Estimating atmospheric pressure loading regression coefficients from GPS observations

The loading exerted by atmospheric pressure on the surface of the Earth causes deformations, mainly in vertical direction. Consequently, these deformations are also subject to pressure variations. At present this effect is only modeled by a few research groups in the post-processing of very long baseline interferometry (VLBI) and global positioning system (GPS) observations. As the displacements may clearly exceed the accuracy goals, we implement vertical pressure loading regression coefficients as a new estimable parameter type in the Bernese GPS software. This development is applied to a network of 60 European permanent GPS stations extending from 35 to 79° northern latitude. The analysis comprises 1,055 days of observations between January 2001 and February 2004. During that period pressure variations as large as 80 hPa occurred at high latitude sites. A least squares solution including all observations and all relevant parameters yields significant regression coefficients for all stations but reveals also some critical issues with regard to the capability of this geodetic approach to verify results based on the geophysical convolution method.An erratum to this article can be found at