GPS Campaigns Research Articles

Insights into present-day crustal motion in the central Mediterranean area from GPS surveys

In this paper we present observations of crustal motion related to a large GPS network located in the central‐western Mediterranean area. Velocities are obtained by the analysis of more than 30 observing sites at which at least three different GPS campaigns were carried out in the time span 1991–1999. The results are presented both in the ITRF96 reference frame and with respect to a Eurasian fixed reference frame. The sites located along the northern African margin, in Sicily and southern Italy show prevalent northeastward movements with a mean velocity of roughly 0.7 cm yr−1. The relatively dense network available in the southern Apennines led us to try a tentative estimate of the average strain rate in this zone, resulting in a maximum extension of 0.021 ± 0.006 × 10−6 yr−1 normal to the chain. In Ionian Greece the two sites located south of the Kephallinia discontinuity (Lefkas and Kastro Ilias) consistently indicate a south to southwestward motion at an average rate of roughly 15 mm yr−1.

Using repeated precise gps campaigns for research of present-day crustal movement and strain in the Qinghai-Tibetan Plateau

The problem of crustal movement in the Qing-hai-Tibetan Plateau is a hotstopic of the present-day international geodetic world. GPS observations in 1993, 1995 and 1997 are used to survey the three-dimensional displacement of GPS monitoring station in the Qinghai-Tibetan Plateau, and to calculate and analyze the present-day crustal movement and strain in the Qinghai-Tibetan Plateau. The result of repeated campaign shows that the crustal movement in the Qinghai-Tibetan Plateau is still compressing in the north and south directions, extending in the west and east directions, and uplifting in the vertical direction. The relative convergence velocity of the Himalayan block is approximately (19.5±1.7)mm/a, the strain velocity is about (5.5±6.0) mm/a, and the uplifting velocity is some (7.6±5.2) mm/a. The compression velocity of the Tibet block relative to Golmud in the north and south directions is (9.3±4.6) mm/a, the strain velocity of the middle part in the west and east directions is (15.6±6.3) mm/a, reflecting the plateau lateral extruding movement. The Himalayan block is mainly compressing strain, and the middle part of the Tibet block is mainly extending strain. The highest extending and the compressing strain are 0.131×10-6/a and -0.189×10-6/a, respectively.

ITRF coordinates and plate velocities from repeated GPS campaigns in Antarctica - an analysis based on different individual solutions

Scientific Committee on Antarctic Research (SCAR) The GPS campaigns of the between 1995 and 1998 provide a valuable data set which is used to link Antarctica with the ITRF (International Terrestrial Reference Frame), and to gain the first detailed insights into the tectonic behaviour of the Antarctic plate. The data analysis is based on seven individual solutions with four different software packages (Bernese, GAMIT/GLOBK, GIPSY/OASIS II, GEONAP), and provides the possibility to study the existing software noise. The combined solution yields an accuracy of 1 cm for the horizontal and 2 cm for the vertical coordinate components within ITRF96. Based on repeated observations during a total time span of three years, it is possible to determine detailed regional deformations in the area of the Antarctic Peninsula and to obtain an estimate for the rotation vector of the whole Antarctic plate. The main findings of a very comprehensive project of, in total, seven working groups are summarized and presented for the first time. The main emphasis is on the geodetic aspects of the project, i.e. the analysis of the same data set with different software packages, and on the final results from repeated observations, namely station coordinates and horizontal velocities in ITRF96.

EFFECTS OF ELECTRIC POWER LINES ON THE ACCURACY OF GPS POSITIONING

A practical experiment of the potential effects of high voltage electric power lines on GPS satellite signals and hence the received GPS data was performed. In this experiment, a network of 10 stations were precisely established underneath and around 380 kV electric power lines. Then, a static GPS campaign was conducted on the network while the power was shut off. The same campaign was repeated in the next day when the power was turned on. The results of the two days were analysed and compared with each other. The results of this experiment demonstrated that there is no statistical evidence of any degradation of GPS positioning accuracy due to the high voltage electric power lines.

