A novel real-time PPP-AR framework for continuous crustal deformation monitoring across Japan

A novel real-time PPP-AR framework for continuous crustal deformation monitoring across Japan | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article A novel real-time PPP-AR framework for continuous crustal deformation monitoring across Japan Naofumi Takamatsu, Yusaku Ohta, Shengping He, Andreas Brack This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9135824/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract Kinematic Precise Point Positioning (PPP) is a powerful technique for Global Navigation Satellite System (GNSS)-based real-time crustal deformation monitoring. However, PPP float solutions are often affected by sub-daily periodic fluctuations, caused by orbit errors and atmospheric delays, thereby motivating the adoption of PPP with ambiguity resolution (PPP-AR). To enhance PPP-AR performance in the sub-daily domain, we investigated the optimized network configuration for phase bias estimation with respect to network scale and receiver type consistency. Using GEONET observations in Japan together with GFZ real-time orbit and clock products, we sequentially estimated satellite phase biases from receiver-type-specific regional networks and, by applying these estimates, processed 30-s GEONET data from approximately 1,300 stations in a simulated real-time kinematic PPP-AR mode. Two weeks of data in 2025 were analyzed to assess noise characteristics, together with case studies of recent seismic and volcanic deformation events. The results demonstrate significant suppression of long-term (> 2–3 h) coordinate fluctuations associated with GPS satellite orbital resonance while retaining short-term (< 2–3 h) noise levels comparable to those of float solutions. With nearly complete ambiguity resolution, horizontal positioning precision reached the millimeter level (daily standard deviation), representing an improvement of up to 40% relative to float solutions. These improvements were achieved through enhanced phase bias estimates using regional networks and user-side PPP-AR that is consistent with the derived phase bias estimates. The enhanced performance enables the capture of not only co-seismic displacements but also deformations ranging from several centimeters to a decimeter that evolve over several hours to several days, including earthquake swarms, volcanic inflation, and very early post-seismic deformation. Based on these findings, we propose a novel PPP-AR framework in which phase biases are estimated from receiver-type-specific regional networks, followed by user-side PPP-AR employing the same receiver type as that used within the network. This framework extends the applicability of real-time PPP to the sub-daily domain and helps bridge the temporal gap in GNSS-based deformation monitoring. GNSS GEONET PPP (precise point positioning) PPP-AR (PPP with ambiguity resolution) Phase bias UPD (uncalibrated phase delay) Regional network Receiver type consistency Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 Figure 17 Figure 18 Figure 19 Figure 20 Figure 21 1 Introduction Static Global Navigation Satellite System (GNSS) positioning based on 24-hour observation data provides station coordinates with millimeter-level precision, making it well suited for monitoring crustal deformation over timescales ranging from several days to years. In contrast, kinematic positioning estimates coordinates epoch-by-epoch with centimeter-level precision, enabling the detection of shorter timescale phenomena, including co-seismic deformation (e.g., Ohta et al., 2006 ; Bilich et al., 2008 ; Miyazaki et al., 2004 ; Murray et al., 2019). However, sub-daily crustal deformation signals are often obscured by various error sources such as orbit inaccuracies, atmospheric delays, and multipath effects (King et al., 2008 ; Geng et al., 2017 ; Geng et al., 2018 ; Genrich and Bock, 1992 ; Larson et al., 2007 ), making the separation of true geophysical signals from estimation errors a persistent challenge. Although previous studies have demonstrated the effectiveness of filtering techniques—including sidereal filtering (e.g., Geng et al., 2021 ; Jiang et al., 2021 ; Malservisi et al., 2015 ; Twardzik et al., 2019 ), independent component analysis (Milliner et al., 2020 ), and empirical orthogonal function analysis (Munekane, 2012 ) —, their application has largely been limited to research environments rather than operational systems. Precise Point Positioning (PPP) (Zumberge et al., 1997 ; Kouba and Héroux, 2001 ) has been widely adopted for crustal deformation detection owing to its low computational demand and operational flexibility. PPP requires rigorous modeling of satellite- and receiver-dependent biases, which are inherently eliminated in relative positioning based on double-differenced observations. For a single-receiver user, the estimable ambiguity parameters cannot be independently separated from satellite phase biases. To recover the integer property of the ambiguities, externally estimated satellite phase biases must be introduced as correction products. Due to the presence of atmospheric delays and imperfect orbit, clock, and bias corrections, fast ambiguity resolution remains challenging, particularly for analyses based on short data spans, including kinematic and sub-daily static processing (Ge et al., 2008 ; Geng et al., 2009 ; Schwarz et al., 2009 ). Several studies have focused on separating phase biases from integer ambiguities (Collins et al., 2010 ; Ge et al., 2008 ; Geng et al., 2019 ; Laurichesse et al., 2009 ; Odijk et al., 2016; Schaer et al., 2021 ). The uncalibrated phase delay (UPD) model proposed by Ge et al. ( 2008 ) is compatible with International GNSS Service (IGS) clock products and has therefore been widely adopted in PPP-AR implementations. Numerous studies have demonstrated its effectiveness in improving PPP performance (e.g., Geng and Bock, 2016 ; Li et al., 2017 ; Li et al., 2018 ; Zhang et al., 2018). The UPD model extracts wide-lane (WL) and narrow-lane (NL) satellite phase biases as the fractional-cycle parts of the corresponding float ambiguities (Ge et al., 2008 ). In this model, receiver-dependent phase biases are first eliminated by forming between-satellite single differences within a reference network. Subsequently, satellite-dependent phase biases are estimated as the form of single-difference satellite UPDs (i.e., inter-satellite differences of UPDs). The resulting phase bias products are distributed to users, who apply them in PPP with ambiguity resolution (PPP-AR) to achieve high precision positioning (Wang et al., 2018 ). To fully unlock the capability of the UPD model, both the network scale and receiver type within the network play critical roles. Previous studies have shown that the quality of phase bias estimates improves as the spatial extent of the network decreases (e.g., Li et al., 2018 ; Wang et al., 2018 ; Zeng et al., 2023 ). This improvement is attributed to the spatial correlation of unmodeled orbit and tropospheric errors across stations (Zhang et al., 2017 ), whereas hardware delays remain independent of network extent. As the network extent decreases, error characteristics tend to be more spatially correlated across stations. Consequently, a regional network can more effectively absorb unmodeled errors than a global network (Li et al., 2018 ; Wang et al., 2018 ). As a result, regional networks achieve higher ambiguity fixing rates and shorter times to first fix (Wang et al., 2018 ; Zeng et al., 2023 ). Moreover, Cui et al. ( 2021 ) demonstrated that differences in receiver specifications (i.e. hardware and firmware) can introduce discrepancies of up to 0.3 cycles in WL phase biases, thereby degrading WL ambiguity fixing performance. They further showed that inconsistencies in receiver specifications between the network- and user-sides can lead to incorrect ambiguity fixing or ambiguity resolution failures. These effects arise from the receiver dependency of pseudorange observations (Hauschild and Montenbruck, 2016 ), which are not fully corrected by conventional differential code bias (DCB) corrections or by the recently proposed observable-specific bias (OSB) corrections (Geng et al., 2022 ). Although the influence of network configuration on UPD estimation and PPP-AR performance has been investigated, its implications for crustal deformation monitoring have not yet been systematically evaluated. Japan provides an ideal field for evaluating the influence of network configuration on deformation monitoring. Geospatial Information Authority of Japan (GSI) operates a nationwide real-time GNSS Continuously Operating Reference Stations (CORS) network known as GEONET (GNSS Earth Observation NETwork System; Tsuji and Hatanaka, 2018 ), with an average spacing of approximately 20 km. It has revealed a wide range of surface deformation such as secular plate motion, co-seismic and post-seismic deformation, slow slip events, and surface loading effects (e.g., Fukuda et al., 2018; Heki, 2001 ; Nishimura et al., 2013 ; Ozawa et al., 2011 ; Sagiya et al., 2000 ; Saito et al., 2025 ; Tang et al., 2025 ). Recent advancements have highlighted near-real-time monitoring of transient volcanic deformation (e.g. Ohno et al., 2024 ) and real-time detection of co-seismic displacements (e.g. Kawamoto et al., 2017 ; Ohta et al., 2012 ; Ohno et al., 2022 ). In addition, there is a growing demand for the rapid and accurate determination of station coordinates to support governmental assessments issued shortly after major earthquakes (e.g. HERP, 2024 ; JMA, 2026 ), since GNSS serve as essential constraints for evaluating the spatiotemporal evolution of interplate coupling and stress changes. This study makes the following contributions: (1) quantifying the performance of kinematic PPP-AR under different network scales for satellite phase bias estimation, with particular focus on the sub-daily domain; (2) evaluating the impact of receiver type consistency on phase bias estimates and ambiguity fixing, while identifying systematic effects arising from receiver type mismatches; and (3) proposing an optimized and operationally feasible network configuration for phase bias estimation and providing practical recommendations for sub-daily deformation monitoring. Section 2 describes the kinematic PPP analysis procedure and the evaluation methodology. Section 3 presents coordinate noise characteristics under different network scales and receiver types and presents case studies of selected deformation events, including co-seismic displacement, earthquake swarms, and volcanic inflation. Section 4 discusses the effects of network scale and receiver type on phase bias estimation and proposes an optimized network configuration together with operational recommendations. Section 5 concludes the study and outlines future perspectives. 2 Methodology 2.1 GNSS data processing The processing workflow is divided into two steps: (1) satellite phase bias estimation using a GNSS reference network and (2) kinematic PPP-AR processing using the derived phase bias estimates. In step 1, satellite phase biases were estimated under three distinct network configurations: (i) a global network comprising 80 stations equipped with mixed receiver types selected from the IGS Real-Time Service (IGS, 2026 ); (ii) a regional network consisting of 69 Trimble Alloy receivers; and (iii) a regional network consisting of 69 TOPCON NET-G5 receivers (Fig. 1). These configurations are hereafter referred to as “gUPD”, “rUPD”, and “rUPDwTPS” respectively. Stations forming the regional networks were selected from GEONET (Fig. 2). A static PPP processing strategy was employed to obtain stable float ambiguity estimates. WL satellite phase biases at each epoch were estimated using Melbourne–Wübbena linear combination averaged over continuous observation epochs. NL satellite phase biases were subsequently derived from the float ambiguities after applying the estimated WL phase bias corrections. In step 2, data from approximately 1,300 GEONET stations were processed by PPP-AR, with the phase biases deduced from each network configuration applied. The PPP float solution (hereafter referred to as “float”) was also obtained during the PPP-AR processing. In both steps, 30-s sampled GPS and Galileo observations were analyzed in post-processing mode using the GFZ in-house GNSS positioning software, RTPPP (Real-Time Precise Point Positioning; Jiang, 2025 ). To emulate real-time conditions, GFZ real-time orbit and satellite clock products were employed (Brack et al., 2025 ). The processing period spanned June 16–30, 2025, during which few antennas and receivers were replaced and GFZ real-time products maintained high availability. The data were processed in two 8-day segments (June 16–23 and June 23–30). The first day of each segment was treated as an initialization period to ensure sufficient convergence of the positioning solutions. Further processing was conducted for three crustal deformation events: the M7.1 earthquake at the Hyuganada Sea on August 8, 2024; the earthquake swarms at Tokara Islands from June 26 to July 3, 2025; and the volcanic inflation episode at Ioto Island from August 28 to September 3, 2025 (see Fig. 2 for locations). Detailed processing settings are summarized in Table 1 . Table 1 The PPP-AR processing strategy in step 2. Positioning float gUPD rUPD rUPDwTPS PPP PPP-AR with global network PPP-AR with Trimble Alloy regional network PPP-AR with TOPCON NET-G5 regional network Satellites GPS, Galileo Orbit and clock GFZ real-time product (Brack et al., 2025 ) whose time intervals are 5 minutes for orbit and 5 seconds for clock Obervation Ionosphere-linear combination Elevation mask 7 degree DCB CODE final bias products (Dach et al., 2025) Satellite antenna calibration igs20_2375.atx Receiver antenna calibration Field calibration considering different specifications in antennas, radomes, and monuments (Nakagawa et al., 2024) Solid tide correction IERS2010 Ocean tide correction NAO.99b(Matsumoto et al., 2000) Zenith tropospheric delay Dry : Saasamoinen (1973) with standard atmosphere Wet : Estimate with GMF (Böhm et al., 2006) Tropospheric gradient Estimate with MacMillan et al. (1997) 2.2 Evaluation For each processing segment, days 2–8 were designated as the evaluation period. During this period, the daily standard deviation (STD) of epoch-by-epoch station coordinates and the daily ambiguity fixing rate (FR) were computed. The FR is defined as the ratio of ambiguity-fixed epochs to the total number of epochs processed (Zhang et al., 2020 ). In addition, amplitude spectra of the coordinate time series were computed over the evaluation period. Stations used for phase bias estimation were excluded from the computation of these performance metrics to ensure fair evaluation. The performance during deformation events was evaluated with reference to 24-h static solutions (F5; Takamatsu et al., 2023 ) and 3-hourly static solutions derived from 6-h observations using the same processing strategy. The deformation field was quantitatively assessed using variance reduction (VR). VR is defined as the residual sum of squares of horizontal displacements normalized by their corresponding standard deviations. For comparison, kinematic relative positioning was performed using the GNSS processing software MALIB (JAXA, 2024 ). Similar to PPP processing, 30-s sampled data and GFZ real-time products were employed. 3 Results 3.1 Noise characteristics at different network scales Figure 3 presents the coordinate time series at station ISHI over one week beginning on June 17, 2025. The float solution exhibits pronounced long-term fluctuations, particularly in the east–west (EW) component. These fluctuations were substantially suppressed in the gUPD solution; however, short-term noise was noticeably amplified, resulting in STDs comparable to those of the float solution. In addition, ambiguity resolution failed for several percent of the processed epochs. In contrast, the rUPD solution effectively mitigated both long-term fluctuations and short-term noise while achieving near complete ambiguity fixing. Similar improvements were observed at stations AIRA, CCJ2, STK2, and even MCIL located outside the regional network (Figure S1 ; see Fig. 1 for station locations). Figure 4 presents the 14-day median of station-specific daily STDs together with regression lines. Compared with the float solution, the gUPD solution shows substantial improvement in the EW component, whereas only marginal improvements are observed in the north–south (NS) and up–down (UD) components. In contrast, the rUPD solution demonstrates further improvement in the EW component and noticeable improvements in the NS and UD components. The daily STDs averaged across all available stations are 0.83 cm, 0.91 cm, and 2.99 cm for the EW, NS, and UD components, respectively (as indicated along the y-axis of the lower panel in Fig. 4). Accordingly, rUPD achieved millimeter-level precision in the horizontal components with an improvement of up to 40% in the EW component relative to the float solution. The histogram of daily FR indicates that rUPD reliably fixes ambiguities whereas gUPD frequently fails to do so (Fig. 5). Noise characteristics in the period domain are illustrated using amplitude spectra (Fig. 6a). In the short-period domain ( 2–3 hours), noise behavior is dominated by GPS orbital resonances (Zajdel et al., 2021 ), manifested as spectral peaks at the GPS sidereal day and its harmonics (a rightmost panel of Fig. 6a). These peaks were substantially attenuated in the rUPD solution, particularly in the EW component. Amplitudes at representative periods further demonstrate that rUPD consistently outperforms the other strategies across the examined frequency bands (Fig. 6b). 3.2 Noise characteristics across different receiver network configurations Figure 7 presents the scatter plots of the 14-day median daily STD, labeled according to the receiver type on the user-side. Adopting consistent receiver types between network- and user-sides (i.e. ALLOY for rUPD and NET-G5 for rUPDwTPS) results in slight improvements in the UD component, while producing minimal changes in the EW and NS components. Daily FR values also clustered within a higher range when receiver types were consistent between the network- and user-sides (Fig. 8). 3.3 Detectability of geophysical signals across multiple timescales In this section, each kinematic analysis strategy is applied to representative crustal deformation events—including co-seismic deformation, earthquake swarms, and volcanic inflation episode—to evaluate the capability of regional PPP-AR in detecting geophysical signals across multiple timescales. 