T A Single Column Model (SCM) for Global and Regional Atmospheric Prediction Enhanced System (GRAPES) is constructed for the purpose of evaluating physical process parameterizations. Two observational datasets including Wangara and the third Global Energy and Water Cycle Experiment Atmospheric Boundary Layer Study (GABLS-3) SCM field observations have been applied to evaluate this SCM. By these two numerical experiments, the GRAPES_SCM is verified to be correctly constructed. Furthermore, the interaction between the land surface process and atmospheric boundary layer (ABL) is discussed through the second experiment. It is found that CASE3 (CoLM land surface scheme coupled with ABL scheme) simulates less sensible heat fluxes and smaller surface temperature which corresponds with its lower potential temperature at the bottom of the ABL. Moreover, CASE3 simulates turbulence that is weaker during the daytime and stronger during nighttime, corresponding with its wind speed at 200 m which is bigger during daytime and smaller during nighttime. However, they are generally opposite in CASE2 (SLAB coupled with ABL). The initial profile of the water vapor mixing ratio is artificially increased by the experiment setup which results in the simulated water vapor mixing becoming wetter than actually observed. CASE1 (observed surface temperature taken as lower thermal forcing) and CASE2 have no soil moisture prediction and simulate a similar water vapor mixing ratio, while CASE3 has a soil moisture prediction and simulates wetter. It is also shown that the time step may affect the stabilization of the ABL when the vertical levels of the SCM are fixed.