First-Order Leveling Reveals Anomalous, Interseismic, Contractile, Transient Strain Across Teton Normal Fault, 1988-2001, Grand Teton National Park, Wyoming

As part of a comprehensive neotectonic study of interseismic behavior of active faults, we have done six first order leveling surveys of 50 permanent bench marks in a 22 km-long base line across the Teton fault to characterize its interseisrnic behavior between 1988 and 2001. This 55 km-long normal fault extends along the eastern base of the Teton Range, exhibits up to 30 m of post-glacial offset, and has one the highest rates of Holocene slip of any fault in the Basin-Range. It is seismically dormant at the M2+ level, however, and presently lies in the center of a 50 km-long seismic gap. Results of five of the six levelings are remarkably similar and suggest that the alluvium-filled valley of northern Jackson Hole (hanging wall) subsided 6-8 mm relative to bedrock of the Teton Range (footwall) relative to the 1989 survey. Height changes were insignificant from 1989 to 1993. In 1997, however, a 2 km-wide zone adjacent to the fault rose 12 mm relative to the 1993 survey, and then dropped 16 mm by the 2001 leveling. This zone coincides with an area of low topography characterized by lakes ponded along the fault and south-flowing streams parallel to the range front, rather than eastward away from the range. This subsidence zone records hanging wall subsidence related to long term faulting. The 1997 uplift of the valley floor and subsidence zone may reflect an unexpected, reverse loading and local crustal shortening between 1993 and 2001. Campaign GPS surveys (1987 to 2000, only briefly summarized in this report) support this hypothesis, indicating that the principal direction of horizontal shortening was locally E-W perpendicular to the fault, and that crustal shortening occurred in the period 1995-2000. Regionally during 1987-1995, subsidence and shortening characterized deformation of the Yellowstone caldera when GPS recorded uplift and extension across the Teton fault, only 30 km to the south. During 1995-2000, subsidence slowed or ceased for much of the caldera, whereas the overall GPS vectors across Jackson Hole were directed west with almost 2 mm/yr of E-W motion (N. America fixed). This shortening strain field implies that the regional stress field was compressional normal to the fault at the time of the 2000 GPS survey. The return of the 2001 leveling signal to pre-1997 values suggests that the strain reversed or relaxed, and that the 1997 leveling anomaly was a contractile strain transient that passed across the fault probably between 1995, when the strain pattern at Yellowstone caldera changed, and the 2000 GPS survey but before the 2001 leveling. Preliminary elastic dislocation models indicate 10-20 mm reverse slip at a depth of 1-2 km. Alternatively the observed leveling changes may reflect a complex combination of other processes including local poroelastic effects, or nearfield drag of the hanging wall as it subsides overall in farfield extension.

Local GPS tropospheric tomography

A formulation of local GPS tropospheric tomography for determining 4-D distribution of refractivity in the troposphere is presented together with a preliminary analysis of local dense GPS campaign data. Dividing the modeling space up to a height of 10 km above GPS receivers into cells, the refractivity in each cell is estimated in a successive time window by a tomographic reconstruction method in a quite similar manner like the seismic velocity in each cell in Earth’s interior is estimated in seismic tomography. The basic data for tomography are GPS slant delays for respective pairs of station and satellite, which are the sum of postfit phase residual, hydrostatic and wet slant delay. On the other hand, the slant delay from a station to a satellite is expressed by the summation of the product of path length and refractivity in each cell along the ray path. In a given time window, we have numerous observed slant delays corresponding to different ray trajectories, and the refractivity in each cell can be estimated by discrete inversion and least squares methods. The observational equations are usually singular so that we use a damped least squares method popular in seismic tomography. An example of real data analysis is given for the 1995 Shigaraki GPS campaign data, which reveals the spatial and temporal change of refractivity corresponding to the passage of ‘cold front’.

