Quality assessment of Continuous Operating Reference Stations(CORS) - GPS statio

Dniel Hernˊndez-Andrde , Rosendo Romero-Andrde ,*, Gopl Shrm ,Mnuel E. Trejo-Soto , Jun L. Cnills-Zvl

a Faculty of Earth and Space Sciences, Autonomous University of Sinaloa, 80013, Culiacˊan Rosales, Sinaloa, Mexico

b North Eastern Space Applications Centre, Umiam, 793103, Meghalaya, India

ABSTRACT The present analysis provided the quality assessment of 65 Continuous Operating Reference Stations(CORS)situated in and around Mexico using the data of about ten years (January 2010 to January 2020).To accomplish the assessment, we considered 4 quality indexes for performing the quality check,incluing the multipath effect on L1 and L2 bands, signal-to-noise ratio in both bands, cycle slips, and integrity.Meanwhile,a new reference parameter named regional values for Mexico(RVM)was calculated for the stations situated in Mexico. Additionally, an exhaustive analysis of the quality indicators in the worst and best cases was performed. The signal degradation was proved by a long-term time series and crustal deformation analysis in GAMIT/GLOBK platform.The results showed a strong correlation between integrity, cycle slips and daily observations time, and the multipath effect is strong in the case of outdated systems. The study indicates that the estimated quality indexes and values could be the basis for establishing new CORS in Mexico, and the errors corrections which cannot be mitigated in the postprocessing stage would greatly help utilize the data for different scientific applications.The results were supported by deformation analysis in part of Baja California Peninsula in Mexico indicating northern and eastern velocity vector of -3.08 ± 0.02 mm/yr, -7.85 ± 0.02 mm/yr and -0.07 ± 0.03 mm/yr, -9.33 ±0.02 mm/yr respectively at MPR1 and INEG stations.© 2022 Editorial office of Geodesy and Geodynamics. Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords:Multipath effect Signal-to-Noise ratio Cycle slip Integrity Regional quality indicators IGS indicators

1. Introduction

With the advance of the Global Navigations Satellite System(GNSS) and permanent geodetic networks, studies such as crustal deformation and geodynamic [1-6], monitoring and mapping [7],ionospheric total electron content (TEC) variations before the occurrence of earthquakes have been carried out[8-15]due to the high accuracy in positioning and a well-known structured signal in GNSS data [16]. Seismic and geodetic data were also used to determine the velocity field in collision intraplate zones for tectonic study and to evaluate of the strain accumulations [17-19]. One of the main aspects to consider in the GNSS related applications is the quality check of the GNSS observations [17-21]. The GNSS signals are affected by different factors such as systematic errors,antenna phase center, ephemeris errors, cycle slips, multipath effect, ionospheric and tropospheric effects,receiver noise,and other factors to degrade the results leading to erroneous estimates [20-25]. Many organizations have installed CORS worldwide, which can share GNSS data such as Global Positioning System (GPS) and other constellations enabling researchers to carry out the scientific analysis. GNSS data are being shared through different geodetic global networks such as International GNSS Service (IGS) and regional networks such as National Geodetic Survey (NGS), University NAVSTAR Consortium (UNAVCO) and Active National Geodetic Network (RGNA). With the introduction of many constellations like Quasi-Zenith Satellite System (QZSS), Galileo, GLONASS, and Beidou/Compass, the received signals in the geodetics networks are significantly impacted.Therefore,a greater number of constellations and satellites, increase the multipath effect and decrease the quality of positioning [26]. Nevertheless, when all constellations (GPS, GLONASS, Beidou and Galileo) are used, the positioning is more precise than a single constellation [25]. The precision achieved by utilizing available constellations also depends on the parameters (that influence the quality) and the positioning method (static relative or real-time kinematic) [27].

Numerous global state-of-art studies proved the necessity to perform a quality check of the received signal considering the multipath effect,data integrity,cycle slips,etc.,in the continuous and campaign mode geodetic networks [17,20,24,26-29].Simultaneously, a quality check was performed to compare the obtained values of signal-to-noise ratio,cycle slips and multipath effect in the observations with a G-NUT/ANUBIS and TEQC software with the inclusion of all constellations [30], showing equivalent solutions. A similar study was attempted to perform a quality check of the geodetic network of Topo-Iberia for a long time series (10 years) considering only the GPS constellation,which demonstrated that the IGS guidelines did not represent a regional network but were a good quality indicator [17]. Likewise,studies were conducted in Mexico analyzing only multipath effect as a principal source of signal degradation [24]. Therefore,it is critical to analyze the quality of data/information in the geodetic networks to achieve the precise positioning required for different applications [18,26,28]. The main quality indicators are multipath effect, integrity, cycle slips, and signal-to-noise ratio[18,26,28,29].

