Global Navigation Satellite System Sites Research Articles

Numerical experiment toward simultaneous estimation of the spatiotemporal evolution of interseismic fault slips and block motions in southwest Japan

Interseismic strain around southwest Japan can be attributed to slip phenomena on the plate interface and processes inside the continental plate, such as block motions and inland fault slips. Although many crustal deformation analyses have been carried out, the simultaneous estimation of these phenomena and comprehensive discussion regarding the effect on the total strain budget remain topics to be investigated. In this study, we conducted numerical experiment to evaluate the possibility of simultaneously monitoring the spatiotemporal evolution of interseismic fault slips and block motions based on state-space modeling. We aimed to estimate fault slips and block motions using the time series of dense continuous Global Navigation Satellite System sites covering southwest Japan, encompassing 25 years from 1996 onward. We calculated synthetic block motions of the forearc region, monotonically increasing back slips on block boundary faults, and long-term slow slip events in the Tokai region, Kii Channel, and Bungo Channel. Subsequently, we generated the expected synthetic displacement time-series at GNSS stations. Applying Kalman filtering, we successfully estimated the spatiotemporal evolution of block motions, cumulative back slips, multiple slow slip events occurring in different regions. Although the contributions of fault slips and block motions showed slight trade-off, main characteristics of the slip distributions and the direction of block motions were well-recovered. We recovered slow slip events with slips of 5–10 cm or larger. We investigated the estimation uncertainty and separation precision of the unknown parameters using the covariance matrix estimated by Kalman filtering. Focusing on the structure of the non-diagonal component of the covariance matrix, we evaluated the complex effects of site and subfault locations on the estimated slip spatial distribution bias. For instance, the extent of the correlation between subfaults suggested that the three slow slip regions have different tendency of the uncertainties of slip areas, extending toward landside, seaward, or both of them. Our framework enables the comprehensive evaluation of the contributions and uncertainties of various deformation sources covering the entire subduction zone.Graphical abstract

Temporal and spatial variations of terrestrial water storage in the northeastern Tibetan Plateau retrieved by GNSS observations

The variation of solid Earth's hydrologic loading could cause the elastic vertical deformation of the crust, and the Global Navigation Satellite System (GNSS) could effectively monitor the vertical displacement of surface loads. However, the widely used Green's function method does not work well in areas where GNSS sites are sparse. Here, the vertical displacement time series of GNSS stations and the Slepian basis function method have been applied to convert displacement signals into spatial spectrum signals. The elastic mass load theory is used to study the changes in terrestrial water storage on the Northeastern Tibetan Plateau (NETP). The temporal and spatial characteristics of seasonal water changes are well-represented by the GNSS, the Gravity Recovery and Climate Experiment (GRACE), and the Global Land Data Assimilation System (GLDAS). Several data points suggest that the change in water storage shows a gradual increase from the northeast to the southwest. The greatest annual amplitude of water storage retrieved by GNSS is ~159 mm, which is greater than the ~47 mm and ~44 mm obtained by GRACE and GLDAS. These results demonstrate that GNSS is capable of capturing small-scale hydrological changes in this region, whereas GRACE and GLDAS data tend to underestimate seasonal variations in water storage. We also used GNSS to describe the hydrological drought conditions in NETP, showing that GNSS could be used as an independent method to characterize hydrological drought events. The findings suggest it could observe water storage with high spatial and temporal resolution and aid in comprehending regional hydrological trends with a sparse GNSS station network.

Selection of noise models for GNSS coordinate time series based on model averaging algorithm

