Tropospheric delays from GNSS for application in coastal altimetry

Abstract

In the scope of the development of an improved methodology for the computation of the wet tropospheric correction for coastal altimetry, based on the use of tropospheric delays derived from GNSS (Global Navigation Satellite Systems), various studies have been conducted aiming to improve the estimation, at global scale, of GNSS-derived tropospheric delays.Amongst these studies, two are presented in this paper: (1) a global assessment of zenith total delays (ZTD) determined at international data centres such as EPN (EUREF Permanent Network) and IGS (International GNSS Service) by comparison with ZTD solutions computed at the University of Porto (U.Porto) using state-of-the-art methodologies and ZTD estimated from ERA Interim, the latest reanalysis dataset from ECMWF (European Centre for Medium-Range Weather Forecasts), (2) evaluation of the accuracy of the hydrostatic component of the tropospheric delay (zenith hydrostatic delay, ZHD) estimation from different sources of surface pressure.When compared with ERA Interim, both IGS and U.Porto ZTD are homogeneous with a mean standard deviation of the differences, for all analysed sites, of 12mm. The U.Porto and IGS ZTD agree within 4mm (1σ), while for EPN the same result is only valid for the period after November 2006. Before that date, the EPN solutions are slightly degraded and require an adequate correction.Aiming to evaluate the accuracy of ZHD determination from various sources of atmospheric pressure, a study is presented that compares ZHD values determined with in situ measurements of surface pressure at a global set of 63 coastal barometric sites (GNSS stations), the corresponding values obtained from ECMWF operational model, ERA Interim sea level pressure (SLP) and ZHD from the Vienna Mapping Functions 1 (VMF1).Results show that the global grids of sea level pressure provided by ECMWF operational model, either at 0.25° or 0.125° spacing, or the ERA Interim reanalysis product at 1.5°, allow the estimation of the hydrostatic component of the tropospheric delay with an accuracy of 1 to 3mm at global scale, provided an adequate model for the height dependence of atmospheric pressure is adopted. In comparison, for VMF1 grids provided at 2.5° spacing, although the overall accuracy of ZHD estimation is 2–4mm in most sites, in regions with high variability and strong seasonal signal in the surface pressure, VMF1 can reveal errors with a clear annual pattern and epochs for which the error exceeds the centimetre level. When used to estimate the wet component of the tropospheric delay (zenith wet delay, ZWD) for coastal altimetry, these errors can translate into errors of similar magnitude in sea level studies.