The present study focused on performing a quality check of 10 years of continuous GPS observations considering the values from the International GNSS Service (IGS) [31]. 65 CORS from several networks were utilized to estimate new values that represented the real situation in Mexico. The study proposed a new approach for estimation of regional values with descriptive statistics of the multipath, signal-to-noise ratio, cycle slips and integrity with a comprehensive data set. Additionally, an in-depth quality check for the worst and the best CORS were correlated. The signal degradation was proved by performing a long-term time series analysis with GAMIT/GLOBK, which was analyzed and compared with a previous study of crustal deformation in the northeast of Mexico [17-19,32-36].

2. Tectonic setting of the region

Mexico is one of the most seismically active countries globally due to the interaction of the Rivera, North American, Pacific and Cocos Plate(Fig.1)[33].The Rivera and Cocos plates subduct to the North American Plate along the Mesoamerican Trench [34]. The Caribbean Plate that includes a part of the State of Chiapas is in contact with the North American Plate by a transforming boundary and is influenced in its western limit by the subduction of the Cocos plate [35]. Most Mexican territory moves together with the North American Plate [34]. The northern part of Mexico (where the Baja California Peninsula is located,a part of the Pacific plate),is delimited by the North American Plate in a transtensional fault system. Previous studies performed through GPS campaign in the Jalisco Block have shown a motion of 2 mm/year to the southwestward with respect to the North American Plate, attributed to the tectonic and interseismic deformations[36].

3. Data used in the analysis

Data from NGS,UNAVCO,IGS and RGNA were utilized in present study.These data are publicly available at https://geodesy.noaa.gov/CORS/data.shtml, https://www.unavco.org/data/gps-gnss/gps-gnss.html, https://igs.org/products/and https://www.inegi.org.mx/app/geo2/rgna/. The data are available in network such as TLALOCNet(Trans-boundary, Land, and Atmosphere Long-term Observational and Collaborative Network), PBO (Plate Boundary Observatory),REGNOM (Northwest Mexico Geodetic Network), CORS and COCONet (Continuously Operating Caribbean GPS Observational Network). In present study, 65 CORS from UNAVCO (49), IGS (1),and NGS (15) were utilized. The data utilized is shown in Table 1 and Fig. 1 (created by ETOPO1 [37] and Generic Mapping Tools(GMT) [38]). The total duration of data used was from January 01,2010, to January 28, 2020 (10.07 years; 155,324 available days), as shown in Fig. 2.

Fig.1. Location of Continuous Reference GPS stations in Mexico utilized in the present study.

4. Methodology

4.1. Data preprocessing and quality check

Shell-C scripts on UNIX were used to download GPS observation,log, and broadcast ephemeris files. The original data were converted from HATANAKA [39] format to the Receiver Independent Exchange Format (RINEX) [40] with CRX2RNX [39]. The sampling rate of the data was modified to 30s. Only GPS observations were selected for the analysis as not all stations have the capability to track multiple constellations(Fig.3).The IGS recommended values[31]were calculated as:the multipath effect(MP)was calculated by the linear combination of carrier phase and pseudorange in Land Lin m with recommended values not above 0.30 m in Land L[41]. The integrity (INT) is theoretically calculated considering the complete observations concerning the possible observations in a percentage[21];the recommended value is ＞95%.Cycle slips were calculated from the carrier phase as per the relationship CSR=1000/(o/slps) [22,30]; where “o/slps” is the number of observations between cycle slips. Theoretically, the recommended value must be less than 1 per 1000 observations[31].The signal-tonoise ratio (SNR) values were considered as 36 dBHz [31] in Land L. All the quality indicators mentioned were calculated with TEQC software and evaluated with the recommended values from IGS [31,41].

4.2. Estimation of realistic values for Mexico CORS

IGS is recommended indicator values for the quality checks, as shown in Table 2.However,in Mexico,only one station agrees with the guideless and belongs to the IGS network. Hence, therecommended values from IGS could not be used as a representative indicator (values) for local GPS CORS in Mexico. Thus, all indicators were analyzed by applying descriptive statistics to remove the outliers and avoid a bias in the final value estimation. Finally,the value for each indicator was estimated(Table 2)according to Eq.(1) in order to give a regional and realistic representation of the actual situation in Mexico CORS.