In the field of global navigation satellite system (GNSS) time series noise analysis, appropriately modeling the noise components plays an important role in determining the velocity of GNSS sites and quantifying the uncertainty associated with the velocity estimation. Over the years, researchers have focused on only one optimal noise model, while other noise models that show similar performance to the optimal model have been ignored. We investigated whether these ignored noise models can be made use of to describe the noise in the GNSS time series after applying a model averaging algorithm. The experimental data were derived from 28 International GNSS Service (IGS) sites in the California region of the United States and 110 IGS sites worldwide. The results showed that for the GNSS time series of 28 IGS sites in the California, 79%, 68%, and 75% of the site components can be applied the model averaging algorithm in the east/north/up (E/N/U) directions, respectively. Based on it, the east direction showed the best performance, with 50% of the site components obtaining more conservative velocity uncertainty after applying the model averaging algorithm compared to the optimal noise model. For GNSS time series of 110 IGS stations worldwide, the model averaging algorithm demonstrates excellent performance in all the E/N/U directions. In the E/N/U directions, 86%, 94%, and 57% of the site components can apply the model averaging algorithm. Building upon this, 77%, 65%, and 62% of the site components achieve more conservative velocity uncertainty in the E/N/U directions compared to the optimal noise model. To fully validate the feasibility of the model averaging algorithm, we also tested GNSS time series of varying lengths and different thresholds of the model averaging algorithm. In summary, the model averaging algorithm performs exceptionally well in the noise analysis of GNSS time series. It helps prevent overly optimistic estimation results.

Dominance of transient rheology in far-field postseismic deformation following the 2012 Mw 7.8 Haida Gwaii and 2013 Mw 7.5 Craig earthquakes

The Mw 7.8 2012 Haida Gwaii earthquake and the Mw 7.5 2013 Craig earthquake occurred along the Queen Charlotte transpressive margin, two months and 340 km apart. On and near Vancouver Island, hundreds of kilometres southeast of the two events, many continuous Global Navigation Satellite Systems sites underwent no discernable coseismic motion but up to 1 cm southeast postseismic motion (away from epicenters) which lasted for only 2 – 3 years. To resolve the postseismic signal of the two earthquakes, we used a minimum of four years of measurements before the Haida Gwaii earthquake to remove the longer-term motion of these sites that reflect mainly the locking of the Cascadia megathrust. To study the geodynamic process that governed the observed far-field postseismic motion, in particular its short duration, we employ a model of an elastic lithosphere with thickness 20 km and rigidity (μE) 32 GPa overlying Burgers-viscoelastic mantle layers in a self-gravitating spherical Earth which captures the regionally integrated effect of Earth rheology despite the complex and controversial tectonics of the study area. We find that the short-lived far-field postseismic motion was controlled entirely by the transient rheology. The observations can be explained by a steady-state (Maxwell) viscosity ηM of the upper mantle no lower than 1×1019 Pa s with a transient (Kelvin) viscosity ηK of about 6×1017 Pa s (Kelvin relaxation time τK ∼1.2 years) extending to large depths such as > 400 km. The far-field postseismic deformation is the result of the elastic plate serving as a stress guide but resisted by the asthenospheric mantle undergoing transient creep. The lithosphere-asthenosphere coupling is quantified by a “plate Kelvin response time” τKE=ηK/μE. Rheological parameters can be scaled to other values, but as long as τK and τKE stay unchanged, the far-field deformation results remain the same. Our observations presented a unique opportunity to quantify the effect of the transient rheology separately from that of the steady-state rheology.

Hydrology Drives Crustal Deformation and Modulates Seismicity in the Matese Massif (Italy)

Abstract We analyze the interplay between hydrology, deformation, and seismicity in the Matese massif, located in the Italian Southern Apennines. We find that this area is characterized by the concurrent action of two hydrologically driven processes: the first is the deformation detected by the Global Navigation Satellite Systems (GNSS) data in the shallowest part (above the elevation of the major springs) of the Earth crust, in phase with the hydrological forcing; the second is the triggering of seismicity at depth with a delay suggesting a downward diffusive process. We study the first process by applying a principal component analysis to the GNSS displacements time series, aiming to identify a common signal describing the largest data variance. We find that the maximum horizontal displacements associated with the first principal component (PC1) are larger than 1 cm in two GNSS sites, and the PC1 temporal evolution is well correlated and in phase with the flow of the largest spring of the region, which we consider as proxy of the water content of the massif. This suggests that the main source of horizontal deformation is the water content fluctuations in the shallow portion of the Matese aquifer, in particular within fractures located in correspondence of the main mapped faults. The deformation rates caused by this process are one order of magnitude larger than the tectonic ones. Finally, we infer the second process by observing the correlation between the background seismicity and the spring discharge with a time lag of 121 days. In our interpretation, downward diffusive processes, driven by aquifer water content variations, propagate pore-pressure waves that affect the fault’s strength favoring the occurrence of microearthquakes. This is supported by the values of hydraulic diffusivity (1.5 m2/s) and rock permeability (3.2–3.8×10−13 m2), which are compatible with what is observed in karstified limestones.

