It has been reported that root characteristics and canopy structure in an agroforestry system directly change the availability of underground and aboveground resources and climatic conditions and seriously affect crop growth and yield (Dufour et al. in J Agron Crop Sci 199: 217–227, 2013; Bouttier et al. in Agroforest Syst 88: 693–706, 2016; Farooq et al. in Appl Ecol Env Res 17: 3553-3566, 2019). An experiment was conducted in northwest China to explore the distribution and characteristics of root systems for alfalfa grown in a poplar/alfalfa silvopastoral system. Field sampling was conducted every 15 days in a 3-yr-old alfalfa + 7-yr-old poplar silvopastoral system, a 3-yr-old sole-alfalfa system, and a 7-yr-old sole-poplar system. The results showed that intercropping reduced the RLD of alfalfa, while the FSRL of intercropped alfalfa was higher than that of sole-cropped alfalfa. In terms of the agronomic traits of alfalfa (plant height, leaf area index, leaf-stem ratio, growth rate), intercropping traits were significantly lower than sole-cropped alfalfa, and the total hay yield during intercropping was 55.27% lower than that of sole-cropped alfalfa. In terms of field microclimate, the dew point temperature and wind speed of sole-cropped alfalfa were 70.68% and 93.83% higher than that of intercropping, but the air humidity of sole-cropped alfalfa was 13.49% lower than that of intercropping, and there was no significant difference in air temperature. The crude protein (CP), crude fat (CF), and relative feed value (RFV) of intercropping alfalfa were 17.78%, 12.33%, and 17.26% higher than sole-cropped alfalfa, respectively. Although the agronomic traits and yield of alfalfa decreased with intercropping, intercropping could improve the microclimate and the quality of alfalfa significantly. The land equivalent ratio (LER = 1.42) in the poplar/alfalfa silvopastoral system was > 1, indicated that the system in advantageous.