Table 1 CORS data used in the analysis.

Fig. 2. Availability of GPS observations in the 10 years evaluated of NGS, IGS, and UNAVCO stations.

5. Results and discussions

5.1. Analysis of the quality indicators with the recommended values from IGS and regional values for Mexico (RVM)

Fig.4 presents the mean values of multipath effect in Land L(MP1 and MP2) of each CORS, which were 0.14 m-1.0 m and 0.19 m-0.94 m respectively. Only 7 CORS such as INEG, MEXI,MTY2,OAX2,TOL2,PLPX and PJZX accomplished the IGS standards.All the 65 CORS in the region were ordered concerning its effectiveness towards multipath, and 36 out of 65 CORS were found to accomplish the RVM standard for multipath effects. The behaviors of the signal deterioration could be related to the hardware and software updates [21,28,40]. It was also observed that the CORS having low multipath effects were the stations that had a recent update between 2016 and 2017.

In the case of SNR (Fig. 5), considering the IGS recommended values,only 17 CORS tracked strong signals on Land L.Besides,48 CORS tracked strong signals on Land weak signals on L. On the other hand, 24 and 18 CORS tracked a strong signal for the RVMSNR1 and RVM-SNR2, respectively. Therefore, considering RVM standard, only 6 CORS accomplished the recommended value.

In terms of MMD1, MMX1, MTP1, MSD1 and MPR1 from CORS network, a strong multipath effect and signal-to-noise ratio was observed, thus fulfilling the RVM recommended value, however,the inverse relation between signal-to-noise ratio and multipath effect, is not observed as mentioned by Kamatham [42]. The present behavior could be related to the antenna location and the nonavailability of recent updates. Equally the same firmware (2.003),antenna, and receiver (NOV WAASGII) were used in the same five CORS mentioned above with no registered updates for detail analysis.The values of the multipath effect and the signal-to-noise ratio indicated that the behavior in these stations was more related to the environment. Considering the IGS recommended values,MTY2, OAX2, TOL2, MEXI and INEGI stations from CORS/RGNA networks, shows an inverse behavior between the signal-to-noise ratio and multipath effect in Land Lwith the same receiver(LEICA GR10), antenna type (LEIAR10) and firmware (4.11/6.523).

RVM and IGS's recommended value for the CSR is similar,major being RVM with 0.04 CSR.Hence,only 48 stations from TLALOCNet,COCOnet, REGNOM, and CORS are under RVM and IGS recommended value. On the other hand,17 stations are over the recommended value of 1.04 CSR of RVM (Fig. 6). The stations with weak multipath effects and strong signals are the same stations that has a high number of cycle slips, which could be related to the software failure and wrong antenna location.

Fig.7 represents the calculated integrity considering the IGS and RVM recommended values. The results show that only 4 stations disagreed with IGS and RVM recommended values, where the station TNMT had the worst behavior with 77.45% of integrity. In addition,it was observed that 16 stations have a low integrity and 49 stations good integrity in concordance with RVM recommended values. Therefore, a total of 61 stations accomplished the IGS standard with good integrity.

Fig. 8 shows the analysis of 4 quality indicators in 6 recommended values, assigning a different color and a value of 1 when accomplishing the recommended value by the IGS and RVM. The maximum value will be 6 with an overlapping bar of 6 colors,which each recommended value is corresponding to MP1, MP2, SNR1,SNR2,CSR,and INT.However,when the recommended value is not accomplished,it will not be assigned a value,in this way only those recommended values that comply with the IGS(Fig.8 A)and RVM(Fig. 8 B) are shown.

Table 3 represents the total stations that agree with the IGS and RVM recommended values, where neither station agrees with allrecommended values. Considering the thresholds of the IGS and the global average of the 65 stations in the fulfilled variables, only 3.30 of 6 variables are accomplished,in the same way,for the RVM recommended values only 3.54 agree with the recommended values. Analysis reveals that there is no substantial difference between the IGS and RVM recommended values,nevertheless,in case of RVM, more stations are in agreement.

Table 2 Recommended values for the indicators MP, SNR, CSR, and INT from IGS and estimated regional values for Mexico (RVM).