3.3.1 M7.1 earthquake at the Hyuganada Sea The Hyuganada Sea is a seismically active region as the Philippine Sea Plate subducts beneath the continental plate. M7–8 class earthquakes have repeatedly occurred in the plate boundary. On August 8, 2024, an M7.1 interplate earthquake occurred at a depth of 31 km (Fig. 9a). Significant crustal deformation was observed near the epicenter, with a maximum displacement of approximately 13 cm toward the east-southeast at station 1088 (Fig. 9b). Residual displacement fields referenced to the F5 solution indicate that gUPD is comparatively noisier than the other kinematic strategies, as reflected by its lower VR value (Fig. 9d). Receiver-type-specific VR values indicate that the deformation field becomes more consistent when receiver types match between the network- and user-sides than when mismatches occur (Fig. 9e, 9f). Similar to the receiver-type-consistent cases of rUPD and rUPDwTPS, displacement field derived from the float solutions remain highly consistent with the F5, regardless of user-side receiver type (Fig. 9c). At station 1086, the eastward displacement signal was well captured by the float and receiver-consistent PPP-AR (i.e. rUPD) solutions. In contrast, the gUPD solution exhibits coordinate jumps during the pre-seismic period, attributable to failures in ambiguity fixing (Fig. 10). Similar jumps were observed in the receiver-inconsistent PPP-AR solution (i.e. rUPDwTPS) during the post-seismic period. 3.3.2 Earthquake swarms at the Tokara Islands The Tokara Islands, situated along the Nansei Islands Trench, are prone to recurrent earthquake swarms (HERP, 2025). In the northwestern offshore region of Takarajima Island, one of the Tokara Islands, seismicity began to intensify around 05:00 UTC on July 2, 2025 (Fig. 11a). The largest event (M5.6) occurred at 06:26 UTC, after which the seismicity diminished (shaded period in Fig. 11c). Static solutions indicate a southward displacement of approximately 4 cm at station 1243 (Fig. 11b). Although the rUPD time series exhibits moderate noise, it clearly reveals a gradual southward displacement during the episode, while the EW component remains stable. Notably, the kinematic solution demonstrates that the deformation was not instantaneous but evolved progressively in association with the earthquake swarms. This is consistent with the findings of Okada et al. ( 2026 ). Unlike position-domain filtering approaches, PPP-AR mitigates hardware delays and orbit errors directly at the observation level. Consequently, long-term noise is reduced without distorting the underlying geophysical signal. In contrast, the float and gUPD solutions remain more affected by long-term noises, thereby reducing the detectability of the evolving deformation. 3.3.3 Volcanic inflation episode at Ioto Island Ioto Island is a tectonically active volcanic island located in the southern part of Japan. There have been observed not only the island-scale long-term uplift at several tens of centimeters per year, but also episodic deformation events associated with earthquakes and volcanic activity. Around August 30, 2025, volcanic earthquakes began to increase. Prior to August 29, daily event counts remained below 20–30 events per day; however, earthquakes increased sharply to 153 events on August 30, 685 on August 31, and 700 on September 1 (based on local time; preliminary counts from JMA, 2025 ). At 10:14 UTC on September 1, a volcanic eruption occurred at Chidorigahama, in the western part of the island (Fig. 12a), followed by continued eruption over several days (JMA, 2025 ). GNSS static analysis indicates that station 0604, located in the eastern part of the island, experienced gradual uplift and northward displacement synchronous with the increase in seismicity, followed by stagnation during the eruption period (Fig. 12b). The kinematic relative positioning solution using a tectonically stable reference station 0603 (baseline length of approximately 220 km) is significantly contaminated by short-term noise. For such a long baseline, troposphere and ionosphere errors are no longer effectively cancelled because of reduced spatial correlation. In contrast, the rUPD solution tracks the deformation signal with higher precision than the relative positioning solution. Even so, the noise level remained higher than that observed in the previous case studies because only GPS observations were available. Divergent behaviors are observed in both the relative positioning and rUPD solutions, likely attributable to orbit or satellite clock instabilities. Notably, PPP-AR enables nationwide unified deformation monitoring once a regional network is established, whereas relative positioning requires careful selection of a suitable reference station for each case. This advantage is particularly significant in geographically constrained areas such as remote islands, where the availability of suitable reference stations is inherently limited. 4 Discussion In the previous chapter, we demonstrated that employing a regional network, together with ensuring receiver-type consistency between the network- and user-sides, is essential for achieving reliable ambiguity resolution and, consequently, extending the applicability of kinematic positioning to sub-daily deformation monitoring. In this chapter, we discuss these improvements with a focus on phase bias estimates and propose an optimized network configuration. 4.1 Phase biases under different network configurations Figures 13 and 14 present the time series of phase bias estimates for GPS and Galileo satellites, respectively. For visual clarity, integer offsets were applied where necessary to maintain continuity in the time series. gUPD provides continuous phase bias estimates owing to its global tracking capability; however, its time series exhibits greater dispersion than that of rUPD, including several divergence episodes in the NL phase biases. In contrast, rUPD yields comparatively stable and smooth time series for both WL and NL phase biases. These differences are consistent with the phase bias time series reported by Wang et al. ( 2018 ), who observed similar contrasts between global and regional networks in the United States. Notably, rUPD exhibits larger fluctuations at the beginning of each arc, reflecting the convergence process from the moment satellites first become visible within the regional network until the phase bias estimates stabilize. These low-quality data are effectively filtered by user-side quality control procedures and therefore do not significantly degrade ambiguity resolution reliability. In addition, simultaneous discontinuities were observed across satellites in the rUPD solutions. These correspond to changes in the reference satellite or reinitialization of the ambiguity datum. As long as the same reference satellite and ambiguity datum are consistently adopted in user-side, reliable ambiguity resolution can still be achieved. Furthermore, no systematic differences were detected among GPS block types or between Galileo satellite generations (IOV versus FOC). Histograms of phase bias residuals were computed over the entire evaluation period, using all available stations and satellites (Fig. 15a, b). The phase bias residual is defined as the fractional part of the float ambiguity after applying the phase bias correction (Li et al., 2018 ). Compared with gUPD, rUPD exhibits sharper distributions centered near zero, indicating that the regional network configuration improves phase bias estimates. The underlying mechanism may be explained by the absorption of spatially correlated unmodeled errors (e.g., orbits and troposphere) into the phase biases (Wang et al., 2018 ; Zeng et al., 2023 ). Meanwhile, a number of observations were excluded from WL ambiguity resolution in rUPD due to Flag U (Fig. 15c). These exclusions correspond to unstable phase bias estimates at the beginning of each arc as mentioned in the previous section. In a regional network, satellites observed at low elevation angles are often tracked only from a single direction, resulting in fewer contributing stations and poor observation geometry. Minor rises are observed in the tails of the WL residual distribution for rUPD (Fig. 15a); these are flagged as Flag F and excluded from the WL ambiguity resolution procedure. These rises comes from larger residuals in GPS Block III observations, as described in the subsequent paragraph. Figure 16 presents the WL phase bias residuals for different satellite observations, color-coded according to user-side receiver type. When receiver types are inconsistent between the network- and user-side, residuals for GPS Block III exhibit broader distribution with half-cycle shift. For GPS Block II series and Galileo, no shift is observed; however, the distributions remain noticeably wider. One plausible explanation for this degraded performance is the receiver-dependency of pseudorange biases (e.g., Hauschild and Montenbruck, 2016 ), which cannot be fully corrected through the application of constant DCB products. These receiver-dependent biases propagate into the WL phase bias estimates through the Melbourne–Wübbena linear combination, thereby impairing the recovery of the integer property of WL ambiguities on the user-side. However, the reason why only Block III residuals exhibit a systematic shift remains unclear. Conversely, when receiver types are consistent, the residuals distributions for all satellite types—including Block III— are more sharply concentrated around zero. This suggests that receiver-dependent contamination in the phase bias estimates is effectively cancelled under a receiver-type-consistent configuration. 4.2 Optimized network configuration for deformation monitoring In this section, we propose a novel PPP-AR framework for monitoring regional deformation across multiple temporal scales. This framework consists of three principal components: station screening, phase bias estimation using receiver-type-specific regional networks, and user-side PPP-AR with the same receiver type employed in phase bias estimation (Fig. 17). Within this framework, spatially correlated unmodeled errors are effectively separated from ambiguity parameters, enabling nationwide detection of instantaneous displacements as well as sub-daily deformations ranging from several centimeters to a decimeter. An operational advantage of the proposed framework is that, once networks are configured, comprehensive network-wide deformation monitoring can be performed without case-specific reconfiguration. Moreover, unlike conventional relative positioning, the framework avoids baseline-dependent biases and enables precise positioning even in geographically constrained regions such as remote islands. Even when a dense network is initially deployed, real-time operation may necessitate phase bias estimation under a sparse network due to data gaps caused by hardware failures or communication disruptions. The tolerance of the framework to station outages within the network was therefore evaluated. Figures 18 and 19 present the STD and FR obtained when phase biases were estimated using progressively fewer stations. The STD show no significant degradation for either sparse network. However, when the network was reduced to only 10 stations, the FR decreased markedly. By contrast, no significant degradation was observed when 20 stations were retained. These results suggest that 20 stations represent a practical lower bound for stable phase bias estimation in Japan. This threshold provides a quantitative reference for developing operational policies related to station maintenance, redundancy planning, and network management. 4.3 Applicability to continuous deformation monitoring In light of the anticipated megathrust earthquake along the Nankai Trough, we investigated whether continuous deformation monitoring remains feasible after network stations experience significant co-seismic displacements. Using the 2011 off the Pacific coast of Tohoku Earthquake (Mw9.0; hereafter referred to as the Tohoku-Oki earthquake) as a case study, two regional networks were constructed: one comprising Trimble 5700 receivers and the other comprising TOPCON NET-G3 receivers (Fig. 20). Phase biases were estimated after excluding near-field stations identified by co-seismic displacements greater than 10 cm, followed by PPP-AR analysis with the corresponding receiver types. The 10 cm threshold is adopted based on past studies that co-seismic displacement of several centimeters can be reliably detected using kinematic positioning techniques (e.g. Kawamoto et al., 2016 ). Consequently, long-term fluctuations in the float solutions were substantially suppressed, enabling detection of very early post-seismic deformation signals, as evidenced by gradual eastward movements at stations 0171 and 0176 (Figure S4 ). A more spatially coherent post-seismic deformation field was obtained compared with that derived from the float solutions, which show systematic shifts at most stations and exhibit exceptional behavior at others, as evident from the residual plot (Fig. 21). Time series at stations sufficiently distant from the epicenter (e.g. 0087 and 0450 in Figure S4 ) showed stable behavior. However, when near-field stations were included in phase bias estimation, occasional coordinate jump and offsets were observed, accompanied by a decreased ambiguity fixing rate (e.g. 0176 and 0450 in Figure S4 ). Moreover, the spatial coherence of the post-seismic deformation field was slightly degraded, manifesting as residuals of 1–2 cm for the NET-G3 receivers (Fig. 21). These subtle degradations would be caused by large co-seismic displacements that compromise the static analysis used for phase bias estimation, leading to unstable phase bias estimates. To ensure high-quality phase bias estimates following a major earthquake, stations affected by significant co-seismic displacement must be promptly identified and screened. We propose a station screening strategy in which stations exhibiting large static offsets—derived from float solutions—are immediately excluded from phase bias estimation. When static offsets are computed as the coordinate difference between 1 minute before and 15 minutes after the earthquake, stations with offsets exceeding 10 cm correspond to the filled symbols shown in Fig. 20. As demonstrated in the Results section, the float solution is capable of detecting instantaneous displacements with accuracy comparable to that of the regional PPP-AR solution. An additional advantage of exploiting the float solution is that it is inherently generated within the PPP-AR processing workflow, allowing the screening strategy to be implemented without sacrificing operational efficiency. It should be noted that, to avoid misidentifying surface-wave-induced displacements as permanent offsets, the time window after the earthquake must be dynamically adjusted based on hypocenter and the spatial extent of network. However, station screening reduces the number of contributing stations, potentially degrading satellite tracking capability within the network. This reduction leads to shorter arcs in the phase bias estimates. For example, slight arc shortening was observed for G12 in 11:00 UTC, G22 in 17:00 UTC, and G32 in 22:00 UTC after the Tohoku-Oki earthquake (Figure S5 a). Consequently, the risk of incorrect ambiguity fixing or failure to achieve ambiguity fixing increases. To mitigate this limitation, further integration of multi-GNSS constellations and the use of temporally extrapolated phase bias estimates would be effective. Moreover, regardless of station screening, phase biases were unavailable for the minimum number of satellites to required for PPP-AR (i.e., five) during the several tens of minutes immediately following the mainshock (Figure S5 b). This limitation arises because traveling ionospheric disturbance excited by the mainshock induced cycle slips in the observation data, resulting in the exclusion of the affected observations. A geometry-free linear combination was employed as a cycle slip detector; however, it is sensitive to elevated ionospheric activity and may generate false detections under disturbed conditions. Multiple studies have proposed advanced cycle slip detection and correction methods under complex ionospheric conditions (Banville and Langley, 2012; Zhang et al., 2023 ; Zhang and Li, 2012 ). Incorporating these improvements would enable more robust and reliable monitoring of crustal deformation during gigantic earthquakes and their subsequent seismic events. 5 Conclusion and outlook We investigated how the network configuration—specifically network scale and the receiver types—in satellite phase bias estimation affects crustal deformation monitoring using kinematic PPP-AR. To evaluate its applicability to real-time operation, we processed roughly 1,300 GEONET stations across Japan with GFZ real-time orbit and clock products in a simulated real-time mode. While PPP float solutions are modulated by long-term noise associated with GPS orbital resonances and the PPP-AR implemented using a global network exhibits insufficient fixing rate (FR) and noisier coordinates time series, regional PPP-AR achieves nearly complete FR and the lowest noise levels across timescales ranging from several minutes to sub-daily periods. The daily STD reached the millimeter level in the horizontal components, representing an improvement of up to 40% in the EW component relative to the float solution. These improvements partly result from the receiver-type consistencies between phase bias estimation and user-side PPP-AR. In contrast, the use of inconsistent receiver types slightly degrades both STD and FR. Case studies of multiple deformation events demonstrate that receiver-type-consistent and regional PPP-AR enables monitoring of sub-daily deformation on the order of several centimeters to a decimeter as well as instantaneous displacements across Japan. These advances were achieved through improved phase bias estimates using the receiver-type-specific regional network, combined with user-side PPP-AR processing consistent with those estimates. Accordingly, we propose a framework in which phase biases are estimated from receiver-type-specific regional networks and user-side PPP-AR is performed using the same receiver type as that employed in the network. This framework enables continuous and comprehensive deformation monitoring within the network, including in geographically constrained regions. However, high quality of phase bias estimates needs to be continuously provided to ensure operational reliability. To achieve this, it is essential to maintain a sufficiently dense network configuration and implement an on-time station screening algorithm capable of addressing large displacement events. Sensitivity tests using sparse networks showed that 20 stations represent a practical lower bound for stable phase bias estimation in Japan. Further studies are required to resolve the systematic offset observed in GPS Block III phase bias residuals under receiver-type-inconsistent conditions. One potential approach is to calibrate this offset at the user-side prior to ambiguity resolution. Cui et al. ( 2021 ) proposed a method in which receivers are grouped using K-means clustering based on WL phase bias residuals, and the resulting inter-group differences are used to correct systematic offsets. Consequently, both the fixing rate and station coordinate precision were restored to levels comparable to those of receiver-type-consistent conditions. Implementing such calibration would further enhance the effective utilization of GPS Block III, whose constellation will expand in the coming years. Therefore, even when receiver-type selection is constrained, station coordinate precision can be improved. The extension of regional phase bias estimation to multi-GNSS constellations should also be considered in future