Accurate identification of key parameters for data assimilation is important in simulating the planetary boundary layer height (PBLH) and structure evolution in numerical weather prediction models. In this study, surface observational data and lidar-derived PBLH on 42 cloudless days from June 2007 to May 2008 are used to quantify the statistical relationships between surface parameters and the PBLH at a semiarid climate observational site in Northwest China. The results indicate that surface upward long wave radiation, surface temperature, and surface sensible heat fluxes show strong correlations with the PBLH with correlation coefficients at a range of 0.63–0.72. But these parameters show varying correlation response time to the different stages of PBL development. Furthermore, the air temperature shows the highest correlation with the PBLH near the surface and the correlation decreases with increasing height.