Fig. 3. Flow-chart of methodology implemented in the present research.

Fig. 4. The average of MP1 and MP2 for each observation station. Green and purple lines represent RVM values and red line represents IGS recommended value.

Fig. 5. The average signal-to-noise ratio observed at each station. Green and purple lines represent RVM values and green line represents IGS recommended value.

5.2. Analysis of the quality indicators in the worst and best cases

The principal problem causing signal degradation is the multipath effect [19,31,41,43,44]. In present analysis INEG and MPR1 CORS were selected(best and worst behavior,respectively)as it has a low and high value of multipath effect to study the correlation between all indicators (SNR, MP, CSR, INT, standard deviations in the daily positioning solutions (σN， σE， σU), and the observation time per day).

Fig.9 represents the behavior of MPR1 CORS in respect to all the quality indicators.Analysis of the multipath effect showed the high value of MP1 and MP2, which was the worst cases with respect to this effect, with an average of 1.00 m and 0.94 m, respectively.These results could be related to a low signal-to-noise ratio.However, strong signals were present in Land L, showing a directly proportional relation. The multipath values were correlated with the integrity where the multipath effect value demonstrated a decrease with the loss of the signal,which was evident in the days with a greater number of cycles slips as revealed at the end of 2011 and early 2018.Similarly,the number of hours observed showed a strong correlation with the percentage of integrity,indicating that a lower integrity and shorter daily observation time generated higher standard deviation, which was found in the year 2013 being nonprecise solutions.

The irregular behavior found in time series in case of MPRI CORS(Fig. 9) could be related to the hardware and software issues, as it has not undergone any updates on receiver/antenna system.Based on the integrity, the reduced number of cycles slips per 1000 observations,and a strong SNR signal record with a strong multipath effect, it was likely that the surrounding environment is the primary source of the behavior.The large amounts of multipath could also be related to the bad antenna location. The MPR1 CORS was located on top of the tower control of the Puerto Vallarta Jalisco International Airport, and according to the National Geodetic Survey [45], the received signals can be adversely affected by radiofrequency.

INEG CORS has an adequate degree of compliance in the quality parameters based on the RVM and the IGS standard,being the only station that is part of the IGS network. The average values of multipath effect were 0.14 m in MP1 and 0.19 m for MP2, with 99.25%integrity and strong signals in both bands with SNR1 of 39.6 dBHz and SNR2 of 42.78 dBHz(Fig.10).Similarly,there was a strong correlation between cycle slips,integrity,and the daily observation time.The lower integrity,the lesser number of cycle slips,which is evident with a shorter time of daily observation. Similar behavior was shown by MEXI, MTY2, OAX2 and TOL2, and these stations conformed to the same quality indicators as INEG for both IGS and RVM values.

Fig. 6. The average of cycle slips per 1000 observations (CSR) for each station. Green and red lines represent RVM and IGS recommended values, respectively.

Fig. 7. The average integrity for each station. Red line: the recommended value of IGS. Green line: regional value for Mexico.

Fig.8. Global average of total indicator accomplished per station.When the indicator is accomplished, the corresponding nominal value will be 1.A)Total variables that meet the IGS standard. B) Total variables that meet the RVM.

Table 3 Total of stations that agree with IGS and RVM recommended values.

Fig. 9. Time series of the quality indicators for MPR1 CORS.

Fig.10. Time series of the quality indicators for INEG station.

The receiver TRIMBLE 5700 was changed atthe beginningof 2011.However,it did not showa significant improvement in the signal-tonoise ratio, even it was found to have instability on the multipath effect and cycle slips in the observations.The next important change was in May 2016, when receiver TRIMBLE 5700 was changed to LEICA GR10, antenna TRM29659.00 NONE to LEIAR10 NONE, and firmware was updated from 1.24 to 3.22/6.521. These changes showed a better response in all quality indicators,which improved the signal-to-noise ratio and had better performance in terms of multipath effect(being lower than 0.25 m),with integrity over 98%and better behavior in cycle slips with less than 1 per 1000 observations.The adequate behavior described in the INEG station since 2010,and improved in May 2016,is not only a function of the system(antenna-receiver-firmware) but also of the environment of the station that has presented a weak multipath effect.

To check the influence of the surrounding changes on the data quality and the daily position, a time series analysis of INEG and MPR1 CORS was carried out using GAMIT/GLOBK [32].

