Abstract
With more continuous Global Navigation Satellite System (cGNSS) network stations becoming available around the world and with improved data processing techniques, it is possible to observe and model subtle motions in the Earth’s crust that were previously undetectable. Critical to studying these subtle motions is understanding the contributions of various signals mixed into the cGNSS time-series, for example non-tidal atmospheric and ocean loading (NTAOL) and hydrologic loading. We investigate the effect that atmospheric and surface mass loading has on the stochastic properties of GPS time series around the Great Lakes (GL) region of the U.S. and Canada. This region is ideal for studying these effects because it is covered by a dense network of GPS stations and it is known to be affected by significant hydrological loading due to water level changes in the GL. We use readily available NTAOL and hydrologic loading models to remove these signals from the cGNSS time-series and track the variance changes in the residual time-series in order to quantify the effect of each loading component. In order to assess whether the loading models fully capture the full magnitude of displacement we also perform common mode filtering in order to extract the remaining spatially correlated signal. We estimate the stochastic parameters (white noise amplitude, power law amplitude and spectral index) and compare between the raw, loading corrected, and filtered loading corrected time series in order to evaluate the contribution of the different loading signals to the time series noise properties. The outcomes of this study will help validate existing loading models and where improvement may be needed. Results will also support GNSS data providers in assessing the quality of available GNSS stations for use in scientific and surveying applications.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have