The Greenland GNSS Network (GNET): Long-Term Stability and Validation of Geodetic-Grade GNSS Measurements of Greenland’s 3D Bedrock Displacement from 1995–2025

The Greenland GNSS Network (GNET) consists of 71 geodetic-grade Global Navigation Satellite Systems
(GNSS) stations mounted directly in stable bedrock along the perimeter of the Greenland Ice Sheet
(GRLS). The first continuously running GNSS (cGNSS) station pre-GNET, was set up in 1995 and has
been operating continuously since then. During the fourth International Polar Year (IPY, 2007–2008),
GNET was established with the addition of 49 remote stations, with additional sites added during the
following years. Over time, the installations have undergone various updates, which have helped to
stabilize and improve observations from the network. Early GPS-only receivers have gradually been
replaced by multi-constellation systems, improving positioning precision. The continuous updates have
resulted in a yearly mean of received observations to be above 95% since around 2020 across the network.
Operating cGNSS stations in the remote high Arctic is challenging and can give rise to downtime for
stations in the network. In this project, we aim to publish the most comprehensive, fully processed
positional time-series from GNET up to date, provided as geodetic coordinates and in a local East,
North, Up (ENU) frame, together with metadata documenting station history and development. The
processing is performed using the Precise Point Positioning (PPP) methodology implemented in the
GipsyX 2.5 software [Bertiger et al., 2020].
To evaluate the quality and performance of the processed time series, two analyses are performed. First,
a long-term stability analysis is carried out by fitting and removing a seasonal trajectory model from each
ENU component individually. The residuals are then used to estimate power-law noise characteristics,
derived from Lomb–Scargle periodograms. The analysis shows that all stations in the network can be
expected to operate with a noise profile in the flicker-noise region, −1 < κ < 1. [Goudarzi et al., 2015]
Second, we compare our processed positional time series with two previously published products from
other processing centers: the Jet Propulsion Laboratory (JPL) [NASA Jet Propulsion Laboratory, 2018]
and the Nevada Geodetic Laboratory (NGL) [Blewitt et al., 2018]. The comparison, based on 39 of 71
stations, uses inter-metric correlation analysis of trajectory model parameters fitted to individual time
series. The results show good agreement among the three products in the vertical direction but poor
correlation in horizontal displacements. The JPL and NGL products exhibit a small, non-zero seasonal
signal in the horizontal components, which is not expected [White et al., 2022; Bian et al., 2023; Materna
et al., 2021]. The amplitudes of these signals, however, are very small, suggesting that these signals likely
originate from specific modeling and processing choices during the PPP processing. Consequently, while
the vertical seasonal signals can be interpreted confidently, horizontal seasonal amplitudes and phases
should be treated with caution when using time-series from NGL or JPL compared to the product we
publish.
Overall, the results highlight the importance of processing strategy, noise characterization, and validation
for high-precision GNSS time series in geoscience applications.