Abstract. Mixing layer height (h) is an important parameter for understanding the transport process in the troposphere, air pollution, weather and climate change. Many methods have been proposed to determine h by identifying the turning point of the radiosonde profile. However, substantial differences have been observed in the existing methods (e.g. the potential temperature (θ ), relative humidity (RH), specific humidity (q) and atmospheric refractivity (N) methods). These differences are associated with the inconsistency of the temperature and humidity profiles in a boundary layer that is not well mixed, the changing measurability of the specific humidity and refractivity with height, the measurement error of humidity instruments within clouds, and the general existence of clouds. This study proposes a method to integrate the information of temperature, humidity and cloud to generate a consistent estimate of h. We apply this method to high vertical resolution (∼ 30 m) radiosonde data that were collected at 79 stations over North America during the period from 1998 to 2008. The data are obtained from the Stratospheric Processes and their Role in Climate Data Center (SPARC). The results show good agreement with those from N method as the information of temperature and humidity contained in N ; however, cloud effects that are included in our method increased the reliability of our estimated h. From 1988 to 2008, the climatological h over North America was 1675 ± 303 m with a strong east-west gradient: higher values (generally greater than 1800 m) occurred over the Midwest US, and lower values (usually less than 1400 m) occurred over Alaska and the US West Coast.