Recent advances of harmonic delay models for the neutral atmosphere (Invited)

The path delay caused by the neutral atmosphere is one of the main factors limiting the accuracy of positioning products from the analysis of space geodetic observations, like those from GNSS (Global Navigation Satellite Systems). For a precise account of this effect, the slant path delays are typically separated into a hydrostatic part and a wet part with each of them being the product of the zenith delay and a mapping function. If available, the analyst is advised to determine the zenith hydrostatic delay from pressure values recorded at the site, to derive the mapping functions from operational analysis fields of numerical weather models, and to estimate the zenith wet delay within the least-squares adjustment. If neither local pressure values nor operational mapping functions can be accessed, it is recommended to use so-called blind models which are based on harmonic functions and do not need any parameters other than approximate position and date. We outline the recent advances made in this field of empirical models for tropospheric slant delays by reference to GPT2, an improved version of the Global Pressure and Temperature (GPT) model and the Global Mapping Functions (GMF). Characterized by an increased horizontal resolution of 5 degrees, a temporal variability down to semi-annual periodicities, and an extended set of meteorological parameters, the model's excellent usability for GNSS applications is demonstrated by validation against in-situ pressure and temperature observations as well as ray-traced delays. In addition, real time positioning and navigation applications profit from another, currently developed augmentation of GPT2 in terms of accurate a priori estimates for zenith wet delays. We discuss the subtleties of this extension and assess its performance by comparing to zenith total delays from GNSS solutions in 2011-2012.