studies. In recent years, several analysis centers of IGS Real-Time Committee have provided real-time multi-GNSS OSB products (IGS, 2026 ). By directly applying OSB corrections to carrier phase observations, users can perform PPP-AR without explicitly accounting for frequency and signal type differences. However, currently available OSB products are global in scope and are not optimized for regional applications. As demonstrated earlier, the application of regional phase bias products yields substantially higher positioning precision than global products. With the large number of existing satellites and the planned expansion of Quasi-Zenith Satellite System, Japan is expected to increasingly benefit from enhanced multi-GNSS capabilities. Leveraging this opportunity, the development of regional real-time multi-GNSS phase bias products will further enhance PPP capabilities and expand applications in socioeconomic activities and geophysical monitoring. Abbreviations DCB differential code bias FR fixing rate GEONET GNSS Earth Observation NETwork System GNSS Global Navigation Satellite System GSI Geospatial Information Authority of Japan HERP Headquarters of Earthquake Research Promotion IGS International GNSS Service JAXA Japan Aerospace Exploration Agency JMA Japan Meteorological Agency NL narrow-lane OSB observable-specific bias PPP precise point positioning PPP-AR precise point positioning with ambiguity resolution RTPPP Real-Time Precise Point Positioning STD standard deviation UPD uncalibrated phase delay VR variance reduction WL wide-lane Declarations Ethics approval and consent to participate Not applicable Consent for publication Not applicable Availability of data and materials The GNSS data of GEONET are available from https://terras.gsi.go.jp/index.php and https://www.gsi.go.jp/ENGLISH/geonet_english.html. The GNSS data of IGS are available from https://cddis.nasa.gov/archive/gnss/data/daily/. Competing interests The authors declare that they have no competing interests. Funding Part of this study was conducted during NT’s stay at GFZ Helmholtz Centre for Geosciences under the overseas research fellowship program for research related to space by the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan. This study was partly supported by the JST FOREST Program (Grant Number: JPMJFR202P, Japan). Funding was also provided by MEXT of Japan under the Third Earthquake and Volcano Hazards Observation and Research Program (Earthquake and Volcano Hazard Reduction Research). This work was supported by MEXT Coordination Funds, Japan Grant Number J013348. Authors' contributions NT designed the study, processed the data, and wrote the manuscript. YO, SH, and AB advised on the interpretation of results. All authors read and approved the final manuscript. Acknowledgements We are grateful to Dr. Hiroshi Munekane and Dr. Tomokazu Kobayashi for insightful discussions. The RTPPP and GFZ real-time orbit and satellite clock products were provided by GFZ Helmholtz Centre for Geosciences. The GNSS data in the global network was obtained from IGS web site. The plate models by Iwasaki et al. (2015) were constructed from topography and bathymetry data by Geospatial Information Authority of Japan (250-m digital map), Japan Oceanographic Data Center (500m mesh bathymetry data, J-EGG500, http://www.jodc.go.jp/jodcweb/JDOSS/infoJEGG_j.html) and Geographic Information Network of Alaska, University of Alaska (Lindquist et al., 2004). English proofreading was provided by Editage (https://www.editage.com/). Authors' information N.T. is a staff member at Geospatial Information Authority of Japan and a visiting scientist of the research group "Real-time GNSS" at GFZ Helmholtz Centre for Geosciences, Germany. Y.O. is a professor at the Graduate School of Science, Tohoku University, Japan. S.H. is a scientist in the research group "Real-time GNSS" at GFZ Helmholtz Centre for Geosciences, Germany. A.B. is the head of the research group "Real-time GNSS" at GFZ Helmholtz Centre for Geosciences, Germany. Endnotes Not applicable References Amante C, Eakins BW (2009) ETOPO1 arc-minute global relief model: procedures, data sources and analysis Banville S, Langley RB (2013) Mitigating the impact of ionospheric cycle slips in GNSS observations. J Geodesy 87(2):179–193. https://doi.org/10.1007/s00190-012-0604-1 Bilich A, Cassidy JF, Larson KM (2008) GPS seismology: Application to the 2002 M w 7.9 Denali fault earthquake. Bull Seismol Soc Am 98(2):593–606. https://doi.org/10.1785/0220190113 Brack A, He S, Du S, Wickert J (2025) Real-time GNSS for geosciences: Initial assessment of new data products from GFZ. EGU General Assembly 2025 Collins P, Bisnath S, Lahaye F, Héroux P (2010) Undifferenced GPS ambiguity resolution using the decoupled clock model and ambiguity datum fixing. Navigation 57(2):123–135. https://doi.org/10.1002/j.2161-4296.2010.tb01772.x Cui B, Li P, Wang J, Ge M, Schuh H (2021) Calibrating receiver-type-dependent wide-lane uncalibrated phase delay biases for PPP integer ambiguity resolution. J Geodesy 95(7):82. https://doi.org/10.1007/s00190-021-01524-6 Fukuda J (2018) Variability of the space-time evolution of slow slip events off the Boso Peninsula, central Japan, from 1996 to 2014. J Geophys Research: Solid Earth 123(1):732–760. https://doi.org/10.1002/2017JB014709 Ge M, Gendt G, Rothacher M al, Shi C, Liu J (2008) Resolution of GPS carrier-phase ambiguities in precise point positioning (PPP) with daily observations. J Geodesy 82(7):389–399. https://doi.org/10.1007/s00190-007-0187-4 Geng J, Teferle FN, Shi C, Meng X, Dodson A, Liu J (2009) Ambiguity resolution in precise point positioning with hourly data. GPS Solutions 13(4):263–270. https://doi.org/10.1007/s10291-009-0119-2 Geng J, Bock Y (2016) GLONASS fractional-cycle bias estimation across inhomogeneous receivers for PPP ambiguity resolution. J Geodesy 90(4):379–396. https://doi.org/10.1007/s00190-015-0879-0 Geng J, Jiang P, Liu J (2017) Integrating GPS with GLONASS for high-rate seismogeodesy. Geophys Res Lett 44(7):3139–3146. https://doi.org/10.1002/2017GL072808 Geng J, Pan Y, Li X, Guo J, Liu J, Chen X, Zhang Y (2018) Noise characteristics of high-rate multi-GNSS for subdaily crustal deformation monitoring. J Geophys Research: Solid Earth 123(2):1987–2002. https://doi.org/10.1002/2018JB015527 Geng J, Chen X, Pan Y, Zhao Q (2019) A modified phase clock/bias model to improve PPP ambiguity resolution at Wuhan University. J Geodesy 1–15. https://doi.org/10.1007/s00190-019-01301-6 Geng J, Xin S, Williams SD, Jiang W (2021) Comparing non-tidal ocean loading around the southern North Sea with subdaily GPS/GLONASS data. J Geophys Research: Solid Earth 126(3). https://doi.org/10.1029/2020JB020685 . e2020JB020685 Geng J, Wen Q, Zhang Q, Li G, Zhang K (2022) GNSS observable-specific phase biases for all-frequency PPP ambiguity resolution. J Geodesy 96(2):11. https://doi.org/10.1007/s00190-022-01602-3 Genrich JF, Bock Y (1992) Rapid resolution of crustal motion at short ranges with the Global Positioning System. J Geophys Research: Solid Earth 97(B3):3261–3269. https://doi.org/10.1029/91JB02997 Hauschild A, Montenbruck O (2016) A study on the dependency of GNSS pseudorange biases on correlator spacing. GPS Solutions 20(2):159–171. https://doi.org/10.1007/s10291-014-0426-0 Heki K (2001) Seasonal modulation of interseismic strain buildup in northeastern Japan driven by snow loads. Science 293(5527):89–92. https://doi.org/DOI:%252010.1126/science.1061056 HERP (2024) Evaluation of the 2024 Noto Peninsula Earthquakes. Earthquake Research Committee, The Headquarters of Earthquake Research Promotion. https://jishin.go.jp/main/index-e.html . Accessed 26 January 2026 HERP, The Headquarters of Earthquake Research Promotion (2025) Seismic Activity in the Ocean Near Tokara Islands (Near Kodakarajima Island). Earthquake Research Committee,. https://jishin.go.jp/main/chousa/25jul_tokara_2/p01-e.htm . Accessed 26 January 2026 IGS (2026) IGS RTS Products. International GNSS Service. https://igs.org/rts/products/ . Accessed 26 January 2026 Iwasaki T, Sato H, Shinohara M, Ishiyama T, Hashima A (2015) Fundamental structure model of island arcs and subducted plates in and around Japan. 2015 Fall Meeting JAXA (2024) MALIB: MADOCA-PPP Library. GitHub repository. https://github.com/JAXA-SNU/MALIB . Accessed 26 January 2026 Jiang J, Bock Y, Klein E (2021) Coevolving early afterslip and aftershock signatures of a San Andreas fault rupture. Sci Adv 7(15):eabc1606. https://doi.org/DOI:%252010.1126/sciadv.abc1606 Jiang X (2025) Multi-GNSS Real-time Precise Satellite Clock Estimation. Dissertation, Technische Universität Berlin JMA (2018) The Seismological Bulletin of Japan. Japan Meteorological Agency. https://www.data.jma.go.jp/eqev/data/bulletin/index_e.html . Accessed 26 January 2026 JMA (2025) Explanatory Report on Volcanic Activity (Ioto Island, Japan). Japan Meteorological Agency. https://www.data.jma.go.jp/vois/data/report/monthly_v-act_doc/monthly_vact_vol.php?id=329 . (in Japanese) Accessed 26 January 2026 JMA (2026) Nankai Trough Earthquake Extra Information. Japan Meteorological Agency. https://www.jma.go.jp/jma/en/Activities/earthquake.html . Accessed 26 January 2026 Kawamoto S, Hiyama Y, Ohta Y, Nishimura T (2016) First result from the GEONET real-time analysis system (REGARD): the case of the 2016 Kumamoto earthquakes. Earth Planets and Space 68:190. https://doi.org/10.1186/s40623-016-0564-4 Kawamoto S, Ohta Y, Hiyama Y, Todoriki M, Nishimura T, Furuya T, Sato Y, Yahagi T, Miyagawa K (2017) REGARD: A new GNSS-based real-time finite fault modeling system for GEONET. J Geophys Research: Solid Earth 122(2):1324–1349. https://doi.org/10.1002/2016JB013485 King MA, Watson CS, Penna NT, Clarke PJ (2008) Subdaily signals in GPS observations and their effect at semiannual and annual periods. Geophys Res Lett 35(3). https://doi.org/10.1029/2007GL032252 Kouba J, Héroux P (2001) Precise point positioning using IGS orbit and clock products. GPS Solutions 5(2):12–28. https://doi.org/10.1007/PL00012883 Larson KM, Bilich A, Axelrad P (2007) Improving the precision of high-rate GPS. J Geophys Research: Solid Earth 112. https://doi.org/10.1029/2006JB004367 Laurichesse D, Mercier F, Berthias J-P, Broca P, Cerri L (2009) Integer ambiguity resolution on undifferenced GPS phase measurements and its application to PPP and satellite precise orbit determination. Navigation 56(2):135–149. https://doi.org/10.1002/j.2161-4296.2009.tb01750.x Li P, Zhang X, Guo F (2017) Ambiguity resolved precise point positioning with GPS and BeiDou. J Geodesy 91(1):25–40. https://doi.org/10.1007/s00190-016-0935-4 Li X, Li X, Yuan Y, Zhang K, Zhang X, Wickert J (2018) Multi-GNSS phase delay estimation and PPP ambiguity resolution: GPS, BDS, GLONASS, Galileo. J Geodesy 92(6):579–608. https://doi.org/10.1007/s00190-017-1081-3 Lindquist KG, Engle K, Stahlke D, Price E (2004) Global topography and bathymetry grid improves research efforts. Eos Trans Am Geophys Union 85(19):186–186. https://doi.org/10.1029/2004EO190003 Malservisi R, Schwartz SY, Voss N, Protti M, Gonzalez V, Dixon TH, Jiang Y, Newman AV, Richardson J, Walter JI (2015) others Multiscale postseismic behavior on a megathrust: The 2012 Nicoya earthquake, Costa Rica. Geochemistry, Geophysics, Geosystems 16(6):1848–1864. https://doi.org/10.1002/2015GC005794 Milliner C, Bürgmann R, Inbal A, Wang T, Liang C (2020) Resolving the kinematics and moment release of early afterslip within the first hours following the 2016 Mw 7.1 Kumamoto earthquake: Implications for the shallow slip deficit and frictional behavior of aseismic creep. J Geophys Research: Solid Earth 125(9). https://doi.org/10.1029/2019JB018928 . e2019JB018928 Miyazaki S, Segall P, Fukuda J, Kato T (2004) Space time distribution of afterslip following the 2003 Tokachi-oki earthquake: Implications for variations in fault zone frictional properties. Geophys Res Lett 31(6). https://doi.org/10.1029/2003GL019410 Munekane H (2012) Coseismic and early postseismic slips associated with the 2011 off the Pacific coast of Tohoku Earthquake sequence: EOF analysis of GPS kinematic time series. Earth Planets and Space 64(12):1077–1091. https://doi.org/10.5047/eps.2012.07.009 Murray JR, Bartlow N, Bock Y, Brooks BA, Foster J, Freymueller J, Hammond WC, Hodgkinson K, Johanson I, López-Venegas A (2020) others Regional global navigation satellite system networks for crustal deformation monitoring. Seismological Research Letters 91(2A):552–572. https://doi.org/10.1785/0220190113 Nishimura T, Matsuzawa T, Obara K Detection of short-term slow slip events along the Nankai Trough, southwest Japan, using GNSS data. Journal of Geophysical Research: Solid Earth 118(6):3112–3125., Odijk D, Zhang B, Khodabandeh A, Odolinski R, Teunissen PJG (2013) (2016) On the estimability of parameters in undifferenced, uncombined GNSS network and PPP-RTK user models by means of S-system theory. Journal of Geodesy 90:15–44. https://doi.org/10.1007/s00190-015-0854-9 Ohno K, Ohta Y, Hino R, Koshimura S, Musa A, Abe T, Kobayashi H (2022) Rapid and quantitative uncertainty estimation of coseismic slip distribution for large interplate earthquakes using real-time GNSS data and its application to tsunami inundation prediction. Earth Planets and Space 74(24):1–18. https://doi.org/10.1186/s40623-022-01586-6 Ohno K, Ohta Y, Takamatsu N, Munekane H, Iguchi M (2024) Real-time modeling of transient crustal deformation through the quantification of uncertainty deduced from GNSS data. Earth Planets and Space 76(1):140. https://doi.org/10.1186/s40623-024-02068-7 Ohta Y, Meilano I, Sagiya T, Kimata T, Hirahara K (2006) Large surface wave of the 2004 Sumatra-Andaman earthquake captured by the very long baseline kinematic analysis of 1-Hz GPS data. Earth Planets and Space 58(2):153–157. https://doi.org/10.1186/BF03353372 Ohta Y, Kobayashi T, Tsushima H, Miura S, Hino R, Takasu T, Fujimoto H, Iinuma T, Tachibana K, Demachi T (2012) others Quasi real-time fault model estimation for near-field tsunami forecasting based on RTK-GPS analysis: Application to the 2011 Tohoku-Oki earthquake (Mw 9.0). Journal of Geophysical Research: Solid Earth 117(B2). https://doi.org/10.1029/2011JB008750 Okada Y, Ohta Y, Ohtate M, Ito Y, Ohzono M, Yakiwara H, Nakao S (2026) Recurrent earthquake swarms and transient crustal deformation off the Tokara Islands, southern Japan: Insights from the 2025 swarm sequence (Preprint). https://doi.org/10.21203/rs.3.rs-8399343/v1 . Research Square Ozawa S, Nishimura T, Suito H, Kobayashi T, Tobita M, Imakiire T (2011) Coseismic and postseismic slip of the 2011 magnitude-9 Tohoku-Oki earthquake. Nature 475(7356):373–376. https://doi.org/10.1038/nature10227 Sagiya T, Miyazaki S, Tada T (2000) Continuous GPS array and present-day crustal deformation of Japan. Pure appl Geophys 157(11):2303–2322. https://doi.org/10.1007/PL00022507 Saito K, Takahashi H, Ohzono M, Ohta Y (2025) Dynamic stress estimation during the M5.7 large remotely triggered earthquake due to the 2016 M7.0 Kumamoto earthquake, Kyushu, Japan, using dense strong motion and high-rate GNSS data. Seismol Res Lett 96(5):2848–2855. https://doi.org/10.1785/0220240311 Schaer S, Villiger A, Arnold D, Dach R, Prange L, Jäggi A (2021) The CODE ambiguity-fixed clock and phase bias analysis products: generation, properties, and performance. J Geodesy 95(7):81. https://doi.org/10.1007/s00190-021-01521-9 Schwarz CR, Snay RA, Soler T (2009) Accuracy assessment of the National Geodetic Survey’s OPUS-RS utility. GPS Solutions 13(2):119–132. https://doi.org/10.1007/s10291-008-0105-0 Takamatsu N, Muramatsu H, Abe S, Hatanaka Y, Furuya T, Kakiage Y, Ohashi K, Kato C, Ohno K, Kawamoto S (2023) New GEONET analysis strategy at GSI: daily coordinates of over 1300 GNSS CORS in Japan throughout the last quarter century. Earth Planets and Space 75(1):49. https://doi.org/10.1186/s40623-023-01787-7 Tang CH, Hsu YJ, Okada Y, Ohta Y, Sagiya T, Jian PR, Tamura Y, Jike T (2025) Geodetic evidence of shallow slow-slip phenomena beneath the southern Ryukyu forearc. Geophysical Research Letters 52, e2025GL114742. https://doi.org/10.1029/2025GL114742 Tsuji H, Hatanaka Y (2018) GEONET as infrastructure for disaster mitigation. J Disaster Res 13(3):424–432. https://doi.org/10.20965/jdr.2018.p0424 Twardzik C, Vergnolle M, Sladen A, Avallone A (2019) Unravelling the contribution of early postseismic deformation using sub-daily GNSS positioning. Sci Rep 9(1):1775. https://doi.org/10.1038/s41598-019-39038-z Wang S, Li B, Li X, Zang N (2018) Performance analysis of PPP ambiguity resolution with UPD products estimated from different scales of reference station networks. Adv Space Res 61(1):385–401. https://doi.org/10.1016/j.asr.2017.09.005 Zajdel R, Sośnica K, Bury G, Dach R, Prange L, Kazmierski K (2021) Sub-daily polar motion from GPS, GLONASS, and Galileo. J Geodesy 95(1):3. https://doi.org/10.1007/s00190-020-01453-w Zeng P, Zhang Z, Wen Y, He X, He L, Li M, Chen W (2023) Properties of multi-GNSS uncalibrated phase delays with considering satellite systems, receiver types, and network scales. Satell Navig 4(1):19. https://doi.org/10.1186/s43020-023-00110-9 Zhang X, Li X (2012) Instantaneous re-initialization in real-time kinematic PPP with cycle slip fixing. GPS Solutions 16(3):315–327. https://doi.org/10.1007/s10291-011-0233-9 Zhang X, Zhu F, Zhang Y, Mohamed F, Zhou W (2019) The improvement in integer ambiguity resolution with INS aiding for kinematic precise point positioning. J Geodesy 93(7). https://doi.org/10.1007/s00190-018-1222-3 Zhang X, Zhang Y, Zhu F (2020) A method of improving ambiguity fixing rate for post-processing kinematic GNSS data. Satell Navig 1(1):20. https://doi.org/10.1186/s43020-020-00022-y Zhang Z, Li B, Shen Y (2017) Comparison and analysis of unmodelled errors in GPS and BeiDou signals. Geodesy Geodyn 8(1):41–48. https://doi.org/10.1016/j.geog.2016.09.005 Zhang Z, Zeng J, Li B, He X (2023) Principles, methods and applications of cycle slip detection and repair under complex observation conditions. J Geodesy 97(5):50. https://doi.org/10.1007/s00190-023-01743-z Zumberge JF, Heflin MB, Jefferson DC, Watkins MM, Webb FH (1997) Precise point positioning for the efficient and robust analysis of GPS data from large networks. J Geophys Research: Solid Earth 102(B3):5005–5017. https://doi.org/10.1029/96JB03860 Supplementary Files FigS120250617CmprV3pppV3ppparV4InGfzrt2Seq.pdf FigS220250617CmprV4V4tpsInGfzrt2Seq.pdf FigS3Ioto.pdf FigS4TsTohoku.pdf FigS5WLUPD.pdf GraphicalAbstract.png Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 22 Apr, 2026 Reviewers invited by journal 30 Mar, 2026 Editor assigned by journal 17 Mar, 2026 First submitted to journal 16 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9135824","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":614400356,"identity":"627c32e0-a8d0-49c3-9c2f-5af31124b6f6","order_by":0,"name":"Naofumi Takamatsu","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0002-2647-8032","institution":"Geospatial Information Authority of Japan/GFZ Helmholtz Centre for Geosciences","correspondingAuthor":true,"prefix":"","firstName":"Naofumi","middleName":"","lastName":"Takamatsu","suffix":""},{"id":614400357,"identity":"0dd58654-1b71-458c-aed5-6c3cbeb378f7","order_by":1,"name":"Yusaku Ohta","email":"","orcid":"","institution":"Tohoku University: Tohoku Daigaku","correspondingAuthor":false,"prefix":"","firstName":"Yusaku","middleName":"","lastName":"Ohta","suffix":""},{"id":614400358,"identity":"67181e7c-dc0c-4737-b342-d762d8f7ec91","order_by":2,"name":"Shengping He","email":"","orcid":"","institution":"GFZ: Deutsches Geoforschungszentrum Potsdam","correspondingAuthor":false,"prefix":"","firstName":"Shengping","middleName":"","lastName":"He","suffix":""},{"id":614400359,"identity":"0e6f30dc-8ff9-44d1-a1d5-0cfc41450804","order_by":3,"name":"Andreas Brack","email":"","orcid":"","institution":"GFZ: Deutsches Geoforschungszentrum Potsdam","correspondingAuthor":false,"prefix":"","firstName":"Andreas","middleName":"","lastName":"Brack","suffix":""}],"badges":[],"createdAt":"2026-03-16 09:32:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9135824/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9135824/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105886439,"identity":"06d20d41-d84f-4220-845c-5946f5ab59c0","added_by":"auto","created_at":"2026-04-01 07:29:27","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":200516,"visible":true,"origin":"","legend":"\u003cp\u003eGNSS stations used for satellite phase bias estimation. Gray dots denote the complete set of IGS Real-Time Service (RTS) stations.