A new method for vertical stratification of zenith tropospheric delay

Tropospheric delay is an important error source in space geodetic techniques. It varies significantly with height and the height dependent tropospheric error remains one of the major limitations in Global Navigation Satellite System (GNSS) techniques especially in Network Real-Time Kinematic (RTK) with significant topography variations. Based on the Gaussian function and the ten-year monthly mean Zenith Tropospheric Delay (ZTD) data from ERA-Interim of European Centre for Medium-Range Weather Forecasts (ECMWF) with a horizontal resolution of 5 degrees x 5 degrees, a new ZTD vertical stratification model with seasonal variations called seasonal Gaussian function model is developed in this paper. Compared with the troposphere products of 259 International GNSS Service (IGS) stations in 2016, the ERA-Interim-derived ZTDs stratified by using the seasonal Gaussian function model show that the annual mean bias and Root Mean Square error (RMS) are -0.2 cm and 1.9 cm respectively, which are better than the results (bias: -0.7 cm, RMS: 2.1 cm) by using the seasonal exponential function model. The validation with 245 radiosonde stations' data in 2016 shows that the bias and RMS of the seasonal Gaussian function model are reduced by 0.3 cm and 0.2 cm, compared with that of the seasonal exponential function model. Thus the seasonal Gaussian function model instead of the exponential function model is more suitable for accurate ZTD stratification in GNSS data analysis. (C) 2018 COSPAR. Published by Elsevier Ltd. All rights reserved.