In this paper, 6 h surface pressure observation and T639 analysis field in the summer (from June to August) 2013 in five different regions are analyzed firstly by wavelet and spectrum methods. Then, a progressive OMB (observationminusbackground) quality control method is put forward, and the surface pressure observations in China and the neighboring regions are studied. The research results indicate that in the regions with different elevations between surface observation sites and numerical model surface, the cycle and amplitude characteristics can be described accurately after the background data are corrected by the barometric height method. By using the barometric height formula and weighted average method, we correct surface pressure from model height to observation sites, making a good foundation for the quality control of surface pressure observation. Progressive OMB quality control method makes the observation increment very close to normal distribution. The progressive OMB quality control method is better than the OMB quality control method, because it can effectively identify outlier data and guarantee the quality of remained data.