GPS measurements of present-day uplift in the Southern Tibet

Wuhan Technical University Surveying and Mapping (WTUSM) carried out three GPS campaigns in the Tibetan Plateau since 1993. To improve the vertical component precision of GPS measurement, we estimated the remaining tropospheric refraction using the piecewise linear method after the SAASTAMOINEN model and corrected the tropospheric refraction by selecting one parameter for every two hours model to consider the variationof tropospheric refraction. Using the IGS combined ephemeris and corresponding EOP Series for 1995 and 1997 GPS campaigns, all baseline vectors were processed by GAMIT (Ver. 9.71) software, and network adjustment was carried out in ITRF97 frame by PowerAdj software (Ver. 3.3) which was developed by WTUSM considering the baseline vectors’ full covariance matrix. The resulting vertical velocity field determined from GPS campaigns in 1995 and 1997 shows that the average uplift rate in ITRF 97 frame is about 7.8 ± 7.6 mm/yr for all 14 repeated surveying stations in Tibet, and the average uplift rate is 16.2 ± 7.9 mm/yr in southern Tibet, which are compared with the results from the gravity measurements, leveling data and fission-track dating (FT).

Active tectonics of the Black Sea with GPS

The Black Sea occurs within the Anatolian sector of the Alpine-Himalayan orogenic system. In this region northward moving African and Arabian plates collide with Eurasian plate. From this collision the Anatolian block moves westward with a rotation pole located approximately north of Sinai peninsula. Tectonic styles and rates in the circum Black Sea and along the Crimea have been poorly known. A GPS project was initiated in 1995 with the collaboration between Istanbul Technical University, MIT and Joint Institute of Physics of the Earth (Moscow). To carry out the project, two GPS campaigns were performed. Data have been analysed by the well-known GPS processing program, GAMIT which is a comprehensive GPS analysis package, and GLOBK, a Kalman filter. This paper presents the important outputs from the two GPS campaigns and gives suggestions for the future investigations in the region. The important outcome of the project is that the N-S motions along mostly in the Eastern margin of the Black Sea are in a few mm/year, such as SINO, a southern coast site, with a rate of 1.4 ± 1.7 mm/year in NW direction, and GELE, a northern coast site, with a rate of 2.2 ± 2.8 mm/year in SE direction, while the velocities in the Anatolian region are approximately 10–20 ± 3–5 mm/year.

Glacial rebound near Vatnajökull, Iceland, studied by GPS campaigns in 1992 and 1996

Since about 1920 the Vatnajökull ice cap in Iceland has experienced a significant retreat, corresponding to a volume reduction of more than 180 km 3. With two GPS campaigns in 1992 and 1996 along the southern border of the glacier preliminary results reveal land uplift rates of 1–6 mm/yr, after a one-parameter (bias) fit with recent earth rheology models. The best fit model suggests that the lithosphere in the area is about 30 km thick and the viscosity of the asthenosphere 5 × 10 18 Pa s. The rms fit of uplift rate at all GPS sites is ±1.4 mm/yr. As the GPS data alone cannot provide the absolute uplift rates, the one-parameter fit to the theoretical modelling implies that the absolute rates were estimated by the matching of the GPS data and model. The resulting uplift rate at station Höfn (1 mm/yr) is not consistent with two independent sources, and we therefore conclude that further GPS epoch and permanent GPS site data are needed to confirm the present geodynamic processes near Vatnajökull.

Estimating present‐day postglacial rebound and horizontal movements in Fennoscandia By Repeated GPS Campaigns in 1993 and 1997

A well‐known geophysical phenomenon, the postglacial rebound in Fennoscandia, is investigated by data from the Baltic Sea Level Project GPS Campaigns in 1993 and 1997. Preliminary results indicate that the estimated uplift rates vary from 1 mm yr−1 to 11 mm yr−1 with standard errors of the order of 1–3 mm yr−1, which is partly consistent with traditional terrestrial data. In addition, there is a horizontal component of plate motion of 21±1 mm yr−1 towards northeast. After subtracting the average velocity, differential horizontal velocities suggest that there is a right‐hand strike slip fault with a relative velocity of about 1.5±0.5 mm yr−1 along a N‐S direction in the middle of the Baltic Sea, which agrees with some geologists' belief of N‐S trending 1.9–1.8 Ga old strike‐slip megashears in the Baltic shield.