\u003c/p\u003e","description":"","filename":"Fig1UpdSitesAvailable.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/ce850789c029f74a5ca7a0b3.png"},{"id":105886451,"identity":"5fd1e78a-7b7f-4b06-a6df-53db32bf7583","added_by":"auto","created_at":"2026-04-01 07:29:31","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1344900,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of GEONET stations as of June 17, 2025. The bathymetric background is derived from ETOPO1 (NOAA National Geophysical Data Center 2009; Amante and Eakins 2009). Black dashed lines indicate trench axes following Iwasaki et al. (2015) and Lindquist et al. (2004).\u003c/p\u003e","description":"","filename":"Fig2GEONET.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/14077a02de5eb482a3694538.png"},{"id":105886280,"identity":"80932671-f12b-4688-85b0-eb7f331b7166","added_by":"auto","created_at":"2026-04-01 07:28:55","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":530869,"visible":true,"origin":"","legend":"\u003cp\u003eCoordinate deviations from the mean at station ISHI for June 17–25, 2025. The STD and FR values for the plotting period are indicated using corresponding color coding. Gray solid segments denote epochs during which integer ambiguities were not resolved. Horizontal dotted lines represent the mean coordinate values for each processing strategy over the plotting period, with offsets applied for visual clarity.\u003c/p\u003e","description":"","filename":"Fig3200920250617CmprV3pppV3ppparV4InGfzrt2Seq.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/7e0b2e7c24e213f22f9cf3c7.png"},{"id":105886438,"identity":"ce13e3a5-dc87-4e40-8d9f-6bb0cd1859d3","added_by":"auto","created_at":"2026-04-01 07:29:27","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":953497,"visible":true,"origin":"","legend":"\u003cp\u003eScatter plots of daily standard deviation (STD) for all available stations. The plots represent 14-day medians computed for 17–30 June 2025. Each panel includes an inset summarizing the full plot. Thick black dashed lines indicate regression lines fitted using an M-estimator with a Cauchy loss function.\u003c/p\u003e","description":"","filename":"Fig4std20250617to20250630CmprV3ppparV4InGfzrt2WrtV3ppp.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/860958e377bec6add77d8c8d.png"},{"id":105886359,"identity":"d2dd293a-8d36-4739-9da0-89790e3539f9","added_by":"auto","created_at":"2026-04-01 07:29:20","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":139324,"visible":true,"origin":"","legend":"\u003cp\u003eFrequency distribution of daily ambiguity fixing rate (FR). The distribution is based on daily FR values from all available stations for June 17–30, 2025.\u003c/p\u003e","description":"","filename":"Fig5frBoxPlotCmprV3V4InGfzrt2.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/d3543082f2579b4db85cb14f.png"},{"id":105886452,"identity":"93c93c7f-2732-4b8f-96c2-78d70d0c0263","added_by":"auto","created_at":"2026-04-01 07:29:31","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":578905,"visible":true,"origin":"","legend":"\u003cp\u003eAmplitude spectra stacked across all available stations for June 17-30, 2025. (a) Amplitude spectra are shown relative to the float solution, which is presented as the reference in the rightmost panel. The magenta dashed lines indicate the five longest harmonics of the GPS sidereal day. (b) Amplitudes at representative periods (i.e. 5 minutes, 1 hour, and 12 hour) are shown. Amplitudes at 5 minutes and 1 hour are averaged over adjacent frequency bins, i.e. ±5 seconds for the 5-minute period and ±5 minutes for the 1-hour period.\u003c/p\u003e","description":"","filename":"Fig6amp20250617to20250624CmprV3ppparV4InGfzrt2WrtV3ppp.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/2476443afa66bcc88b189722.png"},{"id":105905460,"identity":"d2ec408b-e80c-4b2e-946c-f18c86ef965f","added_by":"auto","created_at":"2026-04-01 10:12:13","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":376955,"visible":true,"origin":"","legend":"\u003cp\u003eSame as Figure 4, but comparing rUPD (x-axis) and rUPDwTPS (y-axis). Markers are color-coded according to user-side receiver type.\u003c/p\u003e","description":"","filename":"Fig7std20250617to20250630V4tpsInGfzrt2WrtV4.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/76231d6fbff9b8fe91769d05.png"},{"id":105886281,"identity":"09cf81a2-cb9d-45a8-a179-5b4402d569bd","added_by":"auto","created_at":"2026-04-01 07:28:55","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":215832,"visible":true,"origin":"","legend":"\u003cp\u003eSame as Figure 5, but comparing rUPD and rUPDwTPS. Bars are color-coded according to user-side receiver type.\u003c/p\u003e","description":"","filename":"Fig8frBoxPlotCmprV4V4tpsInGfzrt2RcvDep.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/9533ce00ddaf55ef07a13685.png"},{"id":105886293,"identity":"06b84072-f600-4411-b409-cbf779bf63a0","added_by":"auto","created_at":"2026-04-01 07:29:02","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":976175,"visible":true,"origin":"","legend":"\u003cp\u003eCase study of the Hyuganada Sea earthquake (M7.1) on August 8, 2024. (a) Geographic overview of the Hyuganada Sea region. The study area is approximately indicated by an orange rectangle in Figure 2. (b) Co-seismic displacement derived from F5 coordinates on August 9 relative to the average coordinates over August 1–7. Panels (c)–(f) present kinematic displacement fields expressed as residuals relative to panel (b). The variance reduction (VR) within the area outlined by the black dotted rectangle in panel (b) is annotated in each panel. Receiver-type-specific VRs are also reported. Kinematic displacements are calculated as the differences between positions 2 minutes before and 5 minutes after the earthquake origin time. Horizontal and vertical residuals exceeding 2 cm and 1 cm, respectively, are omitted for visual clarity.\u003c/p\u003e","description":"","filename":"Fig9HyuganadaSea.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/02aaa9d6a9446f46c2a989e1.png"},{"id":105905485,"identity":"5b133205-22db-4a2c-adcd-ff6e44d4224a","added_by":"auto","created_at":"2026-04-01 10:12:18","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":151644,"visible":true,"origin":"","legend":"\u003cp\u003eEastward coordinate time series at station 1086 spanning the Hyuganada Sea earthquake. FR values during the plotting period are indicated using corresponding color coding. Gray dots denote epochs during which integer ambiguities were not resolved. The vertical dotted line marks the earthquake origin time. Horizontal dotted lines represent the mean coordinate for each processing strategy over the plotting period, with offsets applied for visual clarity.\u003c/p\u003e","description":"","filename":"Fig101086.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/2ae8f9db082fc89f941aaaf0.png"},{"id":105886361,"identity":"a718a01c-0aca-43c3-bde2-dc4c905a3765","added_by":"auto","created_at":"2026-04-01 07:29:21","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":1430479,"visible":true,"origin":"","legend":"\u003cp\u003eCase study of the earthquake swarms at the Tokara Islands. (a) Seismicity map from 09:00 UTC on July 1, 2025 to 09:00 UTC on July 3, 2025. The study area is approximately indicated by an orange rectangle in Figure 2. Seismicity data are derived from the Japan Meteorological Agency unified catalog (JMA, 2018). The yellow dot marks station 1243. (b) Coordinates deviation from the mean at station 1243. FR values during the plotting period are indicated using corresponding color coding, with offsets applied for visual clarity. Orange shaded frames denote periods during which seismic activity increased within region A in panel (a). (c) Temporal evolution of seismicity. Red circles with vertical lines indicate earthquakes that occurred within region A in panel (a).\u003c/p\u003e","description":"","filename":"Fig11Takarajima.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/3cf8963d24c1fbc2f66be902.png"},{"id":105886366,"identity":"6f6bfb6f-77d2-4a16-a2c4-e0e3a37699ae","added_by":"auto","created_at":"2026-04-01 07:29:22","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":1198195,"visible":true,"origin":"","legend":"\u003cp\u003eCase study of volcanic inflation at Ioto Island. (a) Geographic overview of Ioto Island. The study area is approximately indicated by an orange rectangle in Figure 2. Topographic contours are derived from DEM10B provided by the Geospatial Information Authority of Japan. The contour interval is 20 m. The yellow dots indicate GNSS stations. (b) Coordinates deviations from the mean at station 0604. FR values during the plotting period are indicated using corresponding color coding. Gray solid segments denote epochs during which integer ambiguities were not resolved. Magenta shaded frames denote the eruption period starting from 10:14 UTC September 1, 2025, marked by magenta vertical dotted lines. Horizontal dotted lines represent the mean coordinate for each processing strategy over the plotting period, with offsets applied for visual clarity.\u003c/p\u003e","description":"","filename":"Fig12Ioto.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/1839c1982d43b07f2568e23d.png"},{"id":105886316,"identity":"25ab9538-2c31-42e0-9c4b-45f7e30eb7e0","added_by":"auto","created_at":"2026-04-01 07:29:08","extension":"png","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":1077319,"visible":true,"origin":"","legend":"\u003cp\u003eGPS phase bias estimates on June 17, 2025. For visual clarity, markers are plotted at 3-hour intervals for gUPD and 1-hour interval for rUPD. The legend is organized according to satellite generation.\u003c/p\u003e","description":"","filename":"Fig13updgps.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/858bfeb18fc1100e21ad32d8.png"},{"id":105906041,"identity":"05c202a7-197d-4044-83a0-b6b4ef922b81","added_by":"auto","created_at":"2026-04-01 10:16:56","extension":"png","order_by":14,"title":"Figure 14","display":"","copyAsset":false,"role":"figure","size":987737,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 14. Same as Figure 13, but for Galileo.\u003c/p\u003e","description":"","filename":"Fig14updgalileo.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/54b001ad363fa10745c48a4a.png"},{"id":105886440,"identity":"d8515fad-30f4-45f9-8783-0a2268b5448f","added_by":"auto","created_at":"2026-04-01 07:29:27","extension":"png","order_by":15,"title":"Figure 15","display":"","copyAsset":false,"role":"figure","size":406856,"visible":true,"origin":"","legend":"\u003cp\u003eFrequency distribution of phase bias residuals. The distribution of (a) wide-lane and (b) narrow-lane phase bias residuals are shown. Light-colored regions represent residuals calculated from all available observations during June 17-30, 2025. Dark-colored hatched regions indicate residuals used for ambiguity resolution. Panel (c) presents the number of observations excluded from wide-lane ambiguity resolution. Labels indicate the reasons for exclusion: F, large residual (absolute value \u0026gt; 0.375); S, large sigma of the Melbourne–Wübbena linear combination; U, insufficient number of stations (\u0026lt;15); E, low elevation angle; L, loss of lock problem. Because the flags may overlap, the sum of the individual counts does not equal the total N value.\u003c/p\u003e","description":"","filename":"Fig15WlNlFloatAmbWUpdallCmprV3V4InGfzrt2.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/81d1afa9bdf54821534e08bc.png"},{"id":105886436,"identity":"f6e40617-1b45-4a98-be2c-ff0637b8239b","added_by":"auto","created_at":"2026-04-01 07:29:26","extension":"png","order_by":16,"title":"Figure 16","display":"","copyAsset":false,"role":"figure","size":517056,"visible":true,"origin":"","legend":"\u003cp\u003eFrequency distributions of wide-lane phase bias residuals for each user-side receiver type. Plots are color-coded according to user-side receiver type. Residuals are computed for (a) GPS Block III, (b) GPS Block II series, and (c) Galileo, and normalized within individual plot. Light-colored regions represent residuals calculated from available observations during June 17-30, 2025. Dark-colored hatched regions indicate residuals used for ambiguity resolution.\u003c/p\u003e","description":"","filename":"Fig16WlFloatAmbWUpdBlockIIIGECmprV4V4tpsInGfzrt2RcvDep.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/edbb38b39f73d57acd791be6.png"},{"id":105886443,"identity":"4411793c-7049-4351-abd3-04f18ef483e4","added_by":"auto","created_at":"2026-04-01 07:29:28","extension":"png","order_by":17,"title":"Figure 17","display":"","copyAsset":false,"role":"figure","size":132502,"visible":true,"origin":"","legend":"\u003cp\u003eKinematic PPP-AR framework for real-time continuous crustal deformation monitoring.\u003c/p\u003e","description":"","filename":"Fig17optimalconfiguration.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/9d2a78ca052bd4bd3346744a.png"},{"id":105886444,"identity":"0318300b-3760-4964-9663-c4e1582335c6","added_by":"auto","created_at":"2026-04-01 07:29:28","extension":"png","order_by":18,"title":"Figure 18","display":"","copyAsset":false,"role":"figure","size":621715,"visible":true,"origin":"","legend":"\u003cp\u003eSame as Figure 4, but comparing rUPD (x-axis) with solutions derived from fewer network stations (y-axis). The plots present 7-day medians for June 17–23, 2025. Only stations equipped with Trimble Alloy receivers are included.\u003c/p\u003e","description":"","filename":"Fig18std20250617to20250623CmprV40V41InGfzrt2WrtV4.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/615c29e49a94e2254de1c031.png"},{"id":105886445,"identity":"3367580f-3da5-4374-b34c-bafea5bec024","added_by":"auto","created_at":"2026-04-01 07:29:28","extension":"png","order_by":19,"title":"Figure 19","display":"","copyAsset":false,"role":"figure","size":158948,"visible":true,"origin":"","legend":"\u003cp\u003eSame as Figure 5, but comparing rUPD with solutions derived from fewer network stations. Data during June 17–23, 2025 are considered. Only stations equipped with Trimble Alloy receivers are considered when calculating the fixing rate.\u003c/p\u003e","description":"","filename":"Fig19frBoxPlotCmprV4V40V41InGfzrt2.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/ea36c147e44e5ce6fb51ed32.png"},{"id":105886318,"identity":"7305e696-0c2d-489d-9820-2f34770dc779","added_by":"auto","created_at":"2026-04-01 07:29:08","extension":"png","order_by":20,"title":"Figure 20","display":"","copyAsset":false,"role":"figure","size":367664,"visible":true,"origin":"","legend":"\u003cp\u003ePhase bias estimation network for the 2011 Tohoku-Oki earthquake. Filled colors denotes three-dimensional co-seismic displacements derived from the F5 solution. Small markers indicate user-side receivers and are color-coded consistently with the corresponding network-side receiver types.\u003c/p\u003e","description":"","filename":"Fig20Tohoku.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/fc0887fd0f599ff3f0aba468.png"},{"id":105886282,"identity":"2546feb3-7242-4366-82aa-84c3400828c1","added_by":"auto","created_at":"2026-04-01 07:28:55","extension":"png","order_by":21,"title":"Figure 21","display":"","copyAsset":false,"role":"figure","size":1162432,"visible":true,"origin":"","legend":"\u003cp\u003eVery early post-seismic deformation field of the 2011 Tohoku-Oki earthquake. Arrows are color-coded according to user-side receiver type: blue denotes Trimble 5700, and orange denotes TOPCON NET-G3. The lower panels display residuals relative to the “w/o near field” strategy.\u003c/p\u003e","description":"","filename":"Fig21TohokuPSD.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/aad0c79b74df53b9904437c8.png"},{"id":106093445,"identity":"f8e6ec4d-ddf9-4e12-8206-a46e5b1379f5","added_by":"auto","created_at":"2026-04-03 11:37:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":12604345,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/0671d1fd-ce2b-49fb-92a6-9771a3765430.pdf"},{"id":105886441,"identity":"f3eb9972-3d06-45d4-8705-5d888fde96e6","added_by":"auto","created_at":"2026-04-01 07:29:27","extension":"pdf","order_by":26,"title":"","display":"","copyAsset":false,"role":"supplement","size":4411685,"visible":true,"origin":"","legend":"","description":"","filename":"FigS120250617CmprV3pppV3ppparV4InGfzrt2Seq.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/1054d41d08b8e415bc270f5c.pdf"},{"id":105886365,"identity":"b14da78e-4351-4f1a-ac6f-fdf418db6e6d","added_by":"auto","created_at":"2026-04-01 07:29:22","extension":"pdf","order_by":27,"title":"","display":"","copyAsset":false,"role":"supplement","size":2533479,"visible":true,"origin":"","legend":"","description":"","filename":"FigS220250617CmprV4V4tpsInGfzrt2Seq.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/60ae946557a676be01268d63.pdf"},{"id":105886375,"identity":"8eb08897-84c9-4522-a4f1-0b5415afd56b","added_by":"auto","created_at":"2026-04-01 07:29:22","extension":"pdf","order_by":28,"title":"","display":"","copyAsset":false,"role":"supplement","size":1816873,"visible":true,"origin":"","legend":"","description":"","filename":"FigS3Ioto.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/6d499ec9425cde923aab59ed.pdf"},{"id":105886294,"identity":"60248ff5-bef6-403e-beed-a416ba3a3f41","added_by":"auto","created_at":"2026-04-01 07:29:02","extension":"pdf","order_by":29,"title":"","display":"","copyAsset":false,"role":"supplement","size":2541095,"visible":true,"origin":"","legend":"","description":"","filename":"FigS4TsTohoku.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/2708e0cfe050a58e37c5c204.pdf"},{"id":105886449,"identity":"29d78a4e-93ab-4e70-a665-95de6c650a0b","added_by":"auto","created_at":"2026-04-01 07:29:31","extension":"pdf","order_by":30,"title":"","display":"","copyAsset":false,"role":"supplement","size":2357653,"visible":true,"origin":"","legend":"","description":"","filename":"FigS5WLUPD.