ZWD2PW – A Global Model for the Conversion of Zenith Wet Delays to Precipitable Water Vapour

Abstract:

Water vapor plays an important role as a basic climate variable in the thermodynamics and dynamics of the storm systems at the atmosphere and in the hydrological cycle on the local, regional and global scales. Recently the precipitable water vapour content (PW) is estimated using the zenith wet delay (ZWD) derived from ground-based GNSS data.

This study introduces a new global model for the conversion of zenith wet delays (ZWD) obtained from GNSS observations to precipitable water vapour (PW). The model was developed using a monthly mean ERA-Interim global numerical weather model datasets of 14 years between 2001-2014. The 1°×1° global grids of 37 pressure levels of temperature, relative humidity and the geopotential were collected from the ECMWF and the ZWD and PW values as well as the mean temperature of the water vapour (Tm) were calculated for each gridpoint.

Afterwards a direct and an indirect method was used to derive the global grids of the parameters used for the computation of the conversion factor between ZWD and PW. In the indirect method the conversion factor is computed as a function of the mean temperature of water vapour, where Tm is estimated as an empirical function of the surface temperature. The direct method models the conversion factor as a polynomial function of the surface temperature. The global grids of the model parameters were derived for both of the approaches.

The results show that the global climate strongly affects the parameters of the conversion formulae. It is well known that the most widely used conversion formulae were derived from North American and European radiosonde observations only. Our results show that the relative differences in terms of the conversion factors reach the level of 10%, which can lead a similar relative error in PW estimation.

The ZWD2PW model is also validated by a global set of radiosonde observations and the results show that it can be efficiently used for the conversion of ZWD to PW globally, especially for places where the availability of radiosonde observations are limited, thus local conversion models can not be derived from such datasets.