Tracking tsunami propagation and Island’s collapse after the Hunga Tonga Hunga Ha’apai 2022 volcanic eruption from multi-space observations

The quantity and accuracy of satellite-geodetic measurements have increased over time, revolutionizing the monitoring of tectonic processes. Global Navigation Satellite System (GNSS) and satellite radar signals provide observations beyond ground deformation, including how earthquake and tsunami processes affect variations in the ionosphere. Here, we study the Hunga Tonga Hunga Ha’apai (HTHH) volcanic eruption 2022 and its associated tsunami propagation with the analysis GNSS derived Total Electron Content (TEC), Synthetic Aperture Radar (SAR) Sentinel-1 data, complemented with tide gauge observations. We utilize GNSS sites data within a ~ 5000 km radius from the volcanic eruption for estimating the ionospheric perturbation as Vertical TEC. We give evidence on the detection of acoustic gravity, internal gravity, and atmospheric Lamb waves signatures in the TEC perturbation. In particular, the internal gravity waves that concentrated in the southwest of Tonga, directly correlates with the observed tsunami propagation direction as accounted by the tide gauge measurements. However, the acoustic gravity wave signature in the TEC is dominant in the north direction suggesting a surface deformation, which could be verified using Sentinel-1A SAR amplitude data. The analysis presented herein shows that within 5 h of the volcanic eruption, the central part of the HTHH island landscape disappeared with the biggest explosion. The unprecedented detail resolved by integrating satellite data yields previously unknown details of the deformation of the 2022 HTHH volcano eruption.

Investigating the ERA5-Based PWV Products and Identifying the Monsoon Active and Break Spells with Dense GNSS Sites in Guangxi, China

Precipitable water vapor (PWV) with high precision and high temporal resolution estimated by Global Navigation Satellite System (GNSS) is widely used in atmospheric research and weather forecasting. However, most previous works are not consensual concerning the characteristics of the PWV at different time scales and the identification of active and break spells during summ er monsoon climate in Guangxi, China. Taking radiosonde (RS) observations as reference, a strong correlation (R > 0.97) exists between GNSS PWV and RS PWV with a mean root mean square error (RMSE) of 2.68 mm. The annual, seasonal, monthly, and diurnal PWV variations of three years (2017, 2018 and 2020) over Guangxi in were comprehensively investigated using 104 GNSS stations and the fifth-generation European Centre for Medium-Range Weather Forecasts (ECMWF) Atmospheric Reanalysis (ERA5). The mean annual bias and RMSE between GNSS PWV and ERA5 PWV are −1.04 mm and 2.63 mm, respectively. The monthly bias and RMSE range are −0.77 to 3.87 mm, 1.32 to 4.45 mm, and the daily range is −1.41 to 1.07 mm and 1.11 to 5.02 mm, respectively. Additionally, the adopted average standardized rainfall anomaly criteria also identified 7/7/3 active spells and 5/3/7 break spells during the summer monsoon (June–September) from 2017 to 2020, respectively. During the three-year period, the daily amplitude ranges for active spells varied from 1.41 to 2.49 mm, 0.69 to 5.4 mm, and 0.88 to 1.41 mm, while the ranges for break spells were 2.45 to 6.76 mm, 1.66 to 8.17 mm, and 1.48 to 2.99 mm, respectively. The results show a superior performance of GNSS PWV compared to ERA5 PWV in Guangxi, and the maximum, minimum and occurrence time of PWV anomaly vary slightly with the season and the topography of stations. Despite temperature primarily exhibiting a negative correlation with rainfall, acting as a dampener, a positive correlation remains evident between PWV and rainfall. Therefore, densely distributed GNSS stations exhibit excellent capabilities in quantifying atmospheric water vapor and facilitating real-time monitoring of small and medium-scale weather phenomena.

