Tropopause height is a key climate change indicator, with accurate long-term trends vital for climate research. Radiosonde data, while reliable, has limited coverage. ERA5 (European Centre for Medium–Range Weather Forecasts Reanalysis v5) is a reanalysis dataset that provides global data, enabling comparisons of tropopause height estimates and then analyzed for long-term trends. Results show a 32 m mean difference (radiosonde – ERA5) with trends of +9 m/year (radiosonde) and +7 m/year (ERA5), crucial for characterizing tropopause changes under climate change.

The most commonly used tropopause height detection algorithm is based on the World Meteorological Organization (WMO) definition from 1957. However, with the increasing vertical resolution of atmospheric data, this definition has been found to fail in high-resolution radiosonde data. Thus, we propose an improved method to address this issue. This method can effectively bypassing thin inversions while preserving the fine–scale structure of the tropopause.

Kelvin–Helmholtz instability (KHI) is indicated by the critical value of the Richardson (Ri) number, which is usually predicted to be 1/4. Compared to high-resolution radiosondes, the threshold value of Ri could be approximated as 1 rather than 1/4 when using ERA5-based Ri as a proxy for KHI. The occurrence frequency of subcritical Ri exhibits significant seasonal cycles over all climate zones and is closely associated with gravity waves and background flows.

The ozone layer protects the life on the Earth by absorbing the ultraviolet (UV) radiation. Beside the long-term trend, there are strong interannual fluctuations in stratospheric ozone. The quasi-biennial oscillation (QBO) is an important interannual mode in the stratosphere. We show some new zonally asymmetric features of its impacts on stratospheric ozone using satellite data, ERA5 reanalysis, and model simulations, which is helpful for predicting the regional UV radiation at the surface.

Skilful predictions of weather and climate on subseasonal to seasonal scales are valuable for decision makers. Here we show the global spatiotemporal variation of the temperature SAO in the UTLS with GNSS RO and reanalysis data. The formation of the SAO is explained by an energy budget analysis. The results show that the SAO in the UTLS is partly modified by the SSTs according to model simulations. The results may provide an important source for seasonal predictions of the surface weather.