Validity and behaviour of tropospheric gradients estimated by GPS in Corsica

Abstract

Estimation of tropospheric gradients in GNSS data processing is a well-known technique to improve positioning (e.g. Bar-Sever et al., 1998; Chen and Herring, 1997). More recently, several authors also focused on the estimation of such parameters for meteorological studies and demonstrated their potential benefits (e.g. Champollion et al., 2004). Today, they are routinely estimated by several global and regional GNSS analysis centres but they are still not yet used for operational meteorology.This paper discusses the physical meaning of tropospheric gradients estimated from GPS observations recorded in 2011 by 13 permanent stations located in Corsica Island (a French Island in the western part of Italy). Corsica Island is a particularly interesting location for such study as it presents a significant environmental contrast between the continent and the sea, as well as a steep topography.Therefore, we estimated Zenith Total Delay (ZTD) and tropospheric gradients using two software: GAMIT/GLOBK (GAMIT version 10.5) and GIPSY-OASIS II version 6.1. Our results are then compared to radiosonde observations and to the IGS final troposphere products. For all stations we found a good agreement between the ZWD estimated by the two software (the mean of the ZWD differences is 1 mm with a standard deviation of 6 mm) but the tropospheric gradients are in less good agreement (the mean of the gradient differences is 0.1 mm with a standard deviation of 0.7 mm), despite the differences in the processing strategy (double-differences for GAMIT/GLOBK versus zero-difference for GIPSY-OASIS).We also observe that gradient amplitudes are correlated with the seasonal behaviour of the humidity. Like ZWD estimates, they are larger in summer than in winter. Their directions are stable over the time but not correlated with the IWV anomaly observed by ERA-Interim. Tropospheric gradients observed at many sites always point to inland throughout the year. These preferred directions are almost opposite to the largest slope of the local topography as derived from the world Digital Elevation Model ASTER GDEM v2. These first results give a physical meaning to gradients but the origin of such directions need further investigations.