Periodic signal extraction of GNSS height time series based on adaptive singular spectrum analysis

Singular spectrum analysis is widely used in geodetic time series analysis. However, when extracting time-varying periodic signals from a large number of Global Navigation Satellite System (GNSS) time series, the selection of appropriate embedding window size and principal components makes this method cumbersome and inefficient. To improve the efficiency and accuracy of singular spectrum analysis, this paper proposes an adaptive singular spectrum analysis method by combining spectrum analysis with a new trace matrix. The running time and correlation analysis indicate that the proposed method can adaptively set the embedding window size to extract the time-varying periodic signals from GNSS time series, and the extraction efficiency of a single time series is six times that of singular spectrum analysis. The method is also accurate and more suitable for time-varying periodic signal analysis of global GNSS sites.

Elucidating the magma plumbing system of Ol Doinyo Lengai (Natron Rift, Tanzania) Using satellite geodesy and numerical modeling

Ol Doinyo Lengai, located in the southern Eastern Branch of the East African Rift had several eruptive episodes with ash falls and lava flows (VEI 3) that caused damage to the nearby communities between 2007 and 2010. The volcano is remote and access is difficult. Although this volcano has been studied for decades, its plumbing system is still poorly understood, in part, because of the lack of precise observations of surface deformation during periods of quiet and unrest. This study investigates the volcanic plumbing system of Ol Doinyo Lengai and its surroundings using data from the network of permanent Global Navigation Satellite System (GNSS) sites monitoring the volcano (the TZVOLCANO network) around the flanks of the volcano and Interferometric Synthetic Aperture Radar (InSAR) observations. We constrain surface motions using 6 GNSS sites distributed around Ol Doinyo Lengai, operating between 2016 and 2021, and InSAR data covering nearly the same time period. Because of the complex local tectonics, the interpretation of the deformation pattern is not straightforward. We first invert the GNSS deformation and InSAR observations independently to infer potential deformation sources. Then we perform a joint inversion of both GNSS and InSAR datasets to verify our findings. We compare the results from the joint inversion with the results from inverting each dataset independently. The GNSS, InSAR, and joint inversion results point to a deflating source, located east of Ol Doinyo Lengai and southwest of the dormant volcano Gelai at a depth of 3.49 ± 0.03 km (GNSS inversion), 5.2 ± 1.2 km (InSAR inversion) and 3.49 ± 0.06 km (joint inversion) relative to the summit (vent) and with a volume change ∆V of −0.04 ± 0.05 × 106 m3 (GNSS inversion), −0.39 ± 0.29 × 106 m3 (InSAR inversion), and − 0.04 ± 0.01 × 106 m3 (joint inversion). Although this is non-unique modeling of geodetic datasets with small signals, the inversion results suggest that Ol Doinyo Lengai could be fed by an offset multi-reservoir system that includes a shallow magma reservoir (<5 km) east of Ol Doinyo Lengai, possibly connected to a deeper magma reservoir.

Surface Subsidence Monitoring in Kunming City with Time-Series InSAR and GNSS

Kunming city is located in the middle of Yunnan Province. Due to large-scale groundwater exploitation and urban development in recent years, this area has been affected by surface subsidence. In this paper, Interferometric Synthetic Aperture Radar (InSAR) and Global Navigation Satellite System (GNSS) data are used to monitor the surface subsidence in Kunming city area for better analysis and understanding. The study used data of Sentinel-1A from 2018 to 2020 with atmospheric correction based on GACOS to calculate the average annual subsidence rate in Kunming city area, and the results show that the maximum subsidence rate is 48 mm/year. The subsidence obtained by InSAR is compared with the vertical deformation information obtained by eight GNSS stations in continuous operation in the study area. The subsidence rate trend show by the two methods is consistent, which further verifies the validity of InSAR data to reflect the local deformation. Experimental results shown that the eastern and northeastern Dianchi lake areas were affected by underground resources mining, and the induced surface subsidence characteristics were obvious, with the surface subsidence rate reachde 48 mm/year and 37 mm/year respectively. The Kunyang Phosphate Mine also had different degrees of mining subsidence disaster, with the maximum subsidence rate reached 36 mm/year. The subsidence rate of InSAR and GNSS has the same trend on the whole. However, GNSS sites are generally located in stable areas, the settlement amount obtained in the same time period is somewhat different from that of InSAR.

