Abstract: In the traditional quality control of meteorological data, it is using statisti cal method to set up control threshold limit. The data regarded as suspect data is usually labeled or regarded as gross data to be deleted if it is outside the controlled threshold. The monthly mean air temperature data of Tuotuo River Stat ion in Qinghai province from October 1985 to March 1986 is the most extremely abnormal one of base stations from 1971 to 2000 in China. As an illustration, complex quality control and analysis method of extremely abnormal meteorological data is presented which perhaps reflects extreme weather event or extreme climat e event. By utilizing complex quality control method，namely consulting Metadata , surrounding stations tests and consistency checks between various observationa l programs, the extreme monthly mean air temperature data of Tuotuo River Statio n from October 1985 to March 1986 is proved to be correct. It reflects the extre me climate event. Thereby, it can be conclude that the so called gross data out side the controlled area perhaps reflects extremely abnormal weather or climate event and is probably correct, though it can hardly come forth according to stat istical probability. So the more advanced analysis to extremely abnormal meteoro logical data should be carried out by complex quality control method in order to retain the fact of extreme weather or climate event. In the end, it is presente d the complex quality control results about the extremely abnormal monthly mean air temperature data of base stations in China from 1971 to 2000.