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/2949778a6b672335500f4ba7.pdf"},{"id":105905381,"identity":"fba7bba9-802d-4442-9cf6-b14715ce945d","added_by":"auto","created_at":"2026-04-01 10:11:58","extension":"png","order_by":31,"title":"","display":"","copyAsset":false,"role":"supplement","size":2042360,"visible":true,"origin":"","legend":"","description":"","filename":"GraphicalAbstract.png","url":"https://assets-eu.researchsquare.com/files/rs-9135824/v1/aa80ca9f129bd07ed38657d6.png"}],"financialInterests":"","formattedTitle":"A novel real-time PPP-AR framework for continuous crustal deformation monitoring across Japan","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eStatic Global Navigation Satellite System (GNSS) positioning based on 24-hour observation data provides station coordinates with millimeter-level precision, making it well suited for monitoring crustal deformation over timescales ranging from several days to years. In contrast, kinematic positioning estimates coordinates epoch-by-epoch with centimeter-level precision, enabling the detection of shorter timescale phenomena, including co-seismic deformation (e.g., Ohta et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Bilich et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Miyazaki et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Murray et al., 2019). However, sub-daily crustal deformation signals are often obscured by various error sources such as orbit inaccuracies, atmospheric delays, and multipath effects (King et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Geng et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Geng et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Genrich and Bock, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e1992\u003c/span\u003e; Larson et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2007\u003c/span\u003e), making the separation of true geophysical signals from estimation errors a persistent challenge. Although previous studies have demonstrated the effectiveness of filtering techniques\u0026mdash;including sidereal filtering (e.g., Geng et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Jiang et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Malservisi et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Twardzik et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), independent component analysis (Milliner et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), and empirical orthogonal function analysis (Munekane, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) \u0026mdash;, their application has largely been limited to research environments rather than operational systems.\u003c/p\u003e \u003cp\u003ePrecise Point Positioning (PPP) (Zumberge et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e1997\u003c/span\u003e; Kouba and H\u0026eacute;roux, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2001\u003c/span\u003e) has been widely adopted for crustal deformation detection owing to its low computational demand and operational flexibility. PPP requires rigorous modeling of satellite- and receiver-dependent biases, which are inherently eliminated in relative positioning based on double-differenced observations. For a single-receiver user, the estimable ambiguity parameters cannot be independently separated from satellite phase biases. To recover the integer property of the ambiguities, externally estimated satellite phase biases must be introduced as correction products. Due to the presence of atmospheric delays and imperfect orbit, clock, and bias corrections, fast ambiguity resolution remains challenging, particularly for analyses based on short data spans, including kinematic and sub-daily static processing (Ge et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Geng et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Schwarz et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2009\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSeveral studies have focused on separating phase biases from integer ambiguities (Collins et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Ge et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Geng et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Laurichesse et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Odijk et al., 2016; Schaer et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The uncalibrated phase delay (UPD) model proposed by Ge et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) is compatible with International GNSS Service (IGS) clock products and has therefore been widely adopted in PPP-AR implementations. Numerous studies have demonstrated its effectiveness in improving PPP performance (e.g., Geng and Bock, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Zhang et al., 2018). The UPD model extracts wide-lane (WL) and narrow-lane (NL) satellite phase biases as the fractional-cycle parts of the corresponding float ambiguities (Ge et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). In this model, receiver-dependent phase biases are first eliminated by forming between-satellite single differences within a reference network. Subsequently, satellite-dependent phase biases are estimated as the form of single-difference satellite UPDs (i.e., inter-satellite differences of UPDs). The resulting phase bias products are distributed to users, who apply them in PPP with ambiguity resolution (PPP-AR) to achieve high precision positioning (Wang et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTo fully unlock the capability of the UPD model, both the network scale and receiver type within the network play critical roles. Previous studies have shown that the quality of phase bias estimates improves as the spatial extent of the network decreases (e.g., Li et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Wang et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Zeng et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). This improvement is attributed to the spatial correlation of unmodeled orbit and tropospheric errors across stations (Zhang et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), whereas hardware delays remain independent of network extent. As the network extent decreases, error characteristics tend to be more spatially correlated across stations. Consequently, a regional network can more effectively absorb unmodeled errors than a global network (Li et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Wang et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). As a result, regional networks achieve higher ambiguity fixing rates and shorter times to first fix (Wang et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Zeng et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Moreover, Cui et al. (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) demonstrated that differences in receiver specifications (i.e. hardware and firmware) can introduce discrepancies of up to 0.3 cycles in WL phase biases, thereby degrading WL ambiguity fixing performance. They further showed that inconsistencies in receiver specifications between the network- and user-sides can lead to incorrect ambiguity fixing or ambiguity resolution failures. These effects arise from the receiver dependency of pseudorange observations (Hauschild and Montenbruck, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), which are not fully corrected by conventional differential code bias (DCB) corrections or by the recently proposed observable-specific bias (OSB) corrections (Geng et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Although the influence of network configuration on UPD estimation and PPP-AR performance has been investigated, its implications for crustal deformation monitoring have not yet been systematically evaluated.\u003c/p\u003e \u003cp\u003eJapan provides an ideal field for evaluating the influence of network configuration on deformation monitoring. Geospatial Information Authority of Japan (GSI) operates a nationwide real-time GNSS Continuously Operating Reference Stations (CORS) network known as GEONET (GNSS Earth Observation NETwork System; Tsuji and Hatanaka, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), with an average spacing of approximately 20 km. It has revealed a wide range of surface deformation such as secular plate motion, co-seismic and post-seismic deformation, slow slip events, and surface loading effects (e.g., Fukuda et al., 2018; Heki, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Nishimura et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Ozawa et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Sagiya et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Saito et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Tang et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Recent advancements have highlighted near-real-time monitoring of transient volcanic deformation (e.g. Ohno et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) and real-time detection of co-seismic displacements (e.g. Kawamoto et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Ohta et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Ohno et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In addition, there is a growing demand for the rapid and accurate determination of station coordinates to support governmental assessments issued shortly after major earthquakes (e.g. HERP, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; JMA, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2026\u003c/span\u003e), since GNSS serve as essential constraints for evaluating the spatiotemporal evolution of interplate coupling and stress changes.\u003c/p\u003e \u003cp\u003eThis study makes the following contributions: (1) quantifying the performance of kinematic PPP-AR under different network scales for satellite phase bias estimation, with particular focus on the sub-daily domain; (2) evaluating the impact of receiver type consistency on phase bias estimates and ambiguity fixing, while identifying systematic effects arising from receiver type mismatches; and (3) proposing an optimized and operationally feasible network configuration for phase bias estimation and providing practical recommendations for sub-daily deformation monitoring. Section \u003cspan refid=\"Sec2\" class=\"InternalRef\"\u003e2\u003c/span\u003e describes the kinematic PPP analysis procedure and the evaluation methodology. Section \u003cspan refid=\"Sec5\" class=\"InternalRef\"\u003e3\u003c/span\u003e presents coordinate noise characteristics under different network scales and receiver types and presents case studies of selected deformation events, including co-seismic displacement, earthquake swarms, and volcanic inflation. Section \u003cspan refid=\"Sec12\" class=\"InternalRef\"\u003e4\u003c/span\u003e discusses the effects of network scale and receiver type on phase bias estimation and proposes an optimized network configuration together with operational recommendations. Section \u003cspan refid=\"Sec16\" class=\"InternalRef\"\u003e5\u003c/span\u003e concludes the study and outlines future perspectives.\u003c/p\u003e"},{"header":"2 Methodology","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 GNSS data processing\u003c/h2\u003e \u003cp\u003eThe processing workflow is divided into two steps: (1) satellite phase bias estimation using a GNSS reference network and (2) kinematic PPP-AR processing using the derived phase bias estimates. In step 1, satellite phase biases were estimated under three distinct network configurations: (i) a global network comprising 80 stations equipped with mixed receiver types selected from the IGS Real-Time Service (IGS, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2026\u003c/span\u003e); (ii) a regional network consisting of 69 Trimble Alloy receivers; and (iii) a regional network consisting of 69 TOPCON NET-G5 receivers (Fig.\u0026nbsp;1). These configurations are hereafter referred to as \u0026ldquo;gUPD\u0026rdquo;, \u0026ldquo;rUPD\u0026rdquo;, and \u0026ldquo;rUPDwTPS\u0026rdquo; respectively. Stations forming the regional networks were selected from GEONET (Fig.\u0026nbsp;2). A static PPP processing strategy was employed to obtain stable float ambiguity estimates. WL satellite phase biases at each epoch were estimated using Melbourne\u0026ndash;W\u0026uuml;bbena linear combination averaged over continuous observation epochs. NL satellite phase biases were subsequently derived from the float ambiguities after applying the estimated WL phase bias corrections. In step 2, data from approximately 1,300 GEONET stations were processed by PPP-AR, with the phase biases deduced from each network configuration applied. The PPP float solution (hereafter referred to as \u0026ldquo;float\u0026rdquo;) was also obtained during the PPP-AR processing. In both steps, 30-s sampled GPS and Galileo observations were analyzed in post-processing mode using the GFZ in-house GNSS positioning software, RTPPP (Real-Time Precise Point Positioning; Jiang, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). To emulate real-time conditions, GFZ real-time orbit and satellite clock products were employed (Brack et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The processing period spanned June 16\u0026ndash;30, 2025, during which few antennas and receivers were replaced and GFZ real-time products maintained high availability. The data were processed in two 8-day segments (June 16\u0026ndash;23 and June 23\u0026ndash;30). The first day of each segment was treated as an initialization period to ensure sufficient convergence of the positioning solutions. Further processing was conducted for three crustal deformation events: the M7.1 earthquake at the Hyuganada Sea on August 8, 2024; the earthquake swarms at Tokara Islands from June 26 to July 3, 2025; and the volcanic inflation episode at Ioto Island from August 28 to September 3, 2025 (see Fig.\u0026nbsp;2 for locations). Detailed processing settings are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe PPP-AR processing strategy in step 2.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003ePositioning\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003efloat\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003egUPD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003erUPD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003erUPDwTPS\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePPP\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePPP-AR with global network\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePPP-AR with Trimble Alloy regional network\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePPP-AR with TOPCON NET-G5 regional network\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSatellites\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eGPS, Galileo\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOrbit and clock\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eGFZ real-time product (Brack et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) whose time intervals are 5 minutes for orbit and 5 seconds for clock\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObervation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eIonosphere-linear combination\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eElevation mask\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003e7 degree\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDCB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eCODE final bias products (Dach et al., 2025)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSatellite antenna calibration\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eigs20_2375.atx\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReceiver antenna calibration\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eField calibration considering different specifications in antennas, radomes, and monuments (Nakagawa et al., 2024)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSolid tide correction\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eIERS2010\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOcean tide correction\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eNAO.99b(Matsumoto et al., 2000)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZenith tropospheric delay\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eDry : Saasamoinen (1973) with standard atmosphere\u003c/p\u003e \u003cp\u003eWet : Estimate with GMF (B\u0026ouml;hm et al., 2006)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTropospheric gradient\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eEstimate with MacMillan et al. (1997)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Evaluation\u003c/h2\u003e \u003cp\u003eFor each processing segment, days 2\u0026ndash;8 were designated as the evaluation period. During this period, the daily standard deviation (STD) of epoch-by-epoch station coordinates and the daily ambiguity fixing rate (FR) were computed. The FR is defined as the ratio of ambiguity-fixed epochs to the total number of epochs processed (Zhang et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In addition, amplitude spectra of the coordinate time series were computed over the evaluation period. Stations used for phase bias estimation were excluded from the computation of these performance metrics to ensure fair evaluation.\u003c/p\u003e \u003cp\u003eThe performance during deformation events was evaluated with reference to 24-h static solutions (F5; Takamatsu et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) and 3-hourly static solutions derived from 6-h observations using the same processing strategy. The deformation field was quantitatively assessed using variance reduction (VR). VR is defined as the residual sum of squares of horizontal displacements normalized by their corresponding standard deviations. For comparison, kinematic relative positioning was performed using the GNSS processing software MALIB (JAXA, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Similar to PPP processing, 30-s sampled data and GFZ real-time products were employed.\u003c/p\u003e \u003c/div\u003e"},{"header":"3 Results","content":"\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Noise characteristics at different network scales\u003c/h2\u003e \u003cp\u003eFigure 3 presents the coordinate time series at station ISHI over one week beginning on June 17, 2025. The float solution exhibits pronounced long-term fluctuations, particularly in the east\u0026ndash;west (EW) component. These fluctuations were substantially suppressed in the gUPD solution; however, short-term noise was noticeably amplified, resulting in STDs comparable to those of the float solution. In addition, ambiguity resolution failed for several percent of the processed epochs. In contrast, the rUPD solution effectively mitigated both long-term fluctuations and short-term noise while achieving near complete ambiguity fixing. Similar improvements were observed at stations AIRA, CCJ2, STK2, and even MCIL located outside the regional network (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e; see Fig.\u0026nbsp;1 for station locations).