This paper will explore the possibility of using vertical profiles of temperature and relative humidity retrieved by a surface-based multifrequency microwave profiling radiometer to determine the bases and tops of clouds as well as amounts (sky cover) of individual layers using a technique originally developed by Chernykh and Eskridge (CE) for radiosonde profiles. The CE method determines the presence of cloud layers by monitoring the second derivative of temperature T(z) and relative humidity R(z) vertical profiles. Necessary conditions for detection of cloud layers are: T"(z) >= 0 and R"(z) <= 0. We have improved this algorithm by using supplementary retrieved data not measured by radiosondes: the cloud base height estimated by an infrared thermometer and the liquid water path L. This extra set of conditions can be summarized as follows: a cloud contains liquid water whenever the retrieved L exceeds a certain threshold value and its cloud base height is lower than some maximum value. The cloud amount is estimated using the minimum dewpoint depression within the detected layer and the corresponding temperature. This dependence of cloud amount on dewpoint depression and temperature is presented in the form of a new Arabey diagram empirically derived for the liquid water and mixed-phase cloud climatology of the Ottawa area. Comparisons will be made between predictions based on this improved cloud detection algorithm for low and midlevel clouds as well as cases of clear skies and co-located "ground truth" surface observations using daytime data collected in Ottawa between the fall of 2004 and the summer of 2005. These sky observations are based on WMO Cloud Atlas descriptions supplemented by hourly surface observations of cloud cover from the Ottawa Airport. Our new algorithm has been shown to significantly reduce the number of false positives.