Figs.11 and 12 represent the time series of the best and worst case of the quality checked stations. The time series were created with TSVIEW software. These results would be very useful and applicable to scientific studies such as subsidence and crustal deformation analysis, that could be related to the geological faults and nearby rifts to the stations [46,47].

Specifically,the station INEG showed an anomalous signal in the vertical component, related to evident post-seismic deformations and the Aguascalientes graben in Mexico. There is a well-known subsidence effect in the Up component (Fig. 11A) [36]. Thus, this station was not considered to determinate the primary and secondary spherical harmonics coefficients associated with solid earth changes [48]. The time series revealed a dispersion on the estimated daily position based on the quality of indicators, the mayor displacement was presented in the east component. The adjusted time series(Fig.11B)was affected by the large M7.1 earthquake on September 19, 2017. Thus, the results would be affected by the surrounding environment even if the antenna was provided with suppression plates for the multipath error.

The time series of MPR1 (Fig. 12A) evidence southwestward displacement with the horizontal components being negative, as well as the stability in the up component with a velocity of 0.98±0.04 mm/yr are shown in Fig. 12A and B. On the other hand,Fig.12B suggests that the data processing stage could not reduce the error derived from the multipath effect (shown in Fig. 9)showing a major dispersion in the daily positioning estimations.As a complement to the time series analysis,time-correlated noise at each site using First-Order Gauss-Markov Extrapolation (FOGMex) algorithm was estimated in order to obtain a realistic velocity uncertainty [49-51]. The real sigma estimation for INEG and MPR1 is shown in Figs.11C and 12C,where a time-correlated noise estimated is shown for the velocity uncertainties with temporal correlations in the time series. The chi-squared-perdegree of freedom increases as the residuals are averaged over successively longer times intervals [49]. The results of the times series for the real sigma estimation show a combination of random-walk and white noise (Figs. 13 and 14). For the MPR1 station, it is confirmed that the noise in the east component is presented (Fig. 13B). If the averaging continues to infinity, the dominant frequencies of the noise are long period, it could be expected to reach a flat line,or at least an asymptote,which would be used to characterize the noise at an infinite frequency (the uncertainty of velocity).It is important to state that most IGS sites such as INEG benefits from a well-documented and stable geodetic and geophysical applications. Other sites such as MPR1 were installed for various purposes, including surveying, meteorological, aerial control, or ionospheric studies. Hence, their monumentation quality is not always known and may not adhere to IGS standards for stability, expect for those installed for highprecision geodetic and geophysical applications.

Fig. 11. A) Time series of INEG station. B) Adjusted (detrend) time series of INEG station. C) Time series with time-correlated noise model. The red line corresponds to the estimated uncertainty ± 1 σ. The green lines refer to the mean square error weighted by±3 times the dispersion.Magenta bar represents a large earthquake on 19 September 2017, of Mw7.1.

The MPR1 station is located on the top of the control tower of the International Airport of Puerto Vallarta with more antennas that obstructs the received signal.Considering the high value of the multipath effect and cycle slip,the final estimated solution is highly affected by the surrounding environment. The magnitude and orientation of the stations MPR1 (North -3.08 ± 0.02 mm/yr,Est -7.85 ± 0.02 mm/yr) and INEG (North -0.07 ± 0.03 mm/yr,Est -9.33 ± 0.02 mm/yr) in the Jalisco Block, was found to be consistent with the results shown by previous researchers[31,33,49]. A comparison of the results is shown in Table 4, where the motion is consistent with the previous studies.

Fig. 12. A) Time series of MPR1 station. B) Adjusted (detrend) time series of MPR1 station. C) Time series with time-correlated noise model. The red line corresponds to the estimated uncertainty ± 1 σ. The green lines refer to the mean square error weighted by ± 3 times the dispersion.

Additionally, Fig. 15 shows the velocities estimated from few stations in the study region using GAMIT/GLOBK, which are consistent with the southwestward motion of the North American Plate [52]. Besides, the interaction of the Pacific and North American Plates in a transform fault and a tectonically active rift[53], shows changes in the velocity vectors (direction and magnitude), specifically on the Baja California Peninsula, confirming that a part of Baja California Peninsula is detached from the Pacific Plate given the differences in the velocity estimates[34].The behavior of MPR1 is consistent with the southwestward motion of the North American Plate notwithstanding the surrounding environment and outdated system that show a major dispersion in the daily solutions.The obtained results suggest that CORS in Mexico is suitable for operational use considering data analysis of the duration from 2010 to 2020. Few researches reported the analysis of the data from Topo-Iberia network, however, those did not consider the relationship of the quality indicators and precise positioning throw time series analysis[20,24]. Present analysis improved solution in few CORS such as MTY2 and INEG CORS,where the multipath effect was reduced by around 80%. These improved time series through quality indicators analysis will be most useful in scientific applications such as positioning, crustal deformations, geodynamic process, structural monitoring, tectonics motions that requires highest precision and accuracy.