\u003c/p\u003e \u003cp\u003eFigure 4 presents the 14-day median of station-specific daily STDs together with regression lines. Compared with the float solution, the gUPD solution shows substantial improvement in the EW component, whereas only marginal improvements are observed in the north\u0026ndash;south (NS) and up\u0026ndash;down (UD) components. In contrast, the rUPD solution demonstrates further improvement in the EW component and noticeable improvements in the NS and UD components. The daily STDs averaged across all available stations are 0.83 cm, 0.91 cm, and 2.99 cm for the EW, NS, and UD components, respectively (as indicated along the y-axis of the lower panel in Fig.\u0026nbsp;4). Accordingly, rUPD achieved millimeter-level precision in the horizontal components with an improvement of up to 40% in the EW component relative to the float solution. The histogram of daily FR indicates that rUPD reliably fixes ambiguities whereas gUPD frequently fails to do so (Fig.\u0026nbsp;5).\u003c/p\u003e \u003cp\u003eNoise characteristics in the period domain are illustrated using amplitude spectra (Fig.\u0026nbsp;6a). In the short-period domain (\u0026lt;\u0026thinsp;2\u0026ndash;3 hours), the rUPD solution retained a noise level comparable to that of the float solution, whereas the gUPD solution exhibited degraded performance. In the long-period domain (\u0026gt;\u0026thinsp;2\u0026ndash;3 hours), noise behavior is dominated by GPS orbital resonances (Zajdel et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), manifested as spectral peaks at the GPS sidereal day and its harmonics (a rightmost panel of Fig.\u0026nbsp;6a). These peaks were substantially attenuated in the rUPD solution, particularly in the EW component. Amplitudes at representative periods further demonstrate that rUPD consistently outperforms the other strategies across the examined frequency bands (Fig.\u0026nbsp;6b).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Noise characteristics across different receiver network configurations\u003c/h2\u003e \u003cp\u003eFigure 7 presents the scatter plots of the 14-day median daily STD, labeled according to the receiver type on the user-side. Adopting consistent receiver types between network- and user-sides (i.e. ALLOY for rUPD and NET-G5 for rUPDwTPS) results in slight improvements in the UD component, while producing minimal changes in the EW and NS components. Daily FR values also clustered within a higher range when receiver types were consistent between the network- and user-sides (Fig.\u0026nbsp;8).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Detectability of geophysical signals across multiple timescales\u003c/h2\u003e \u003cp\u003eIn this section, each kinematic analysis strategy is applied to representative crustal deformation events\u0026mdash;including co-seismic deformation, earthquake swarms, and volcanic inflation episode\u0026mdash;to evaluate the capability of regional PPP-AR in detecting geophysical signals across multiple timescales.\u003c/p\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e3.3.1 M7.1 earthquake at the Hyuganada Sea\u003c/h2\u003e \u003cp\u003eThe Hyuganada Sea is a seismically active region as the Philippine Sea Plate subducts beneath the continental plate. M7\u0026ndash;8 class earthquakes have repeatedly occurred in the plate boundary. On August 8, 2024, an M7.1 interplate earthquake occurred at a depth of 31 km (Fig.\u0026nbsp;9a). Significant crustal deformation was observed near the epicenter, with a maximum displacement of approximately 13 cm toward the east-southeast at station 1088 (Fig.\u0026nbsp;9b). Residual displacement fields referenced to the F5 solution indicate that gUPD is comparatively noisier than the other kinematic strategies, as reflected by its lower VR value (Fig.\u0026nbsp;9d). Receiver-type-specific VR values indicate that the deformation field becomes more consistent when receiver types match between the network- and user-sides than when mismatches occur (Fig.\u0026nbsp;9e, 9f). Similar to the receiver-type-consistent cases of rUPD and rUPDwTPS, displacement field derived from the float solutions remain highly consistent with the F5, regardless of user-side receiver type (Fig.\u0026nbsp;9c). At station 1086, the eastward displacement signal was well captured by the float and receiver-consistent PPP-AR (i.e. rUPD) solutions. In contrast, the gUPD solution exhibits coordinate jumps during the pre-seismic period, attributable to failures in ambiguity fixing (Fig.\u0026nbsp;10). Similar jumps were observed in the receiver-inconsistent PPP-AR solution (i.e. rUPDwTPS) during the post-seismic period.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e3.3.2 Earthquake swarms at the Tokara Islands\u003c/h2\u003e \u003cp\u003eThe Tokara Islands, situated along the Nansei Islands Trench, are prone to recurrent earthquake swarms (HERP, 2025). In the northwestern offshore region of Takarajima Island, one of the Tokara Islands, seismicity began to intensify around 05:00 UTC on July 2, 2025 (Fig.\u0026nbsp;11a). The largest event (M5.6) occurred at 06:26 UTC, after which the seismicity diminished (shaded period in Fig.\u0026nbsp;11c). Static solutions indicate a southward displacement of approximately 4 cm at station 1243 (Fig.\u0026nbsp;11b). Although the rUPD time series exhibits moderate noise, it clearly reveals a gradual southward displacement during the episode, while the EW component remains stable. Notably, the kinematic solution demonstrates that the deformation was not instantaneous but evolved progressively in association with the earthquake swarms. This is consistent with the findings of Okada et al. (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). Unlike position-domain filtering approaches, PPP-AR mitigates hardware delays and orbit errors directly at the observation level. Consequently, long-term noise is reduced without distorting the underlying geophysical signal. In contrast, the float and gUPD solutions remain more affected by long-term noises, thereby reducing the detectability of the evolving deformation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e3.3.3 Volcanic inflation episode at Ioto Island\u003c/h2\u003e \u003cp\u003eIoto Island is a tectonically active volcanic island located in the southern part of Japan. There have been observed not only the island-scale long-term uplift at several tens of centimeters per year, but also episodic deformation events associated with earthquakes and volcanic activity. Around August 30, 2025, volcanic earthquakes began to increase. Prior to August 29, daily event counts remained below 20\u0026ndash;30 events per day; however, earthquakes increased sharply to 153 events on August 30, 685 on August 31, and 700 on September 1 (based on local time; preliminary counts from JMA, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). At 10:14 UTC on September 1, a volcanic eruption occurred at Chidorigahama, in the western part of the island (Fig.\u0026nbsp;12a), followed by continued eruption over several days (JMA, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eGNSS static analysis indicates that station 0604, located in the eastern part of the island, experienced gradual uplift and northward displacement synchronous with the increase in seismicity, followed by stagnation during the eruption period (Fig.\u0026nbsp;12b). The kinematic relative positioning solution using a tectonically stable reference station 0603 (baseline length of approximately 220 km) is significantly contaminated by short-term noise. For such a long baseline, troposphere and ionosphere errors are no longer effectively cancelled because of reduced spatial correlation. In contrast, the rUPD solution tracks the deformation signal with higher precision than the relative positioning solution. Even so, the noise level remained higher than that observed in the previous case studies because only GPS observations were available. Divergent behaviors are observed in both the relative positioning and rUPD solutions, likely attributable to orbit or satellite clock instabilities. Notably, PPP-AR enables nationwide unified deformation monitoring once a regional network is established, whereas relative positioning requires careful selection of a suitable reference station for each case. This advantage is particularly significant in geographically constrained areas such as remote islands, where the availability of suitable reference stations is inherently limited.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eIn the previous chapter, we demonstrated that employing a regional network, together with ensuring receiver-type consistency between the network- and user-sides, is essential for achieving reliable ambiguity resolution and, consequently, extending the applicability of kinematic positioning to sub-daily deformation monitoring. In this chapter, we discuss these improvements with a focus on phase bias estimates and propose an optimized network configuration.\u003c/p\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Phase biases under different network configurations\u003c/h2\u003e \u003cp\u003eFigures 13 and 14 present the time series of phase bias estimates for GPS and Galileo satellites, respectively. For visual clarity, integer offsets were applied where necessary to maintain continuity in the time series. gUPD provides continuous phase bias estimates owing to its global tracking capability; however, its time series exhibits greater dispersion than that of rUPD, including several divergence episodes in the NL phase biases. In contrast, rUPD yields comparatively stable and smooth time series for both WL and NL phase biases. These differences are consistent with the phase bias time series reported by Wang et al. (\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), who observed similar contrasts between global and regional networks in the United States. Notably, rUPD exhibits larger fluctuations at the beginning of each arc, reflecting the convergence process from the moment satellites first become visible within the regional network until the phase bias estimates stabilize. These low-quality data are effectively filtered by user-side quality control procedures and therefore do not significantly degrade ambiguity resolution reliability. In addition, simultaneous discontinuities were observed across satellites in the rUPD solutions. These correspond to changes in the reference satellite or reinitialization of the ambiguity datum. As long as the same reference satellite and ambiguity datum are consistently adopted in user-side, reliable ambiguity resolution can still be achieved. Furthermore, no systematic differences were detected among GPS block types or between Galileo satellite generations (IOV versus FOC).\u003c/p\u003e \u003cp\u003eHistograms of phase bias residuals were computed over the entire evaluation period, using all available stations and satellites (Fig.\u0026nbsp;15a, b). The phase bias residual is defined as the fractional part of the float ambiguity after applying the phase bias correction (Li et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Compared with gUPD, rUPD exhibits sharper distributions centered near zero, indicating that the regional network configuration improves phase bias estimates. The underlying mechanism may be explained by the absorption of spatially correlated unmodeled errors (e.g., orbits and troposphere) into the phase biases (Wang et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Zeng et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Meanwhile, a number of observations were excluded from WL ambiguity resolution in rUPD due to Flag U (Fig.\u0026nbsp;15c). These exclusions correspond to unstable phase bias estimates at the beginning of each arc as mentioned in the previous section. In a regional network, satellites observed at low elevation angles are often tracked only from a single direction, resulting in fewer contributing stations and poor observation geometry. Minor rises are observed in the tails of the WL residual distribution for rUPD (Fig.\u0026nbsp;15a); these are flagged as Flag F and excluded from the WL ambiguity resolution procedure. These rises comes from larger residuals in GPS Block III observations, as described in the subsequent paragraph.\u003c/p\u003e \u003cp\u003eFigure 16 presents the WL phase bias residuals for different satellite observations, color-coded according to user-side receiver type. When receiver types are inconsistent between the network- and user-side, residuals for GPS Block III exhibit broader distribution with half-cycle shift. For GPS Block II series and Galileo, no shift is observed; however, the distributions remain noticeably wider. One plausible explanation for this degraded performance is the receiver-dependency of pseudorange biases (e.g., Hauschild and Montenbruck, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), which cannot be fully corrected through the application of constant DCB products. These receiver-dependent biases propagate into the WL phase bias estimates through the Melbourne\u0026ndash;W\u0026uuml;bbena linear combination, thereby impairing the recovery of the integer property of WL ambiguities on the user-side. However, the reason why only Block III residuals exhibit a systematic shift remains unclear. Conversely, when receiver types are consistent, the residuals distributions for all satellite types\u0026mdash;including Block III\u0026mdash; are more sharply concentrated around zero. This suggests that receiver-dependent contamination in the phase bias estimates is effectively cancelled under a receiver-type-consistent configuration.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Optimized network configuration for deformation monitoring\u003c/h2\u003e \u003cp\u003eIn this section, we propose a novel PPP-AR framework for monitoring regional deformation across multiple temporal scales. This framework consists of three principal components: station screening, phase bias estimation using receiver-type-specific regional networks, and user-side PPP-AR with the same receiver type employed in phase bias estimation (Fig.\u0026nbsp;17). Within this framework, spatially correlated unmodeled errors are effectively separated from ambiguity parameters, enabling nationwide detection of instantaneous displacements as well as sub-daily deformations ranging from several centimeters to a decimeter. An operational advantage of the proposed framework is that, once networks are configured, comprehensive network-wide deformation monitoring can be performed without case-specific reconfiguration. Moreover, unlike conventional relative positioning, the framework avoids baseline-dependent biases and enables precise positioning even in geographically constrained regions such as remote islands.\u003c/p\u003e \u003cp\u003eEven when a dense network is initially deployed, real-time operation may necessitate phase bias estimation under a sparse network due to data gaps caused by hardware failures or communication disruptions. The tolerance of the framework to station outages within the network was therefore evaluated. Figures\u0026nbsp;18 and 19 present the STD and FR obtained when phase biases were estimated using progressively fewer stations. The STD show no significant degradation for either sparse network. However, when the network was reduced to only 10 stations, the FR decreased markedly. By contrast, no significant degradation was observed when 20 stations were retained. These results suggest that 20 stations represent a practical lower bound for stable phase bias estimation in Japan. This threshold provides a quantitative reference for developing operational policies related to station maintenance, redundancy planning, and network management.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Applicability to continuous deformation monitoring\u003c/h2\u003e \u003cp\u003eIn light of the anticipated megathrust earthquake along the Nankai Trough, we investigated whether continuous deformation monitoring remains feasible after network stations experience significant co-seismic displacements. Using the 2011 off the Pacific coast of Tohoku Earthquake (Mw9.0; hereafter referred to as the Tohoku-Oki earthquake) as a case study, two regional networks were constructed: one comprising Trimble 5700 receivers and the other comprising TOPCON NET-G3 receivers (Fig.\u0026nbsp;20). Phase biases were estimated after excluding near-field stations identified by co-seismic displacements greater than 10 cm, followed by PPP-AR analysis with the corresponding receiver types. The 10 cm threshold is adopted based on past studies that co-seismic displacement of several centimeters can be reliably detected using kinematic positioning techniques (e.g. Kawamoto et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Consequently, long-term fluctuations in the float solutions were substantially suppressed, enabling detection of very early post-seismic deformation signals, as evidenced by gradual eastward movements at stations 0171 and 0176 (Figure \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e). A more spatially coherent post-seismic deformation field was obtained compared with that derived from the float solutions, which show systematic shifts at most stations and exhibit exceptional behavior at others, as evident from the residual plot (Fig.\u0026nbsp;21). Time series at stations sufficiently distant from the epicenter (e.g. 0087 and 0450 in Figure \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e) showed stable behavior. However, when near-field stations were included in phase bias estimation, occasional coordinate jump and offsets were observed, accompanied by a decreased ambiguity fixing rate (e.g. 0176 and 0450 in Figure \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e). Moreover, the spatial coherence of the post-seismic deformation field was slightly degraded, manifesting as residuals of 1\u0026ndash;2 cm for the NET-G3 receivers (Fig.\u0026nbsp;21). These subtle degradations would be caused by large co-seismic displacements that compromise the static analysis used for phase bias estimation, leading to unstable phase bias estimates.\u003c/p\u003e \u003cp\u003eTo ensure high-quality phase bias estimates following a major earthquake, stations affected by significant co-seismic displacement must be promptly identified and screened. We propose a station screening strategy in which stations exhibiting large static offsets\u0026mdash;derived from float solutions\u0026mdash;are immediately excluded from phase bias estimation. When static offsets are computed as the coordinate difference between 1 minute before and 15 minutes after the earthquake, stations with offsets exceeding 10 cm correspond to the filled symbols shown in Fig.\u0026nbsp;20. As demonstrated in the Results section, the float solution is capable of detecting instantaneous displacements with accuracy comparable to that of the regional PPP-AR solution. An additional advantage of exploiting the float solution is that it is inherently generated within the PPP-AR processing workflow, allowing the screening strategy to be implemented without sacrificing operational efficiency. It should be noted that, to avoid misidentifying surface-wave-induced displacements as permanent offsets, the time window after the earthquake must be dynamically adjusted based on hypocenter and the spatial extent of network.