Abstract

Accurate a priori Zenith Hydrostatic Delays (ZHDs) are required during the processing of space geodetic observations. The data provided by the European Centre for Medium-Range Weather Forecasts (ECMWF) are routinely processed and translated in terms of ZHDs by the Vienna University of Technology. The usual way to compute gridded ZHDs at a particular location is to correct the four nearest nodes of the grid for their difference in height with respect to the site height and to then interpolate these reduced values. This paper compares and discusses the performance achieved by five methods of reduction that have been proposed in the literature. The interpolated ZHD values are compared with the site-specific ones for a global network of 363 sites over a three-year period. The methods that only use vertical profiles of atmospheric pressure lead to annual signals that are correlated with ground temperature, while the methods that take temperature into account do not contain such annual signals. The reduction methods that use a constant temperature lead to errors of ~2 mm in terms of equivalent height. We also find that the a posteriori errors of the reduced gridded ZHDs are strongly correlated with the pressure over temperature ratio. We recommend that gridded ZHDs be reduced with the combination of both pressure and temperature when processing space geodetic observations.