Le système de positionnement GPS en zone de montagne : effet de la troposphère sur la précision GPS verticale

Precise estimation of the vertical deformation is a challenge for understanding geodynamical processes. In mountainous areas, the main limitation of precise GPS vertical measurement lies in the difficulty of estimating tropospheric delay due to high differential elevation and large weather variability. We carried out two GPS campaigns in July 1994 (8 days) and March 1996 (11 days), during extreme weather conditions, on a 1 390 m height difference baseline. The zenith delay was calibrated with a standard atmospheric model (STD), Water Vapor Radiometric measurements (WVR) and surface pressure measurements, or computed through a least squares estimation (LSE). Half daily repeatability of the vertical LSE solutions has a precision of 11–14 mm with a small difference of 1 mm between the two expeditions.

Unification of vertical datums by GPS and gravimetric geoid models with application to Fennoscandia

The second Baltic Sea Level (BSL) GPS campaign was run for one week in June 1993. Data from 35 tide gauge sites and five fiducial stations were analysed, for three fiducial stations (Onsala, Metsahovi and Wettzell) fixed at the ITRF93 system. On a time-scale of 5 days, precision was several parts in 109 for the horizontal and vertical components. Accuracies were about 1 cm in comparison with the International GPS Geodynamical Service (IGS) coordinates in three directions. To connect the Swedish and the Finnish height systems, our numerical application utilises three approaches: a rigorous approach, a bias fit and a three-parameter fit. The results between the Swedish RH70 and the Finnish N 60 systems are estimated to −19.3 ± 6.5, −17 ± 6 and −15 ± 6 cm, respectively, by the three approaches. The results of the three indirect methods are in an agreement with those of a direct approach from levelling and gravity measurements.

Quality aspects of a GPS reference network in Antarctica - a simulation study

GPS has become an essential tool for the precise determination of point positions. Since GPS campaigns for geodynamic purposes – such as the monitoring of recent crustal movements – require major financial efforts, it is essential to ensure already in the planning phase a good network quality and the attainment of the scientific goals in a reasonable time. The paper outlines an operationally oriented procedure for these purposes based on the simulation and processing of GPS carrier-phase observations. Sensitivity analysis techniques are applied to describe both the network strain as it may be induced by gross errors and the detectability of point movements between two sessions or two campaigns. In addition, the eigenvalue decomposition of the variance-covariance matrix of the GPS coordinates will be used to identify weakly determined network components. The design of the SCAR Epoch 95 Campaign GPS network is discussed throughout the paper. It was realized in the Antarctic summer 1994/1995 on the Antarctic peninsula.

Evidence of active crustal deformation of the Colli Albani volcanic area (central Italy) by GPS surveys

Colli Albani is a volcanic complex located in central Italy, very close to the city of Rome. Its last eruption is dated at 0.03-0.02 Ma. Now it displays a recurrent seismicity, sporadic gas emissions from soils, wells and springs, and post volcanic hydrothermal circulation. Moreover an uplift of about 30 cm over the last 43 years was recently detected by comparing the height differences between some vertices of repeated leveling surveys, thus suggesting a recent active magmatic process beneath the volcano. In 1995 a first-epoch GPS campaign was carried out with last generation receivers, the repetition campaign was performed in 1996. Data coming from the two repeated surveys were carefully processed and analyzed. A statistical analysis rigorously applied between the 1995 and 1996 adjusted coordinates shows significant coordinate differences. Some sites of the network, in particular, the sites of Vivaro (VVR), Cava di Ciampino (CVA) and Istituto Nazionale di Geofisica (INGR) display a significant subsidence of 2.6 ± 0.7, 2.7 ± 0.8 and 2.0 ± 0.8 cm, respectively, with a confidence interval at 95 percent level. This subsidence seems to be induced by a water table level lowering in Vivaro, while in the area where the sites of Cava di Ciampino and INGR are located, this motion is also confirmed by historical levelling data that show a subsidence of ~ 2 cm/yr during the last 45 years and could be due to local tectonics. Moreover, significant planimetric deformations are exhibited by the site of Castel Romano (CSR). It is remarkable that the stations of CVA, INGR and CSR are located in the area that was struck by the June 12, 1995, M d = 3.8 earthquake, although the displacement vectors are quite small. The motion of the Capranica Prenestina (CPR) station, which shows an uplift of 2.7 ± 0.7 cm, must be considered separately, being located outside the volcanic structure and placed over the carbonatic outcrops of the Prenestini Mts. Furthermore, the accuracy obtained in height determinations does not allow us to confirm if the central part of the GPS network corresponding to the area investigated by Amato and Chiarabba is uplifting within the analyzed time span.