Fig.13. Real sigma (green line) for MPR1. A) North(NRMS realistic 9.39; correlation time 256 days). B) East (NRMS realistic 4.56; correlation time 32 days). C) Up (NRMS realistic 3.69; correlation time 128 days) components. Red squares denote calculated values.

Fig.14. Real sigma(green line)for INEG.A)North(NRMS realistic 16.25;correlation time 256 days).B)East(NRMS realistic 14.66;correlation time 256 days).C)Up(NRMS realistic 49.96; correlation time 512 days) components. Red squares denote calculated values.

Table 4 Velocities estimated in mm/yr in the present study with similar studies carried out in the region.

Fig.15. Horizontal velocities of GPS stations relative to reference frame (IGS14).

6. Conclusions

The analysis reveals that IGS station INEG showed the best behavior considering the IGS and regional Mexican recommended values.The station showed high values of cycle slips.Similarly,the worst behavior was seen in MPR1 station that was located on the aerial control antenna. Considering the IGS recommended values(calculated with more than half of the total IGS CORS stations around the world) as the best quality indicators for the satellite navigation, MEXI, MTY2, OAX2, and TOL2 stations fulfil the same quality indicators as INEG station with similar values of multipath(principal indicator of quality), strong SNR and integrity superior to 95%.These four stations can be included in the IGS continuous network due to their observed behaviors. It was found that the multipath effect is high in the case of outdated system (antennareceiver-firmware)specifically when Lsignal is degraded(signalto-noise ratio with weak signals).

The recommended values for multipath effect and cycle slips are more tolerant in the RVM values than the IGS standard,nevertheless, they tend to be highly strict for the signal-to-noise ratio and integrity values. On the other hand, considering the IGS recommended values, the global average of variables that were fulfilled was 3.30 of the 6 possible variables in the 65 stations, nevertheless, with the RVM values agreed with 3.54 variables of the same indicators. The difference between the IGS and RVM recommended values, numerically seemed to be small, but more stations agreed with the RVM recommended values, fulfilling 4-5 variables.

The results show inverse relations between SNR and MP in the case of INEG,MEXI,MTY2,OAX2,and TOL2 stations.The results also show a direct relation between MP and SNR in 60 stations,with the highest effect in Lsignal. Despite the careful selection of location and use of mitigating antennas for the multipath effect, it is not possible to mitigate multipath errors completely. Therefore, it is important to select an optimum location to establish a GNSS continuous observations.The study suggests cGPS/CORS in Mexico possess a good behavior towards the recommended values with over 85% of the stations fulfilling more than 3 variables in the quality indicators. Finally, a strong correlation is determined between integrity,cycle slips and daily observation time.It was found that with shorter observation time,integrity reduced with a greater number of cycles slip per every thousand observations leading to imprecise solutions in the positioning, nevertheless, the velocity estimations are consistent with the interaction of the four principal tectonic plates in Mexico.

The results obtained by RVM identify those CGPS with irregular behavior and represent the real situation in the country (the RVM calculation could be used in other regions), additionally, an exhaustive quality check was performed to obtain possible sources of signal degradation and its implication in crustal deformation analysis. The velocity and direction obtained was found to be similar to previous reported research in the study region in Mexico.It is concluded that the analysis of several indicators (multipath,signal-to-noise ratio,integrity,and cycle slips)is utmost important as it is strongly related with the quality of the results that can be obtained.

Conflicts of interest

The authors declare that there is no conflicts of interest.

Acknowledgment

The authors acknowledge the data provided by the GAGE Facility, operated by UNAVCO, Inc., with support from the National Science Foundation and the National Aeronautics and Space Administration under NSF Cooperative Agreement EAR-1724794.The authors are also grateful to universities, consortia, and government agencies that contributed to make geodetic data publicly available that were utilized in the present analysis. The authors thank the anonymous reviewers for their comments and suggestions, which improved the present analysis manifold.