Crop growth is influenced by the energy partition and water–heat transfer in the soil and canopy, while crop growth affects the land surface energy distribution and soil water-heat dynamics. In order to simulate the above processes and their interactions, a new model, named CropSPAC, was developed considering both the growth of winter wheat and the water–heat transfer in Soil-Plant-Atmosphere Continuum (SPAC). In CropSPAC, the crop module depicts the dynamic changes of leaf area index (LAI), crop height, and the root distribution and outputs them to the SPAC module, while the latter outputs soil moisture conditions for the crop module. CropSPAC was calibrated and validated by field experiment of winter wheat in Yongledian, Beijing, with five levels of irrigation treatments, namely W0 (0 mm), W1 (60 mm), W2 (110 mm), W3 (170 mm), and W4 (230 mm). Results show that CropSPAC could predict the soil water and temperature distribution, and winter wheat growth with acceptable accuracy. For example, for the 0–1 m soil water storage, the R2
for W0, W1, W2, W3, and W4 is 0.90, 0.88, 0.90, 0.91, and 0.79, and the root mean square error (RMSE) is 17.24 mm, 27.65 mm, 20.47 mm, 22.35 mm, and 12.88 mm, respectively. For soil temperature along the soil profile, the R2
ranges between 0.96 and 0.98, and the RMSE between 1.22 °C and 1.94 °C. For LAI, the R2
varied from 0.76 to 0.96, and the RMSE from 0.52 to 0.67. We further compared the simulation results by CropSPAC and its two detached modules, i.e., crop and the SPAC modules. Results demonstrate that the coupled model could better reflect the interactions between crop growth and soil moisture condition, more suitable to be used under deficit irrigation conditions.