Abstract. Kelvin Helmholtz instability (KHI) is most likely to be the primary source for clear-air turbulence that is of importance in pollution transfer and diffusion and aircraft safety. It is exemplarily indicated by the critical value of Richardson (Ri) number, which is typically taken as 1/4. However, Ri is fairly sensitive to the vertical resolution of the dataset: a higher resolution systematically leads to a finer structure. The study aims to evaluate the performance of ERA5 reanalysis (137 model levels) in determining KHI spatial-temporal variabilities, by comparing it against a near-global high-resolution (10-m) radiosonde dataset during years 2017 to 2022, and to further highlight the global climatology and dynamical environment of KHIs. Overall, the occurrence frequency of Ri < 1/4 in the free atmosphere is inevitably underestimated by the ERA5 reanalysis over all climate zones, compared to radiosonde, due largely to the severe underestimation in wind shears. Otherwise, the occurrence frequency of KHI indicated by Ri < 1 in ERA5 is climatologically consistent with that from radiosondes in the free troposphere, especially over the midlatitude and subtropics in the Northern/Southern Hemisphere. Therefore, we infer that the threshold value of Ri should be approximated as 1, rather than 1/4, when using ERA5 for the KHI estimation. KHI occurrence frequencies revealed by both datasets exhibit significant seasonal cycles over polar, midlatitude, and subtropics regions, and they are consistently strong at heights of 10–15 km in the tropic region. In addition, the frequency at low-levels is positively correlated with the standard derivation of orography, and it is exceptionally strong over the Niño 3 region at heights of 6–13 km. Furthermore, the dynamical environment of KHI favors strong wind shears probably induced by the mean flows and the propagation of orographic or non-orographic gravity waves.