A point-surface fusion method for vegetation water content retrievalconsidering optimization of GNSS sites and modeling elements

The normalized microwave reflectance index (NMRI) based on global navigation satellite system (GNSS) interferometric reflectometry has been proven to reflect the changes in vegetation water content (VWC) effectively, but it is limited to point data. A spatially continuous NMRI product can be obtained by fusing NMRI data (point) and moderate-resolution imaging spectroradiometer (MODIS) data (surface). However, the time resolution of the existing results is limited to 16 d, and the research on the selection of modeling elements is not deep enough. In this paper, a point-surface fusion method for VWC retrieval considering the optimization of GNSS sites and elements is proposed. This method is aimed at using MODIS multi-band to synthesize vegetation indices with 8 d spatial-temporal resolution and establishing the initial element set by combining meteorological elements, followed by optimizing the GNSS sites and modeling elements for best modeling scheme, finally constructing the point-surface fusion method for VWC retrieval, and comprehensively evaluating the performance of the method. The results indicate that optimizing GNSS sites and modeling elements are particularly critical to improving modeling accuracy. Compared with other vegetation elements, normalized difference vegetation index (NDVI), gross primary productivity, and leaf area index are essential elements that affect the modeling effect. Among them, NDVI is the critical element. The NMRI products with 8 d/500 m resolution obtained by GA-BPNN can better reflect the change of VWC. Furthermore, the spatial performance of NMRI products is consistent with the fire forecast products and is suitable for drought and fire forecasts.

Using Seafloor Geodesy to Detect Vertical Deformation at the Hikurangi Subduction Zone: Insights From Self‐Calibrating Pressure Sensors and Ocean General Circulation Models

AbstractSeafloor pressure sensor data is emerging as a promising approach to resolve vertical displacement of the seafloor in the offshore reaches of subduction zones, particularly in response to slow slip events (SSEs), although such signals are challenging to resolve due to sensor drift and oceanographic signals. Constraining offshore SSE slip distribution is of key importance to understanding earthquake and tsunami hazards posed by subduction zones. We processed seafloor pressure data from January to October 2019 acquired at the Hikurangi subduction zone, offshore New Zealand, to estimate vertical displacement associated with a large SSE that occurred beneath the seafloor array. The experiment included three self‐calibrating sensors designed to remove sensor drift, which, together with ocean general circulation models, were essential to the identification and correction of long‐period ocean variability remaining in the data after applying traditional processing techniques. We estimate that long‐period oceanographic signals that were not synchronous between pressure sensors and reference sites influenced our inferred displacements by 0.3–2.6 cm, suggesting that regionally deployed reference sites alone may not provide sufficient ocean noise correction. After incorporating long‐period ocean variability corrections into the processing, we calculate 1.0–3.3 cm of uplift during the SSE offshore Gisborne at northern Hikurangi, and 1.1–2.7 cm of uplift offshore the Hawke's Bay area at central Hikurangi. Some Hawke Bay displacements detected by pressure sensors near the trench were delayed by 6 weeks compared to the timing of slip onset detected by onshore Global Navigation Satellite System sites, suggesting updip migration of the SSE.

An optimized hydrological drought index integrating GNSS displacement and satellite gravimetry data