Repeated GPS surveys across the Ionian Sea: evidence of crustal deformations

SUMMARY Geodetic evidence of crustal deformation in the Ionian area detected by GPS surveys is given in this paper. A network consisting of nine geodetic sites crossing the Ionian sea from Calabria (southern Italy) to northwestern Greece was repeatedly surveyed, starting in 1991, within the framework of the TYRGEONET project. The results, obtained from processing data from three GPS campaigns performed on the same network in 1991, 1994 and recently in 1995, show significant changes in the positions of five sites. The deformations in terms of displacements (coordinate differences), are obtained by applying a rigorous statistical approach that analyses their significance by an original iterative procedure based on the classical F (Fisher) test. Although more substantial geophysical conclusions require more observations, the estimated relative displacement pattern is generally coherent with the features of the main tectonic structures identified for this area. The detected deformations for the Greek sites confirm the activity of the Kephalonia right-lateral transform fault, and a rightlateral motion of the Mattinata fault (northern Apulia) seems to have been detected by the Italian sites. Moreover, within the time span analysed, the site of Matera shows a different behaviour from the three Adriatic sites (Tremiti, M.S. Angelo and Specchia Cristi), since relative displacements among Matera and these sites were detected. This fact may indicate the weakness of the assumption, reported in some papers, that Matera could be a representative site of the motion of the whole Adriatic plate. Furthermore, the site of Specchia Cristi shows the maximum relative displacement in the network, with a vector magnitude of about 5 cm with a 3 cm confidence interval at the 95 per cent level. Therefore, even if some additional GPS observations are needed to achieve a clearer picture of the tectonic behaviour of the Ionian Sea area, the detected relative displacement pattern of the Italian sites agrees with a possible rotational behaviour of the Apulian platform with respect to the Adriatic plate, as already hypothesized by Finetti (1982).

Viscosity structure beneath northeast Iceland

The dynamics of crustal rifting in Iceland depend strongly on the lower crustal rheology, which controls the intensity of upper crustal stress concentration and scale time of heat diffusion from the underlying mantle plume. While magnetotelluric surveys suggest the presence of a pervasive hot and highly ductile lowermost crust with possibly high fraction of partial melt, observations of low seismic attenuation and strong shear wave transmission suggest a much cooler lower crust and upper mantle. Since viscosity is also sensitive to the degree of partial melt present, viscosity estimates for these regions could shed light on the factors responsible for these observations. In this study we utilize horizontal and vertical displacement vectors determined in GPS campaigns in northeast Iceland since 1986. These are modeled in terms of steady state tectonic loading plus postseismic/postdiking relaxation following the 1975–1984 Krafla rifting episode, as first proposed by Foulger and others. With the elastic part of the model fixed by external constraints, these data have a high sensitivity to the viscosity structure beneath Iceland. Lower crust and upper mantle viscosities of about 3 × 1019 Pa s and 3 × 1018 Pa s, respectively, yield the closest agreement with the data. Our lower crustal viscosity estimate is consistent with the low attenuation and low (subsolidus) temperature for the lower crust inferred in recent studies. Inversions for fissure opening during the Krafla rifting episode yield about 7 m of opening centered on the Krafla rift, as is observed. Allowing for contemporaneous deep rifting on vertical faults along the Askja segment partially accounts for the observed increase in separation across the rift during 1987–1992 but does not account for large displacements in the southeastern part of the network or the large relative subsidence around the Askja rift during 1987–1990. Recent deep normal faulting beneath the Askja rift and further south might explain all of these remaining features.

