The ionosphere range delay is one of the dominating error sources for the space-borne GPS single-frequency real-time orbit determination. When corrected by ionosphere correction models, the residual error has obvious influence on the accuracy of real-time orbit determination. In this paper, we use single-frequency pseudo-range and carrier phase measurements, calculated with two different ionosphere correction models, and then we analyze their impacts on the results. Our study shows that when the altitude of the low earth orbit (LEO) satellites is 500 km or higher, the position and velocity accuracy (3DRMS) are 0.86 m and 0.9 mm/s, which is close to the accuracy of dual-frequency pseudo-range real-time orbit determination. When using the group and phase ionosphere correction (GRAPHIC), the position and velocity accuracy can be 0.54 m and 0.55 mm/s. Therefore, when an appropriate ionosphere correction model is selected, low-cost single-frequency space-borne GPS receivers can be used in real-time orbit determination for micro-satellites.
This paper analyzes the error characteristics of high-rate single-epoch GPS time series and uses the median filtering method to eliminate the high rate noise. Through calculation we determine that the accidental error in the horizontal direction is ±4.0-4.5 mm and in the elevation direction it is ±9 mm. By periodically repeating solutions, comparing the time series, and maximizing the correlation coefficient, the periodic system error is discovered in the high-rate single-epoch GPS time series. The repetition period is from 86 153 to 86 165 seconds. Through modeling or differencing the system errors, the system components can be eliminated from the time series, thus improving the data processing accuracy of high-rate single-epoch GPS.
The atomic clocks onboard navigation satellites provide the reference frequency for the navigation signals, and the stability of the clock will directly influence the accuracy of timing and positioning performance. In this paper, the IGS MEGX clock products for Beidou satellites are used for analysis,and the Hadamard variance is used to calibrate the clock stability of all orbiting Beidou satellites. The results show that the clocks onboard Beidou satellites are relatively stable over 1 year’s analysis; and the Hadamard variance is about 10-13 on 300 s interval with slight differences between different Beidou satellites. Compared with GPS satellite clocks, the stability performance is between those of GPS ⅡR-M and GPS ⅡF satellite clock performance. These results are beneficial for navigation and positioning users.
In order to improve the accuracy of IGS ultra-rapid predicted products(IGU-P), in response to the data characteristic of ultra-rapid ephemeris composed of both determined and random items, a new prediction algorithm is proposed. This proposed algorithm is based on a polynomial model, which models random items with adding-weight one-rank local-region method of chaos. The results from the calculating examples show that the accuracy of adding-weight one-rank local-region method of chaos is superior to that of the IGU-P.
This paper studies a modified TCAR method and analyzes the noise error that affects the ambiguity resolution of wide-lane(WL) and NL, and points out that the noise standard deviation of the geometry-free and ionosphere-free combinations given by this paper all equal 8.262 8 times the B3 carrierwave length. Finally, the real measured Beidou navigation satellite system(BDS) triple-frequency data is used to verify and analyze the conclusion.
To accomplish spatial rectangular coordinate transformation with big rotation angles, the improved coordinate transformation model is proposed. This model is based on scale parameter estimation using the matrix singular value decomposition (SVD). Three estimation methods, including direction cosine, unit quaternion and Rodrigues are presented for the improved coordinate transformation model. Finally, simulated and practical experiments of coordinate transformation with big angle are tested to compare and analyze the adaptability of the three methods using the total number of singular points, the condition number of normal matrix, the number of iterations and the RMS, etc. Experimental results show that the improved model is better than the extant seven parameters model, that the SVD solution of rotation matrix is superior to unit matrix as an initial value, and that the direction cosine method is preferable to the other two methods.
This paper proposes a new filtering method for LiDAR Data. The proposed approach constructs the connectivity of a grid over the LiDAR point-cloud in order to perform multi-scale data decomposition. This is realized by forming a top-hat scale-space using differential morphological profiles (DMPs) on points’ residuals from the approximated surface. The International Society for Photogrammetry and Remote Sensing (ISPRS) reference dataset is used to test the method. The experimental results show that the proposed method can effectively remove non-ground points, keep the ground points, and is effective at minimizing total error rates.
