A geophysical interpretation of the secular displacement and gravity rates observed at Ny-Ålesund, Svalbard in the Arctic-effects of post-glacial rebound and present-day ice melting

We have analysed the Ny-Ålesund very long baseline interferometry (VLBI) data over the period 1994 August to 2004 May, and we obtain secular displacement rates relative to a NNR-NUVEL-1A reference frame of and for the north, east and vertical directions, respectively. The corresponding global positioning system (GPS) station displacement rates relative to the same reference frame for the north, east, and vertical directions are at NYA1 and at NALL, where these GPS rates were derived from the ITRF2000 velocity solution of Heflin. From the comparison at 25 globally distributed collocated sites, we found that the difference in uplift rate between VLBI and GPS at Ny-Ålesund is mainly due to a GPS reference frame scale rate error corresponding to 1.6 mm yr−1 in the GPS vertical rates. The uplift rate was estimated to be 5.2 ± 0.3 mm yr−1 from the analysis of the tide gauge data at Ny-Ålesund. Hence the uplift rates obtained from three different kinds of data are very consistent each other. The absolute gravity (AG) measurements at Ny-Ålesund, which were carried out four times (period: 1998–2002) by three different FG5 absolute gravimeters, lead to a decreasing secular rate of . In this analysis, the actual data obtained from a superconducting gravimeter at Ny-Ålesund were used in the corrections for the gravity tide (including the ocean tide effect) and for the air pressure effect. We have estimated three geophysical contributions to examine the observed rates: (1) the effect of the sea-level (SL) change on a timescale of a few decades, (2) the effect of the present-day ice melting (PDIM) in Svalbard and (3) the sensitivity of the computed post-glacial rebound (PGR) effects to different choices of the models of past ice history and Earth's viscosity parameters. Our analysis indicates that the effect of SL change can be neglected as the main source of the discrepancy. On the other hand, the effect of PDIM cannot be ignored in explaining the mutual relation between the observed horizontal and vertical rates and the predicted ones. A large melting rate of the order of −75 cm yr11 (i.e. roughly 1.6 times larger than the mean rate derived from glaciology over Svalbard) would explain the observed uplift but only half of the gravity changes. Our comparison results clearly point out the importance of both the estimation accuracy of the elastic deformations and better observation accuracy to constrain the size of PGR effects in the northwestern Svalbard more tightly.