It has been recently proposed by Clayson and Kantha (2008) to evaluate the climatology of atmospheric turbulence in the free atmosphere by applying a Thorpe analysis on standard radiosoundings obtained with relatively low resolution (LR) in the vertical. Since then, several studies based on this idea have been published. However, the impact of instrumental noise on the detection of turbulent layers was completely ignored in these works. The present study aims to evaluate the feasibility of turbulence detection from radiosoundings. For this purpose, we analyzed data of two field campaigns during which high-resolution (HR) soundings (10–20 cm) were performed simultaneously with standard LR soundings. We here used the raw data of standard radiosondes, the vertical resolution ranging from 5 to 8 m. A Thorpe analysis was performed on both LR and HR potential temperature profiles. A denoising procedure was first applied in order to reduce the probability of occurrence of artificial inversions, i.e. inversions due to instrumental noise only. We then compare the empirical probability of the sizes of the selected overturns from LR and HR profiles. From HR profiles in the troposphere, the scales of the detected turbulent overturns range from 4 to ~1000 m. The shape of the distribution of the size of overturns is found to sharply decrease with increasing scales. From LR profiles, the smallest scale of detected overturns is ~32 m, a similar decrease in the shape of the size distribution being observed. These results suggest that turbulent events can indeed be detected from standard radiosondes measurements in the troposphere. However these events are rather rare as they belong to the tail of the size distribution of the turbulent overturns: they only represent the 7% largest events. Similar conclusions are obtained from radiosondes data collected in the lower stratosphere, but the fraction of the detectable events is even smaller than in the troposphere since they are the 4% largest events.