Space geodetic techniques are rapidly emerging as powerful tools for drought monitoring, including the Global Navigation Satellite System (GNSS) positioning data, and the gravimetric data from Gravity Recovery And Climate Experiment (GRACE) and GRACE Follow-On (GRACE-FO) satellite missions. However, while GNSS-based 3D displacements have high temporal resolution, they are impacted by diverse local effects especially for tectonically active regions. Satellite gravimetry derived terrestrial water storage changes (ΔTWS) are continuous in temporal sampling, however, at much coarser spatial resolutions, and there are systematic noise and a large 11-month data gap between GRACE and GRACE-FO observations. Here for the first time, we developed an Optimized GNSS/GRACE/GRACE-FO Hydrological Drought Index (OGHDI), by optimally combining two drought indices derived from GNSS vertical displacements (VDs) and GRACE/GRACE-FO ΔTWS, to better characterize decadal evolution of droughts over Southwest China, 2011–2020. The optimal relative weights of the two drought indices are determined through maximizing the correlations with the Composite Index (CI) commonly used in the study region. Then OGHDI is compared to the self-calibrating Palmer Drought Severity Index (scPDSI). The results show that VDs agree well with ΔTWS at most of the 30 analyzed GNSS sites, with a mean correlation of −0.44. Compared to the drought index developed using only GNSS data, the optimized technique improved the OGHDI-scPDSI correlations at 90% (27 sites) of the sites, with the highest correlation reaches 0.73. Further, more significant improvement is found at the regional scale than at individual sites, and the regional mean OGHDI-CI and OGHDI-scPDSI correlations increase from 0.37 to 0.57 and from 0.42 to 0.68, respectively. All drought indices show that Southwest China experienced two sustained and excessive drought periods from 2011 to mid-2013, and after 2019. We conclude that the novel integration of two distinct contemporary and complemenatry geodetic datasets, GNSS and GRACE/GRACE-FO, improved the ability of drought monitoring in Southwest China.

Machine learning-based prediction of Alpine foehn events using GNSS troposphere products: first results for Altdorf, Switzerland

Abstract. Remote sensing of water vapour using the Global Navigation Satellite System (GNSS) is a well-established technique and reliable data source for numerical weather prediction (NWP). However, one of the phenomena rarely studied using GNSS are foehn winds. Since foehn winds are associated with significant humidity gradients between two sides of a mountain range, tropospheric estimates from GNSS are also affected by their occurrence. Time series reveal characteristic features like distinctive minima and maxima as well as a significant decrease in the correlation between the stations. However, detecting such signals becomes increasingly difficult for large datasets. Therefore, we suggest the application of machine learning algorithms for the detection and prediction of foehn events by means of GNSS troposphere products. This initial study develops a new, machine learning-based method for detection and prediction of foehn events at the Swiss station Altdorf by utilising long-term time series of high-quality GNSS troposphere products. Data from the Automated GNSS Network Switzerland (AGNES) and various GNSS sites from neighbouring countries as well as records of an operational foehn index are used to investigate the performance of several different classification algorithms based on appropriate statistical metrics. The two best-performing algorithms are fine tuned and tested in four dedicated experiments using different feature setups. The results are promising, especially when reprocessed GNSS products are utilised and the most dense station setup is used. Detection- and alarm-based measures reach levels between 66 %–80 % for both tested algorithms and thus are comparable to those from studies using data from meteorological stations and NWP. For operational prediction, limitations due to the availability and quality of GNSS products in near-real time (NRT) exist. However, they might be mitigated to a significant extent by provision of additional NRT products and improved data processing in the future. Results also outline benefits for the results when including geographically relevant stations (e.g. high-altitude stations) in the utilised datasets.

A new method for tropospheric tomography using GNSS and Fengyun-4A data

Global Navigation Satellite Systems (GNSS) tropospheric tomography can reflect a three-dimensional distribution of water vapour. However, the common ill-posed problem for GNSS tropospheric tomography needs to be addressed. The relatively newly launched Chinese Fengyun-4A (FY-4A) satellite water vapour data have a high spatiotemporal resolution and may be used to assist in resolving the ill-posed problem. In this study, a new tropospheric tomographic method using both GNSS and FY-4A data was proposed. Tomographic schemes for the new method with the coverage rate of GNSS sites (CRGS, which is defined as the ratio of the number of GNSS sites to the number of horizontal grids in the tropospheric tomography) of 0.2, 0.28, 0.36 and 0.44 were compared with traditional tomographic schemes using the same CRGS. Test results showed that adding FY-4A data, in addition to GNSS data, to the tomographic system could effectively increase the number of voxels crossed by signals, which benefits the ill-posed problem. When CRGS equated 0.28, the new method had the largest improvement compared with the traditional method, in which the mean of the root mean squared error (RMSE) of the water vapour density (WVD) was reduced from 2.44 and 1.38 g/m3 to 2.05 and 1.22 g/m3 using radiosonde and ERA5 data as the references, respectively. Statistical results showed that the new method that used both GNSS and FY-4A data outperformed the traditional method that only used GNSS data.