\u003c/p\u003e \u003cp\u003eHowever, station screening reduces the number of contributing stations, potentially degrading satellite tracking capability within the network. This reduction leads to shorter arcs in the phase bias estimates. For example, slight arc shortening was observed for G12 in 11:00 UTC, G22 in 17:00 UTC, and G32 in 22:00 UTC after the Tohoku-Oki earthquake (Figure \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003ea). Consequently, the risk of incorrect ambiguity fixing or failure to achieve ambiguity fixing increases. To mitigate this limitation, further integration of multi-GNSS constellations and the use of temporally extrapolated phase bias estimates would be effective. Moreover, regardless of station screening, phase biases were unavailable for the minimum number of satellites to required for PPP-AR (i.e., five) during the several tens of minutes immediately following the mainshock (Figure \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003eb). This limitation arises because traveling ionospheric disturbance excited by the mainshock induced cycle slips in the observation data, resulting in the exclusion of the affected observations. A geometry-free linear combination was employed as a cycle slip detector; however, it is sensitive to elevated ionospheric activity and may generate false detections under disturbed conditions. Multiple studies have proposed advanced cycle slip detection and correction methods under complex ionospheric conditions (Banville and Langley, 2012; Zhang et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Zhang and Li, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Incorporating these improvements would enable more robust and reliable monitoring of crustal deformation during gigantic earthquakes and their subsequent seismic events.\u003c/p\u003e \u003c/div\u003e"},{"header":"5 Conclusion and outlook","content":"\u003cp\u003eWe investigated how the network configuration\u0026mdash;specifically network scale and the receiver types\u0026mdash;in satellite phase bias estimation affects crustal deformation monitoring using kinematic PPP-AR. To evaluate its applicability to real-time operation, we processed roughly 1,300 GEONET stations across Japan with GFZ real-time orbit and clock products in a simulated real-time mode. While PPP float solutions are modulated by long-term noise associated with GPS orbital resonances and the PPP-AR implemented using a global network exhibits insufficient fixing rate (FR) and noisier coordinates time series, regional PPP-AR achieves nearly complete FR and the lowest noise levels across timescales ranging from several minutes to sub-daily periods. The daily STD reached the millimeter level in the horizontal components, representing an improvement of up to 40% in the EW component relative to the float solution. These improvements partly result from the receiver-type consistencies between phase bias estimation and user-side PPP-AR. In contrast, the use of inconsistent receiver types slightly degrades both STD and FR. Case studies of multiple deformation events demonstrate that receiver-type-consistent and regional PPP-AR enables monitoring of sub-daily deformation on the order of several centimeters to a decimeter as well as instantaneous displacements across Japan.\u003c/p\u003e \u003cp\u003eThese advances were achieved through improved phase bias estimates using the receiver-type-specific regional network, combined with user-side PPP-AR processing consistent with those estimates. Accordingly, we propose a framework in which phase biases are estimated from receiver-type-specific regional networks and user-side PPP-AR is performed using the same receiver type as that employed in the network. This framework enables continuous and comprehensive deformation monitoring within the network, including in geographically constrained regions. However, high quality of phase bias estimates needs to be continuously provided to ensure operational reliability. To achieve this, it is essential to maintain a sufficiently dense network configuration and implement an on-time station screening algorithm capable of addressing large displacement events. Sensitivity tests using sparse networks showed that 20 stations represent a practical lower bound for stable phase bias estimation in Japan.\u003c/p\u003e \u003cp\u003eFurther studies are required to resolve the systematic offset observed in GPS Block III phase bias residuals under receiver-type-inconsistent conditions. One potential approach is to calibrate this offset at the user-side prior to ambiguity resolution. Cui et al. (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) proposed a method in which receivers are grouped using K-means clustering based on WL phase bias residuals, and the resulting inter-group differences are used to correct systematic offsets. Consequently, both the fixing rate and station coordinate precision were restored to levels comparable to those of receiver-type-consistent conditions. Implementing such calibration would further enhance the effective utilization of GPS Block III, whose constellation will expand in the coming years. Therefore, even when receiver-type selection is constrained, station coordinate precision can be improved.\u003c/p\u003e \u003cp\u003eThe extension of regional phase bias estimation to multi-GNSS constellations should also be considered in future studies. In recent years, several analysis centers of IGS Real-Time Committee have provided real-time multi-GNSS OSB products (IGS, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). By directly applying OSB corrections to carrier phase observations, users can perform PPP-AR without explicitly accounting for frequency and signal type differences. However, currently available OSB products are global in scope and are not optimized for regional applications. As demonstrated earlier, the application of regional phase bias products yields substantially higher positioning precision than global products. With the large number of existing satellites and the planned expansion of Quasi-Zenith Satellite System, Japan is expected to increasingly benefit from enhanced multi-GNSS capabilities. Leveraging this opportunity, the development of regional real-time multi-GNSS phase bias products will further enhance PPP capabilities and expand applications in socioeconomic activities and geophysical monitoring.\u003c/p\u003e "},{"header":"Abbreviations","content":"\u003cp\u003eDCB differential code bias\u003c/p\u003e \u003cp\u003eFR fixing rate\u003c/p\u003e \u003cp\u003eGEONET GNSS Earth Observation NETwork System\u003c/p\u003e \u003cp\u003eGNSS Global Navigation Satellite System\u003c/p\u003e \u003cp\u003eGSI Geospatial Information Authority of Japan\u003c/p\u003e \u003cp\u003eHERP Headquarters of Earthquake Research Promotion\u003c/p\u003e \u003cp\u003eIGS International GNSS Service\u003c/p\u003e \u003cp\u003eJAXA Japan Aerospace Exploration Agency\u003c/p\u003e \u003cp\u003eJMA Japan Meteorological Agency\u003c/p\u003e \u003cp\u003eNL narrow-lane\u003c/p\u003e \u003cp\u003eOSB observable-specific bias\u003c/p\u003e \u003cp\u003ePPP precise point positioning\u003c/p\u003e \u003cp\u003ePPP-AR precise point positioning with ambiguity resolution\u003c/p\u003e \u003cp\u003eRTPPP Real-Time Precise Point Positioning\u003c/p\u003e \u003cp\u003eSTD standard deviation\u003c/p\u003e \u003cp\u003eUPD uncalibrated phase delay\u003c/p\u003e \u003cp\u003eVR variance reduction\u003c/p\u003e \u003cp\u003eWL wide-lane\u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe GNSS data of GEONET are available from\u0026nbsp;https://terras.gsi.go.jp/index.php and https://www.gsi.go.jp/ENGLISH/geonet_english.html. The GNSS data of IGS are available from https://cddis.nasa.gov/archive/gnss/data/daily/.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePart of this study was conducted during NT\u0026rsquo;s stay at GFZ Helmholtz Centre for Geosciences under the overseas research fellowship program for research related to space by the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan. This study was partly supported by the JST FOREST Program (Grant Number: JPMJFR202P, Japan). Funding was also provided by MEXT of Japan under the Third Earthquake and Volcano Hazards Observation and Research Program (Earthquake and Volcano Hazard Reduction Research). This work was supported by MEXT Coordination Funds, Japan Grant Number J013348.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNT designed the study, processed the data, and wrote the manuscript. YO, SH, and AB advised on the interpretation of results. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are grateful to Dr. Hiroshi Munekane and Dr. Tomokazu Kobayashi for insightful discussions. The RTPPP and GFZ real-time orbit and satellite clock products were provided by GFZ Helmholtz Centre for Geosciences. The GNSS data in the global network was obtained from IGS web site. The plate models by Iwasaki et al. (2015) were constructed from topography and bathymetry data by Geospatial Information Authority of Japan (250-m digital map), Japan Oceanographic Data Center (500m mesh bathymetry data, J-EGG500, http://www.jodc.go.jp/jodcweb/JDOSS/infoJEGG_j.html) and Geographic Information Network of Alaska, University of Alaska (Lindquist et al., 2004). English proofreading was provided by Editage (https://www.editage.com/).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eN.T. is a staff member at Geospatial Information Authority of Japan and a visiting scientist of the research group \u0026quot;Real-time GNSS\u0026quot; at GFZ Helmholtz Centre for Geosciences, Germany.\u003c/p\u003e\n\u003cp\u003eY.O. is a professor at the Graduate School of Science, Tohoku University, Japan.\u003c/p\u003e\n\u003cp\u003eS.H. is a scientist in the research group \u0026quot;Real-time GNSS\u0026quot; at GFZ Helmholtz Centre for Geosciences, Germany.\u003c/p\u003e\n\u003cp\u003eA.B. is the head of the research group \u0026quot;Real-time GNSS\u0026quot; at GFZ Helmholtz Centre for Geosciences, Germany.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEndnotes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAmante C, Eakins BW (2009) ETOPO1 arc-minute global relief model: procedures, data sources and analysis\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBanville S, Langley RB (2013) Mitigating the impact of ionospheric cycle slips in GNSS observations. J Geodesy 87(2):179\u0026ndash;193. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00190-012-0604-1\u003c/span\u003e\u003cspan address=\"10.1007/s00190-012-0604-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBilich A, Cassidy JF, Larson KM (2008) GPS seismology: Application to the 2002 M w 7.9 Denali fault earthquake. Bull Seismol Soc Am 98(2):593\u0026ndash;606. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1785/0220190113\u003c/span\u003e\u003cspan address=\"10.1785/0220190113\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrack A, He S, Du S, Wickert J (2025) Real-time GNSS for geosciences: Initial assessment of new data products from GFZ. EGU General Assembly 2025\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCollins P, Bisnath S, Lahaye F, H\u0026eacute;roux P (2010) Undifferenced GPS ambiguity resolution using the decoupled clock model and ambiguity datum fixing. Navigation 57(2):123\u0026ndash;135. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/j.2161-4296.2010.tb01772.x\u003c/span\u003e\u003cspan address=\"10.1002/j.2161-4296.2010.tb01772.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCui B, Li P, Wang J, Ge M, Schuh H (2021) Calibrating receiver-type-dependent wide-lane uncalibrated phase delay biases for PPP integer ambiguity resolution. J Geodesy 95(7):82. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00190-021-01524-6\u003c/span\u003e\u003cspan address=\"10.1007/s00190-021-01524-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFukuda J (2018) Variability of the space-time evolution of slow slip events off the Boso Peninsula, central Japan, from 1996 to 2014. J Geophys Research: Solid Earth 123(1):732\u0026ndash;760. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/2017JB014709\u003c/span\u003e\u003cspan address=\"10.1002/2017JB014709\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGe M, Gendt G, Rothacher M al, Shi C, Liu J (2008) Resolution of GPS carrier-phase ambiguities in precise point positioning (PPP) with daily observations. J Geodesy 82(7):389\u0026ndash;399. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00190-007-0187-4\u003c/span\u003e\u003cspan address=\"10.1007/s00190-007-0187-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeng J, Teferle FN, Shi C, Meng X, Dodson A, Liu J (2009) Ambiguity resolution in precise point positioning with hourly data. GPS Solutions 13(4):263\u0026ndash;270. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10291-009-0119-2\u003c/span\u003e\u003cspan address=\"10.1007/s10291-009-0119-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeng J, Bock Y (2016) GLONASS fractional-cycle bias estimation across inhomogeneous receivers for PPP ambiguity resolution. J Geodesy 90(4):379\u0026ndash;396. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00190-015-0879-0\u003c/span\u003e\u003cspan address=\"10.1007/s00190-015-0879-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeng J, Jiang P, Liu J (2017) Integrating GPS with GLONASS for high-rate seismogeodesy. Geophys Res Lett 44(7):3139\u0026ndash;3146. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/2017GL072808\u003c/span\u003e\u003cspan address=\"10.1002/2017GL072808\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeng J, Pan Y, Li X, Guo J, Liu J, Chen X, Zhang Y (2018) Noise characteristics of high-rate multi-GNSS for subdaily crustal deformation monitoring. J Geophys Research: Solid Earth 123(2):1987\u0026ndash;2002. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/2018JB015527\u003c/span\u003e\u003cspan address=\"10.1002/2018JB015527\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeng J, Chen X, Pan Y, Zhao Q (2019) A modified phase clock/bias model to improve PPP ambiguity resolution at Wuhan University. J Geodesy 1\u0026ndash;15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00190-019-01301-6\u003c/span\u003e\u003cspan address=\"10.1007/s00190-019-01301-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeng J, Xin S, Williams SD, Jiang W (2021) Comparing non-tidal ocean loading around the southern North Sea with subdaily GPS/GLONASS data. J Geophys Research: Solid Earth 126(3). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2020JB020685\u003c/span\u003e\u003cspan address=\"10.1029/2020JB020685\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. e2020JB020685\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeng J, Wen Q, Zhang Q, Li G, Zhang K (2022) GNSS observable-specific phase biases for all-frequency PPP ambiguity resolution. J Geodesy 96(2):11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00190-022-01602-3\u003c/span\u003e\u003cspan address=\"10.1007/s00190-022-01602-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGenrich JF, Bock Y (1992) Rapid resolution of crustal motion at short ranges with the Global Positioning System. J Geophys Research: Solid Earth 97(B3):3261\u0026ndash;3269. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/91JB02997\u003c/span\u003e\u003cspan address=\"10.1029/91JB02997\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHauschild A, Montenbruck O (2016) A study on the dependency of GNSS pseudorange biases on correlator spacing. GPS Solutions 20(2):159\u0026ndash;171. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10291-014-0426-0\u003c/span\u003e\u003cspan address=\"10.1007/s10291-014-0426-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeki K (2001) Seasonal modulation of interseismic strain buildup in northeastern Japan driven by snow loads. Science 293(5527):89\u0026ndash;92. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/DOI:%252010.1126/science.1061056\u003c/span\u003e\u003cspan address=\"https://doi.org/DOI:%252010.1126/science.1061056\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHERP (2024) Evaluation of the 2024 Noto Peninsula Earthquakes. Earthquake Research Committee, The Headquarters of Earthquake Research Promotion. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://jishin.go.jp/main/index-e.html\u003c/span\u003e\u003cspan address=\"https://jishin.go.jp/main/index-e.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 26 January 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHERP, The Headquarters of Earthquake Research Promotion (2025) Seismic Activity in the Ocean Near Tokara Islands (Near Kodakarajima Island). Earthquake Research Committee,. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://jishin.go.jp/main/chousa/25jul_tokara_2/p01-e.htm\u003c/span\u003e\u003cspan address=\"https://jishin.go.jp/main/chousa/25jul_tokara_2/p01-e.htm\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 26 January 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIGS (2026) IGS RTS Products. International GNSS Service. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://igs.org/rts/products/\u003c/span\u003e\u003cspan address=\"https://igs.org/rts/products/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 26 January 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIwasaki T, Sato H, Shinohara M, Ishiyama T, Hashima A (2015) Fundamental structure model of island arcs and subducted plates in and around Japan. 2015 Fall Meeting\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJAXA (2024) MALIB: MADOCA-PPP Library. GitHub repository. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/JAXA-SNU/MALIB\u003c/span\u003e\u003cspan address=\"https://github.com/JAXA-SNU/MALIB\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 26 January 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang J, Bock Y, Klein E (2021) Coevolving early afterslip and aftershock signatures of a San Andreas fault rupture. Sci Adv 7(15):eabc1606. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/DOI:%252010.1126/sciadv.abc1606\u003c/span\u003e\u003cspan address=\"https://doi.org/DOI:%252010.1126/sciadv.abc1606\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang X (2025) Multi-GNSS Real-time Precise Satellite Clock Estimation. Dissertation, Technische Universit\u0026auml;t Berlin\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJMA (2018) The Seismological Bulletin of Japan. Japan Meteorological Agency. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.data.jma.go.jp/eqev/data/bulletin/index_e.html\u003c/span\u003e\u003cspan address=\"https://www.data.jma.go.jp/eqev/data/bulletin/index_e.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 26 January 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJMA (2025) Explanatory Report on Volcanic Activity (Ioto Island, Japan). Japan Meteorological Agency. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.data.jma.go.jp/vois/data/report/monthly_v-act_doc/monthly_vact_vol.php?id=329\u003c/span\u003e\u003cspan address=\"https://www.data.jma.go.jp/vois/data/report/monthly_v-act_doc/monthly_vact_vol.php?id=329\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. (in Japanese) Accessed 26 January 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJMA (2026) Nankai Trough Earthquake Extra Information. Japan Meteorological Agency. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.jma.go.jp/jma/en/Activities/earthquake.html\u003c/span\u003e\u003cspan address=\"https://www.jma.go.jp/jma/en/Activities/earthquake.