The CASA'93 GPS campaign for crustal deformation research along the South Caribbean plate boundary

The results of the 1993 GPS campaign in the Central and South America (CASA) project are presented. 21 stations have been observed along the Boconó-El Pilar fault system in Venezuela. By simultaneously occupying a fiducial site at Richmond/Florida the regional Venezuelan network is related to the International Terrestrial Reference Frame (ITRF). Several alternatives in the evaluation procedure are discussed. The accuracy of the final solution is about ± 5 mm in latitudinal and in longitudinal position components, and ± 10 mm in vertical direction. A comparison with the 1988 survey of five identical stations shows coordinate shifts of stations on both sides of the Boconó fault in rough accordance with predictions of the Caribbean and South American plate motions.

Baltic Sea Level project with GPS

In order to study the Baltic Sea Level change and to unify national height systems a two week GPS campaign was performed in the region in Autumn 1990. Parties from Denmark, Finland, Germany, Poland and Sweden carried out GPS measurements at 26 tide gauges along the Baltic sea and 8 VLBI and SLR fiducial stations with baseline lengths ranging from 230 km to 1600 km. The observations were processed in the network mode with the Bernese version 3.3 software using orbit improvement techniques. To get rid of the scale error introduced by the ionospheric refraction from single-frequency data, the local models of the ionosphere were estimated using L4 observations. The tropospheric zenith corrections were also considered. The preliminary results show average root mean square (RMS) errors of about ±3 cm in the horizontal position and ±7 cm in the vertical position relative to the Potsdam SLR station in ITRF89 system. After transformation of the GPS results to geoid heights using the levelled heights, an absolute comparison with gravimetric geoid heights using the least squares modification of Stokes' formula (LSMS), the modified Molodensky and the NKG Scandinavian geoid 1989 (NGK-89) models gives a standard deviation of the difference of ±7cm to ±9cm for the NKG-89 model and of ±9cm to ±30cm for the LSMS and the modified Molodensky model. The Swedish height system is found to be about 8-37cm higher than those of the other Baltic countries for NKG-89 model.

Realization of a Regional Terrestrial Reference Frame: A GPS Campaign

A terrestrial reference frame can be defined by the coordinates of a set of stations on the surface of the earth. It is possible to realize a reference frame through the GPS positioning. In this sense two basic approaches are available to realizing a terrestrial reference frame using GPS: absolute and relative. By the absolute approach we mean the point positioning, whereas by the relative approach is meant the relative positioning. At the present day the absolute approach should realize a reference frame at a meter or submeter level precision. By contrast, the relative approach could realize a reference frame with a centimeter or even better precision.

Comparison of baseline results for the TI-4100 and Trimble 4000SDT geodetic GPS receivers

Many GPS networks which were initially surveyed with Texas Instruments TI-4100 receivers have now been resurveyed with mixtures of TI-4100 and Trimble 4000 receivers or exclusively with Trimble receivers. In order to make confident tectonic interpretation of displacements observed between such surveys, it is necessary to understand any biases which may be introduced by using different receiver types or by mixing receivers within a network. Therefore, one of the primary objectives of the Ecuador 1990 GPS campaign (February 1990) was to provide a direct long baseline comparison between the TI-4100 and Trimble 4000SDT GPS receivers. p ]During this campaign, TI and Trimble receivers were co-located at each end of a 1323 kilometer baseline (Jerusalen to Baltra). Solutions for this baseline show no variation with receiver type. Zero-length baseline solutions showed no evidence for any intrinsic bias caused by mixing the two receiver types. Short baseline solutions indicate a bias of -34±10 mm in the baseline vertical component; the sign of the bias indicates that either the assumed phase center location for the TI is too low or the assumed location for the Trimble is too high. The bias is explainable if the phase centers of the Trimble SDT and SST antennas are similarly located. p ]Solutions for baselines measured with codeless receivers (such as the Trimble) should be as precise as those for baselines measured with P-code receivers (such as the TI) as long as it is possible to resolve ambiguities. Resolution of the widelane ambiguity is the limiting factor in ambiguity resolution with any codeless receiver, and in the February 1990 campaigns it was not successful fore baselines longer than 100 km. Without explicit modeling of the ionospheric effect on the widelane, ambiguity resolution with codeless receivers will not be successful for baselines longer than about 100 km, depending on the local ionospheric conditions.