Systematic errors of mapping functions which are based on the VMF1 concept

Abstract

Precise global navigation satellite system (GNSS) positioning requires an accurate mapping function (MF) to model the tropospheric delay. To date, the most accurate MF is the Vienna mapping function 1 (VMF1). It utilizes data from a numerical weather model and therefore captures the short-term variability of the atmosphere. Still, the VMF1, or any other MF that is based on the VMF1 concept, is a parameterized mapping approach, and this means that it is tuned for specific elevation angles, station heights, and orbital altitudes. In this study, we analyze the systematic errors caused by such tuning on a global scale. We find that, in particular, the parameterization of the station height dependency is a major concern regarding the application in complex terrain or airborne applications. At this time, we do not provide an improved parameterized mapping approach to mitigate the systematic errors but instead we propose a (ultra-) rapid direct mapping approach, the so-called Potsdam mapping factors (PMFs). Since for any station–satellite link the ratio of the tropospheric delay in the slant and zenith direction is computed directly, the PMFs effectively eliminate the systematic errors.