On the Improvement of Mass Load Inversion With GNSS Horizontal Deformation: A Synthetic Study in Central China

AbstractWe carry out synthetic experiments of mass load inversion using Global Navigation Satellite System (GNSS) vertical and horizontal displacements in Central China. Using two synthetic mass load models from a checkerboard mass distribution and a more realistic distribution derived from the Gravity Recovery and Climate Experiment (GRACE) observations, the inverted mass changes are determined based on the load theory and then compared with the known mass‐change inputs. We find that the combination of horizontal displacements with the commonly used vertical displacements significantly improves the inversion results when the number of sites is larger than one third of the number of total grids with relatively uniform distribution over the region. Synthetic tests demonstrate that data from the Crustal Movement Observation Network of China, including the total number of GNSS sites, spatial distribution, and precision, are sufficient to infer the annual amplitudes of mass changes in the region with comparable spatial resolution as that of GRACE observations. For the current precision level of the GNSS surface displacements (3.0 mm in the vertical and 1.0 mm in the horizontal components), including horizontal displacements leads to ∼10% improvement in mass inversion, compared to using vertical displacements only. With a higher precision level of 0.50 and 0.17 mm (for vertical and horizontal components, respectively), an improvement of ∼20% can be achieved.

Modelling and prediction of GNSS time series using GBDT, LSTM and SVM machine learning approaches

Global navigation satellite system (GNSS) site coordinate time series provides essential data for geodynamic and geophysical studies, realisation of a regional or global geodetic reference frames, and crustal deformation research. The coordinate time series has been conventionally modelled by least squares (LS) fitting with harmonic functions, alongside many other analysis methods. As a key limitation, the traditional modelling approaches simply use the functions of time variable, despite good knowledge of various underlying physical mechanisms responsible for the site displacements. This paper examines the use of machine learning (ML) models to reflect the effects or residential effects of physical variables related to Sun and the Moon ephemerides, polar motion, temperature, atmospheric pressure, and hydrology on the site displacements. To form the ML problem, these variables are constructed as the input vector of each ML training sample, while the vertical displacement of a GNSS site is regarded as the output value. In the evaluation experiments, three ML approaches, namely the gradient boosting decision tree (GBDT) approach, long short-term memory (LSTM) approach, and support vector machine (SVM) approach, are introduced and evaluated with the time series datasets collected from 9 GNSS sites over the period of 13 years. The results indicate that all three approaches achieve similar fitting precision in the range of 3–5 mm in the vertical displacement component, which is an improvement in over 30% with respect to the traditional LS fitting precision in the range of 4–7 mm. The prediction of the vertical time series with the three ML approaches shows the precision in the range of 4–7 mm over the future 24- month period. The results also indicate the relative importance of different physical features causing the displacements of each site. Overall, ML approaches demonstrate better performance and effectiveness in modelling and prediction of GNSS time series, thus impacting maintenance of geodetic reference frames, geodynamics, geophysics, and crustal deformation analysis.

Fusion of Spatially Heterogeneous GNSS and InSAR Deformation Data Using a Multiresolution Segmentation Algorithm and Its Application in the Inversion of Slip Distribution