Using wavelet analysis method to process aeromagnetic data in the Qinghai-Tibet plateau, aeromagnetic anomaly maps with different scales are obtained. Results from the first order wavelet analysis reveals the structure feature of shallow crust. In the sutureⅠ and sutureⅡ regions, there exist several clear arcuate structures with NWW-striking, as shown by alternatively positive and negative magnetic anomaly belts. There exists a negative magnetic anomaly belt with near EW-striking in suture Ⅳ near Golmud. The results from the third order wavelet analysis show the structure feature of deep crust. The main feature of deep structure in this region is represented by an obvious negative magnetic anomaly belt with near NS-striking between 87°N and 93°N. The Nepal M8.1 earthquake and its three strong aftershocks with magnitude 7 all occurred near the southern end of this magnetic anomaly belt, where the EW strike Himalaya tectonic belt intersects with the NS-striking negative magnetic anomaly belt. In general, the aeromagnetic data in Qinghai-Tibet plateau shows that, the main feature of shallow structures is near EW-striking, but the deeper structure is SN-striking. It implies that the intersection of the complex overpass structures is the tectonic background of the Nepal M8.1 earthquake.
For the purpose of detailed study of the Xianshuihe fault activity, the disclination-dislocation model is put forward in this paper. This model combines GPS data in the area of 2001-2004 with article swarm optimization to inter-fault sliding and rotation at different segments of the fault zone.Wethen verify the reliability of fault parameters based on forward calculation. Our results show: 1) The Xianshuihe fault zone, which presents left-lateral strike-slip movement on the whole and extrusion on the local scale, gives priority to sliding characteristics; 2) The Xianshuihe fault zone presents more rotation characteristics in the southeast segment than in the northwest segment.
We processed time-series data,using a sub-pixel cross-correlation analysis method, to obtain the temporal evolution of the surface displacements of the west Kunlun mountain glaciers in the north-western Tibetan plateau. This work co-registers periodically acquired Landsat-8 images using the sub-pixel cross-correlation technique at an accuracy of 0.01 pixels, implying a horizontal displacements detecting precision of the optical image 0.15 m. By cross-correlation analysis of the Landsat-8 images acquired from July 2013 to August 2014 and time-series inversion, this study constructs the displacement and velocity fields of two glaciers in the west Kunlun mountains. It demonstrates that the glaciers in the study area are slipping fast and uniformly (no obvious acceleration or deceleration). Based on Landsat-8 images, the method of this study can potentially be used for measuring ground deformation caused by dunes shifting, earthquakes, landslides, or volcanoes etc.
High-rate GPS receivers are popular in real-time dynamic monitoring, but the multi-path effect is one of the key obstacles that restricts its development. However, a sidereal filter can eliminate this effect and improve positioning accuracy. The sidereal filter requires the preset time of each satellite compared to the day before. In this research, 8 526 days broadcast ephemerides are used to calculate the preset time. The Fourier series of 8 orders trigonometric functions is used to fit the series of the preset time. It shows a simple and effective way to estimate the preset time, instead of a more complex calculation. Using this method, the root mean square values of the residual coordinate sequences of N, E and U directions are reduced, and the ability of extracting deformation information is improved.This method therefore can be applied in engineering dynamic deformation monitoring.
Local mean decomposition (LMD) method is applied in monitoring data. Experimental data show that the fitting effect of the LMD-GM (1, 1) model is better than those of the EMD-GM (1,1) and GM (1,1) models. Therefore, LMD-GM (1,1) model has a higher application value in deformation monitoring.
A new kind of on-line predictor is constructed by combining Kalman filtering with the echo state network. The method of Kalman filtering is applied to the echo state network output weights training, directly on-line updating the network output weights, overcoming the defects in traditional recurrent neural network(RNN) which is needed to collect a large number of samples.The examples demonstrate the effectiveness of the proposed method.
