25.9 % of the total added green manure N over the 2-year experiment. Straw addition significantly decreased the green manure residual soil N but increased the cumulative losses. The straw N taken up by wheat, residual N in soil, and N loss were 8.3, 31.0, and 60.7 %, respectively. Green manure addition significantly decreased the straw N taken up by wheat, increased the residual soil N, and reduced the N loss. Furthermore, the fertilizer N taken up by wheat, residual N in soil, and N loss were 32.4, 32.3, and 35.2 %, respectively. Crop residue additions significantly increased the uptake of fertilizer N by wheat in the second year. The application of inorganic N fertilizer in combination with appropriate crop residues may be an effective approach to improve the long-term fertilizer N use efficiency, soil quality, and crop yield in wheat–summer fallow rotation systems in dryland.