hide
Free keywords:
-
Abstract:
Radiosondes observe pressure, temperature, humidity and wind at different height levels from the earth’s surface to the upper air. These observations are very useful to understand the dynamic process of weather and climate changes. There are several radiosonde archives, e.g. ‘University of Wyoming Atmospheric Science Radiosonde Archive’ and ‘Integrated Global Radiosonde Archive’, where the users can get the historical radiosonde data publicly. During the use of these data, we found that the altitudes of the first level of radiosonde data were not accurate for several stations (errors up to >1000 m). The inaccuracy of first level altitudes will affect the height-dependent studies and applications, and thus it is very important to check and correct the first level altitudes of radiosonde profiles. We presented a method to detect and recover the first level altitudes. We used the pressure, temperature, and relative humidity of both the first and second level and the altitude of the second level to estimate the first level height. Experiments show that the accuracies of the estimated heights are 2-3 m, which meets the requirement of most conventional applications. For some stations with inaccurate altitudes of the first level, we recovered their first level heights and used the recovered data to generated water vapor weighted mean temperature models for GNSS meteorology. The models generated from the recovered data are more accurate than those from original data. Since the implementation of this method does not need the support of external data, it is easy to use in practice.