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 26 January 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKawamoto S, Hiyama Y, Ohta Y, Nishimura T (2016) First result from the GEONET real-time analysis system (REGARD): the case of the 2016 Kumamoto earthquakes. Earth Planets and Space 68:190. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s40623-016-0564-4\u003c/span\u003e\u003cspan address=\"10.1186/s40623-016-0564-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKawamoto S, Ohta Y, Hiyama Y, Todoriki M, Nishimura T, Furuya T, Sato Y, Yahagi T, Miyagawa K (2017) REGARD: A new GNSS-based real-time finite fault modeling system for GEONET. J Geophys Research: Solid Earth 122(2):1324\u0026ndash;1349. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/2016JB013485\u003c/span\u003e\u003cspan address=\"10.1002/2016JB013485\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKing MA, Watson CS, Penna NT, Clarke PJ (2008) Subdaily signals in GPS observations and their effect at semiannual and annual periods. Geophys Res Lett 35(3). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2007GL032252\u003c/span\u003e\u003cspan address=\"10.1029/2007GL032252\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKouba J, H\u0026eacute;roux P (2001) Precise point positioning using IGS orbit and clock products. GPS Solutions 5(2):12\u0026ndash;28. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/PL00012883\u003c/span\u003e\u003cspan address=\"10.1007/PL00012883\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLarson KM, Bilich A, Axelrad P (2007) Improving the precision of high-rate GPS. J Geophys Research: Solid Earth 112. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2006JB004367\u003c/span\u003e\u003cspan address=\"10.1029/2006JB004367\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLaurichesse D, Mercier F, Berthias J-P, Broca P, Cerri L (2009) Integer ambiguity resolution on undifferenced GPS phase measurements and its application to PPP and satellite precise orbit determination. Navigation 56(2):135\u0026ndash;149. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/j.2161-4296.2009.tb01750.x\u003c/span\u003e\u003cspan address=\"10.1002/j.2161-4296.2009.tb01750.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi P, Zhang X, Guo F (2017) Ambiguity resolved precise point positioning with GPS and BeiDou. J Geodesy 91(1):25\u0026ndash;40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00190-016-0935-4\u003c/span\u003e\u003cspan address=\"10.1007/s00190-016-0935-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi X, Li X, Yuan Y, Zhang K, Zhang X, Wickert J (2018) Multi-GNSS phase delay estimation and PPP ambiguity resolution: GPS, BDS, GLONASS, Galileo. J Geodesy 92(6):579\u0026ndash;608. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00190-017-1081-3\u003c/span\u003e\u003cspan address=\"10.1007/s00190-017-1081-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLindquist KG, Engle K, Stahlke D, Price E (2004) Global topography and bathymetry grid improves research efforts. Eos Trans Am Geophys Union 85(19):186\u0026ndash;186. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2004EO190003\u003c/span\u003e\u003cspan address=\"10.1029/2004EO190003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMalservisi R, Schwartz SY, Voss N, Protti M, Gonzalez V, Dixon TH, Jiang Y, Newman AV, Richardson J, Walter JI (2015) others Multiscale postseismic behavior on a megathrust: The 2012 Nicoya earthquake, Costa Rica. Geochemistry, Geophysics, Geosystems 16(6):1848\u0026ndash;1864. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/2015GC005794\u003c/span\u003e\u003cspan address=\"10.1002/2015GC005794\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMilliner C, B\u0026uuml;rgmann R, Inbal A, Wang T, Liang C (2020) Resolving the kinematics and moment release of early afterslip within the first hours following the 2016 Mw 7.1 Kumamoto earthquake: Implications for the shallow slip deficit and frictional behavior of aseismic creep. J Geophys Research: Solid Earth 125(9). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2019JB018928\u003c/span\u003e\u003cspan address=\"10.1029/2019JB018928\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. e2019JB018928\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiyazaki S, Segall P, Fukuda J, Kato T (2004) Space time distribution of afterslip following the 2003 Tokachi-oki earthquake: Implications for variations in fault zone frictional properties. Geophys Res Lett 31(6). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2003GL019410\u003c/span\u003e\u003cspan address=\"10.1029/2003GL019410\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMunekane H (2012) Coseismic and early postseismic slips associated with the 2011 off the Pacific coast of Tohoku Earthquake sequence: EOF analysis of GPS kinematic time series. Earth Planets and Space 64(12):1077\u0026ndash;1091. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5047/eps.2012.07.009\u003c/span\u003e\u003cspan address=\"10.5047/eps.2012.07.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMurray JR, Bartlow N, Bock Y, Brooks BA, Foster J, Freymueller J, Hammond WC, Hodgkinson K, Johanson I, L\u0026oacute;pez-Venegas A (2020) others Regional global navigation satellite system networks for crustal deformation monitoring. Seismological Research Letters 91(2A):552\u0026ndash;572. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1785/0220190113\u003c/span\u003e\u003cspan address=\"10.1785/0220190113\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNishimura T, Matsuzawa T, Obara K Detection of short-term slow slip events along the Nankai Trough, southwest Japan, using GNSS data. Journal of Geophysical Research: Solid Earth 118(6):3112\u0026ndash;3125., Odijk D, Zhang B, Khodabandeh A, Odolinski R, Teunissen PJG (2013) (2016) On the estimability of parameters in undifferenced, uncombined GNSS network and PPP-RTK user models by means of S-system theory. Journal of Geodesy 90:15\u0026ndash;44. https://doi.org/10.1007/s00190-015-0854-9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOhno K, Ohta Y, Hino R, Koshimura S, Musa A, Abe T, Kobayashi H (2022) Rapid and quantitative uncertainty estimation of coseismic slip distribution for large interplate earthquakes using real-time GNSS data and its application to tsunami inundation prediction. Earth Planets and Space 74(24):1\u0026ndash;18. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s40623-022-01586-6\u003c/span\u003e\u003cspan address=\"10.1186/s40623-022-01586-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOhno K, Ohta Y, Takamatsu N, Munekane H, Iguchi M (2024) Real-time modeling of transient crustal deformation through the quantification of uncertainty deduced from GNSS data. Earth Planets and Space 76(1):140. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s40623-024-02068-7\u003c/span\u003e\u003cspan address=\"10.1186/s40623-024-02068-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOhta Y, Meilano I, Sagiya T, Kimata T, Hirahara K (2006) Large surface wave of the 2004 Sumatra-Andaman earthquake captured by the very long baseline kinematic analysis of 1-Hz GPS data. Earth Planets and Space 58(2):153\u0026ndash;157. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/BF03353372\u003c/span\u003e\u003cspan address=\"10.1186/BF03353372\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOhta Y, Kobayashi T, Tsushima H, Miura S, Hino R, Takasu T, Fujimoto H, Iinuma T, Tachibana K, Demachi T (2012) others Quasi real-time fault model estimation for near-field tsunami forecasting based on RTK-GPS analysis: Application to the 2011 Tohoku-Oki earthquake (Mw 9.0). Journal of Geophysical Research: Solid Earth 117(B2). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2011JB008750\u003c/span\u003e\u003cspan address=\"10.1029/2011JB008750\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOkada Y, Ohta Y, Ohtate M, Ito Y, Ohzono M, Yakiwara H, Nakao S (2026) Recurrent earthquake swarms and transient crustal deformation off the Tokara Islands, southern Japan: Insights from the 2025 swarm sequence (Preprint). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.21203/rs.3.rs-8399343/v1\u003c/span\u003e\u003cspan address=\"10.21203/rs.3.rs-8399343/v1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Research Square\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOzawa S, Nishimura T, Suito H, Kobayashi T, Tobita M, Imakiire T (2011) Coseismic and postseismic slip of the 2011 magnitude-9 Tohoku-Oki earthquake. Nature 475(7356):373\u0026ndash;376. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nature10227\u003c/span\u003e\u003cspan address=\"10.1038/nature10227\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSagiya T, Miyazaki S, Tada T (2000) Continuous GPS array and present-day crustal deformation of Japan. Pure appl Geophys 157(11):2303\u0026ndash;2322. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/PL00022507\u003c/span\u003e\u003cspan address=\"10.1007/PL00022507\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaito K, Takahashi H, Ohzono M, Ohta Y (2025) Dynamic stress estimation during the M5.7 large remotely triggered earthquake due to the 2016 M7.0 Kumamoto earthquake, Kyushu, Japan, using dense strong motion and high-rate GNSS data. Seismol Res Lett 96(5):2848\u0026ndash;2855. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1785/0220240311\u003c/span\u003e\u003cspan address=\"10.1785/0220240311\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchaer S, Villiger A, Arnold D, Dach R, Prange L, J\u0026auml;ggi A (2021) The CODE ambiguity-fixed clock and phase bias analysis products: generation, properties, and performance. J Geodesy 95(7):81. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00190-021-01521-9\u003c/span\u003e\u003cspan address=\"10.1007/s00190-021-01521-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchwarz CR, Snay RA, Soler T (2009) Accuracy assessment of the National Geodetic Survey\u0026rsquo;s OPUS-RS utility. GPS Solutions 13(2):119\u0026ndash;132. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10291-008-0105-0\u003c/span\u003e\u003cspan address=\"10.1007/s10291-008-0105-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTakamatsu N, Muramatsu H, Abe S, Hatanaka Y, Furuya T, Kakiage Y, Ohashi K, Kato C, Ohno K, Kawamoto S (2023) New GEONET analysis strategy at GSI: daily coordinates of over 1300 GNSS CORS in Japan throughout the last quarter century. Earth Planets and Space 75(1):49. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s40623-023-01787-7\u003c/span\u003e\u003cspan address=\"10.1186/s40623-023-01787-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang CH, Hsu YJ, Okada Y, Ohta Y, Sagiya T, Jian PR, Tamura Y, Jike T (2025) Geodetic evidence of shallow slow-slip phenomena beneath the southern Ryukyu forearc. Geophysical Research Letters 52, e2025GL114742. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2025GL114742\u003c/span\u003e\u003cspan address=\"10.1029/2025GL114742\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTsuji H, Hatanaka Y (2018) GEONET as infrastructure for disaster mitigation. J Disaster Res 13(3):424\u0026ndash;432. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.20965/jdr.2018.p0424\u003c/span\u003e\u003cspan address=\"10.20965/jdr.2018.p0424\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTwardzik C, Vergnolle M, Sladen A, Avallone A (2019) Unravelling the contribution of early postseismic deformation using sub-daily GNSS positioning. Sci Rep 9(1):1775. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-019-39038-z\u003c/span\u003e\u003cspan address=\"10.1038/s41598-019-39038-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang S, Li B, Li X, Zang N (2018) Performance analysis of PPP ambiguity resolution with UPD products estimated from different scales of reference station networks. Adv Space Res 61(1):385\u0026ndash;401. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.asr.2017.09.005\u003c/span\u003e\u003cspan address=\"10.1016/j.asr.2017.09.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZajdel R, Sośnica K, Bury G, Dach R, Prange L, Kazmierski K (2021) Sub-daily polar motion from GPS, GLONASS, and Galileo. J Geodesy 95(1):3. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00190-020-01453-w\u003c/span\u003e\u003cspan address=\"10.1007/s00190-020-01453-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZeng P, Zhang Z, Wen Y, He X, He L, Li M, Chen W (2023) Properties of multi-GNSS uncalibrated phase delays with considering satellite systems, receiver types, and network scales. Satell Navig 4(1):19. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s43020-023-00110-9\u003c/span\u003e\u003cspan address=\"10.1186/s43020-023-00110-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang X, Li X (2012) Instantaneous re-initialization in real-time kinematic PPP with cycle slip fixing. GPS Solutions 16(3):315\u0026ndash;327. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10291-011-0233-9\u003c/span\u003e\u003cspan address=\"10.1007/s10291-011-0233-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang X, Zhu F, Zhang Y, Mohamed F, Zhou W (2019) The improvement in integer ambiguity resolution with INS aiding for kinematic precise point positioning. J Geodesy 93(7). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00190-018-1222-3\u003c/span\u003e\u003cspan address=\"10.1007/s00190-018-1222-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang X, Zhang Y, Zhu F (2020) A method of improving ambiguity fixing rate for post-processing kinematic GNSS data. Satell Navig 1(1):20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s43020-020-00022-y\u003c/span\u003e\u003cspan address=\"10.1186/s43020-020-00022-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Z, Li B, Shen Y (2017) Comparison and analysis of unmodelled errors in GPS and BeiDou signals. Geodesy Geodyn 8(1):41\u0026ndash;48. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.geog.2016.09.005\u003c/span\u003e\u003cspan address=\"10.1016/j.geog.2016.09.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Z, Zeng J, Li B, He X (2023) Principles, methods and applications of cycle slip detection and repair under complex observation conditions. J Geodesy 97(5):50. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00190-023-01743-z\u003c/span\u003e\u003cspan address=\"10.1007/s00190-023-01743-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZumberge JF, Heflin MB, Jefferson DC, Watkins MM, Webb FH (1997) Precise point positioning for the efficient and robust analysis of GPS data from large networks. J Geophys Research: Solid Earth 102(B3):5005\u0026ndash;5017. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/96JB03860\u003c/span\u003e\u003cspan address=\"10.1029/96JB03860\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"earth-planets-and-space","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"epsp","sideBox":"Learn more about [Earth, Planets and Space](http://earth-planets-space.springeropen.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/epsp/default.aspx","title":"Earth, Planets and Space","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"GNSS, GEONET, PPP (precise point positioning), PPP-AR (PPP with ambiguity resolution), Phase bias, UPD (uncalibrated phase delay), Regional network, Receiver type consistency","lastPublishedDoi":"10.21203/rs.3.rs-9135824/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9135824/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eKinematic Precise Point Positioning (PPP) is a powerful technique for Global Navigation Satellite System (GNSS)-based real-time crustal deformation monitoring. However, PPP float solutions are often affected by sub-daily periodic fluctuations, caused by orbit errors and atmospheric delays, thereby motivating the adoption of PPP with ambiguity resolution (PPP-AR). To enhance PPP-AR performance in the sub-daily domain, we investigated the optimized network configuration for phase bias estimation with respect to network scale and receiver type consistency. Using GEONET observations in Japan together with GFZ real-time orbit and clock products, we sequentially estimated satellite phase biases from receiver-type-specific regional networks and, by applying these estimates, processed 30-s GEONET data from approximately 1,300 stations in a simulated real-time kinematic PPP-AR mode. Two weeks of data in 2025 were analyzed to assess noise characteristics, together with case studies of recent seismic and volcanic deformation events. The results demonstrate significant suppression of long-term (\u0026gt;\u0026thinsp;2\u0026ndash;3 h) coordinate fluctuations associated with GPS satellite orbital resonance while retaining short-term (\u0026lt;\u0026thinsp;2\u0026ndash;3 h) noise levels comparable to those of float solutions. With nearly complete ambiguity resolution, horizontal positioning precision reached the millimeter level (daily standard deviation), representing an improvement of up to 40% relative to float solutions. These improvements were achieved through enhanced phase bias estimates using regional networks and user-side PPP-AR that is consistent with the derived phase bias estimates. The enhanced performance enables the capture of not only co-seismic displacements but also deformations ranging from several centimeters to a decimeter that evolve over several hours to several days, including earthquake swarms, volcanic inflation, and very early post-seismic deformation. Based on these findings, we propose a novel PPP-AR framework in which phase biases are estimated from receiver-type-specific regional networks, followed by user-side PPP-AR employing the same receiver type as that used within the network. This framework extends the applicability of real-time PPP to the sub-daily domain and helps bridge the temporal gap in GNSS-based deformation monitoring.\u003c/p\u003e","manuscriptTitle":"A novel real-time PPP-AR framework for continuous crustal deformation monitoring across Japan","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-01 07:28:21","doi":"10.21203/rs.3.rs-9135824/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2026-04-22T13:23:54+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-30T05:24:29+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-17T13:04:12+00:00","index":"","fulltext":""},{"type":"submitted","content":"Earth, Planets and Space","date":"2026-03-16T05:30:37+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"earth-planets-and-space","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"epsp","sideBox":"Learn more about [Earth, Planets and Space](http://earth-planets-space.springeropen.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/epsp/default.aspx","title":"Earth, Planets and Space","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"06bc20d4-1202-4d44-8ad5-080e24c5c17d","owner":[],"postedDate":"April 1st, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-01T07:28:25+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-01 07:28:21","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9135824","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9135824","identity":"rs-9135824","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}