Integrating GPS with GLONASS for high‐rate seismogeodesy

Abstract

High‐rate GPS is a precious seismogeodetic tool to capture coseismic displacements unambiguously and usually improved by sidereal filtering to mitigate multipath effects dominating the periods of tens of seconds to minutes. We further introduced GLONASS (Globalnaya navigatsionnaya sputnikovaya sistema) data into high‐rate GPS to deliver over 2000 24 h displacements at 99 stations in Europe. We find that the major displacement errors induced by orbits and atmosphere on the low‐frequency band that are not characterized by sidereal repeatabilities can be amplified markedly by up to 40% after GPS sidereal filtering. In contrast, integration with GLONASS can reduce the noise of high‐rate GPS significantly and near uniformly over the entire frequency band, especially for the north components by up to 40%, suggesting that this integration is able to mitigate more errors than only multipath within high‐rate GPS. Integrating GPS with GLONASS outperforms GPS sidereal filtering substantially in ameliorating displacement noise by up to 60% over a wide frequency band (e.g., 2 s–0.5 days) except a minor portion between 100 and 1000 s. High‐rate multi‐GNSS (Global Navigation Satellite System) can be enhanced further by sidereal filtering, which should however be carefully implemented to avoid adverse complications of the noise spectrum of displacements.

Noise Characteristics of High‐Rate Multi‐GNSS for Subdaily Crustal Deformation Monitoring

“…Multipath signatures within GPS displacements at a quasi‐static station repeat from day to day, which can be exploited to reduce high‐rate GPS noise using data from neighboring days that is sidereal filtering (Genrich & Bock, ). Over the periods of longer than tens of minutes, in contrast, Geng et al () pointed out that the spatially correlated errors (e.g., orbit errors, atmospheric refractions, and tides) contributed significantly to high‐rate GPS noise and even surpassed the amplitudes of multipath errors. It was also highlighted that integrating GPS with GLONASS (GLObalnaya NAvigatsionnaya Sputnikovaya Sistema) data led to a substantial reduction of the high‐rate displacement noise by up to 40% within European regions.…”

Comparison of Three Methods for Estimating GPS Multipath Repeat Time

“…Since the repeat time of other GNSS satellite may be no longer near a sidereal day, the methods should be implemented with caution when they are used for the non-GPS systems. For example, the GLONASS satellites run 17 revolutions in 8 days and hence the MRT of GLONASS satellite is near 8 days [23]. To eliminate the multipath of GLONASS satellites for the day of interest, we should use the residual time series from eight days ago to generate multipath correction model rather than from one day before.…”

Strong‐Motion Seismogeodesy by Deeply Coupling GNSS Receivers with Inertial Measurement Units

“…However, high-rate GNSS is tumbled by its several orders of magnitude higher noise than that of conventional inertial seismograph, especially in case of very strong dynamic stress of up to 2 g or even more (e.g., Geng et al, 2017Geng et al, , 2018. Based on a GPS signal simulator, Ebinuma and Kato (2012) found that the Trimble NetR8 receiver's recordings for a 5-Hz sinusoidal signal suffered harshly a 125% amplitude error and RESEARCH LETTER 10.1029/2020GL087161 Key Points: • A new GNSS receiver architecture is developed by embedding both accelerometer and gyroscope to capture fierce seismic displacements • GNSS displacement error and phase lag are both reduced by 70% and 85% to 2 mm and 8.0 ms, respectively, compared to conventional receivers • Six-degree-of-freedom seismogeodesy is achieved with the displacement error and phase lag reduced further to 0.9 mm and 3.5 ms, respectively 10.1029/2020GL087161 a 74 • (or 40 ms, milliseconds) phase lag in case of an acceleration as high as 2 g. In a shake table experiment instead, Wang et al (2012) also told that the Trimble NetRS receiver's displacement measurements were likely to overshoot in amplitude by about 2 cm or almost 100% in contrast to the benchmark displacements, whenever the acceleration was over 1 g. While such amplitude errors and phase lags were observed frequently with regard to Trimble equipments struck by great accelerations and jerks, Ebinuma and Kato (2012) and Berglund et al (2015) illustrated that many mainstream GNSS receivers (e.g., Septentrio, Javad and Topcon) could also experience similar problems under extreme dynamic stress.…”