The fusion of global navigation satellite system (GNSS) and interferometric synthetic aperture radar (InSAR) deformation data can leverage the advantages of GNSS high temporal resolution and InSAR high spatial resolution, and obtain more abundant deformation data for constraints on geophysical structural and mechanical parameters. Existing studies seldom consider the spatial heterogeneity of largescale deformation data, which easily leads to obvious spatial aggregation of errors in the results of fusion. Here, we propose a novel multiresolution segmentation fusion (MRSF) method that uses a multiresolution segmentation algorithm to automatically classify the spatial heterogeneity of InSAR deformation data with similar deformation characteristics. We applied the MRSF method to the fusion of GNSS and InSAR deformation data covering the central valley aquifer system (CVAS) in southern California to verify its precision and robustness. Results show that the MRSF method can accurately reflect spatiotemporal evolution characteristics of displacement data and reliably estimate deformation for the times and locations of missing data. We then tested this method for geophysical parameter estimation by constructing three different sets of data, including dense GNSS sites, sparse GNSS sites, and sparse GNSS sites fused with InSAR data using MRSF, to invert the slip distribution of the Cascadia subduction zone. Results show that the inverted slip of the fused InSAR and GNSS data is comparable to that of the dense GNSS sites. Therefore, the MRSF method can obtain deformation results with high precision and high spatiotemporal resolution and effectively compensate for the lack of data caused by sparse GNSS sites during the geophysical inversion process.

High-precision GNSS PWV retrieval using dense GNSS sites and in-situ meteorological observations for the evaluation of MERRA-2 and ERA5 reanalysis products over China

Precipitable water vapor (PWV) product with high accuracy and high spatiotemporal resolution is important for climate change research. Hourly PWV products with a high spatiotemporal resolution can be provided by global navigation satellite system (GNSS) and global reanalysis data. The National Aeronautics and Space Administration (NASA) and the European Center for Medium-Range Weather Forecasts (ECMWF) have recently released their global reanalysis of the Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2) and the fifth-generation ECMWF Reanalysis (ERA5), respectively. A comprehensive evaluation of these two PWV reanalysis products over China has yet to be fully investigated. In this study, the PWV products of MERRA-2 and ERA5 reanalysis over China have been systematically evaluated based on 339 GNSS sites and in-situ ground meteorological observations from year 2016 to 2018. To mitigate the uncertainties of this evaluation, three enhanced vertical correction models for temperature, pressure and PWV that we developed previously are adopted. Four weighted mean temperature (Tm) models are established for the four regions of China (North China, South China, Tibet Plateau, and Northwest China), and they exhibit better accuracy than the other classical Tm models in China. The evaluation results of the MERRA-2- and ERA5-derived PWV show that the correlation coefficient, biases, and root mean square error (RMSE) in China are 0.98/0.99, −0.51/0.38 mm, and 2.50/1.99 mm, respectively. They both exhibit relatively large difference when applied to South China. For the same grid spatial resolution, ERA5 PWV product exhibits similar performance with MERRA-2. The monthly-mean-reanalysis-derived PWV exhibits obvious seasonality and has the largest difference in summer. ERA5 PWV can provide more details of the spatial pattern compared to MERRA-2 PWV. The diurnal anomaly variation for ERA5 is more consistent with GNSS data, and some differences at certain moments during autumn and winter have been observed. These results show that ERA5 PWV and MERRA-2 PWV exhibits excellent applicability in China and can help us to efficiently use PWV products for climate research in China.

Geodetic Analysis inside the South Korean Peninsula and Impact of the 2011 Tohoku–Oki (TO) Earthquake

AbstractThe most powerful Tohoku–Oki (TO) earthquake that occurred in Japan on 11 March 2011 affected Japan as well as South Korea. In the current study, we investigated contemporary geodetic deformation inside South Korea before and after the TO earthquake using Global Navigation Satellite System (GNSS) measurements from 01 January 2008 to 31 December 2017. Measured velocities of GNSS sites are modeled by Auto‐regressive moving average (ARMA) method to analyze the long‐term GNSS time‐series variation and to investigate the secular tectonic crustal deformation. We found that the maximum co‐seismic displacements during the TO earthquake reached up to 36.82 ± 0.21 mm in the east and 5.90 ± 0.08 mm in the north directions. The geometric model of the co‐seismic thrust surface was characterized by a rectangular plane with a dip of 12.0° and strike 200°. The thrust is situated at 25 km hypocenter depth, with an area roughly ∼470 km long and ∼120 km wide. The seismicity pattern after the earthquake indicated that the compressional strain started to be replaced by the extensional strain during the post TO earthquake period from 2011 to 2014. Further, the strain became predominantly extensional during the period 2015 to 2017, revealing an effective rotational change that occurred inside the Korean Peninsula.