Based on teleseismic SKS waveform data recorded from January 2012 to July 2014 at 23 broad-band seismic network stations of Heilongjiang province, we use Splitlab software to study the shear wave splitting in the region. The delay times between fast and slow wave arrival at each station are determined using both the minimum energy and waveform rotation-correlation method and the minimum feature value method. We calculate the SKS fast wave polarization direction and the fast and slow wave delay time for all observation stations, and acquire the image of upper mantle anisotropy in Heilongjiang region. The results show that the Heilongjiang regional upper mantle seismic anisotropy is more obvious and the fast wave polarization direction and principal tensile stress direction are basically consistent. Furthermore, as obtained by the GPS velocity field in the same direction, the coupling of the crust and upper mantle in the region present vertical coherence.
Using P/S amplitude ratio and P-wave first motion data obtained by the digital telemetered seismic network of induced earthquake monitoring system in the Three Gorges area, we analyze the focal mechanisms of earthquakes occurring in the key monitoring area of the Three Gorges reservoir. The main features are near vertical P axes and pure normal faulting. Water loading and its fluctuation in the reservoir, together with the chemical effects of water, might cause the unstable planes to lose balance, resulting in small earthquakes. This is supported by the focal mechanisms of the induced earthquakes that show oblique P axes, which are different from the regional stress fields and focal mechanisms of the earthquakes that occurred before the first impoundment.
The risk of landslides, which have been treated in phase Ⅱ and Ⅲ of the Three Gorges project, are increasing because of the cyclical loading and unloading of the Three Gorges reservoir. The reactivated landslides and new landslides would threaten people’s lives and the safety of property. In order to mitigate the risk, remote sensing (RS) and geographic information systems (GIS) technology are used to study the correlation between the distribution of landslides and other factors, such as topography, lithology, faults, water basins and anthropogenic influences. The results show that the elevation, aspect and elevation variation coefficient in topography have higher correlation with the landslides. Moreover, faults have no obvious effect on the quantity of landslides. The spatial distribution of the landslides in the study area is restricted by the combinations of factors. According to the weights of these restrictions, a risk assessment of landslides is carried out.
In this paper, we use the wave equation finite-difference method for forward modeling of the active fault model by setting different observation system parameters. Combining this with the application of practical urban active fault detection data, we obtain acquisition parameters that are suitable for shallow urban active fault detection. The results show that using the observation methods of small group intervals, mini-shot intervals and appropriate seismic frequency gets high quality shallow seismic data, which is conducive to improving the positioning accuracy of the breakpoint.
In this paper, approximately 11 years of monthly GRACE measurements (from 2003-01 to 2013-12) are used to estimate the real glacier mass variations in Antarctica, after correction of the leakage errors, using global forward modeling. Compared with direct GRACE estimations, our results, after correcting the leakage effect, show the 20.3% of the signals in whole Antarctic glacier mass changes are attenuated in GRACE inversion procedure. 26.4% signals of ice loss in west Antarctica are leaked into the surrounding region, while the leakage errors are up to 70% in east Antarctica. Therefore, the leakage errors should be removed in quantifying the mass loss of Antarctic glacier. The recovered Antarctic glacier mass variations show that most ice losses are in west Antarctica and the Antarctic peninsula, and glacier mass accumulation occurs at Ronne ice shelf and Amery Ronne ice shelf in east Antarctica. The mass loss is about 152.47±2.00 Gt/a in west Antarctica, which is nearly equal to the total mass loss rate of Antarctica. The mass change rate of Antarctic peninsula is about -27.44±0.75 Gt/a, equaling the gaining mass with the rate of 27.27±5.12 Gt/a in east Antarctica, mostly in coastal regions. The total ice mass is being lost at a rate of -152.64±700 Gt/a, with acceleration of -18.85±4.87 Gt/a2,which will cause the sea level to rise atthe rate of 0.41 mm/a.
Based on GPS coordinate time series from this decade, the velocity field of the Sichuan-Yunnan region is obtained. According to geological data, we divide this region into 7 blocks, then find the three-dimensional finite element model by ABAQUS. Through extracting the boundary value of the velocity field as model constraint conditions, we obtain numerical simulation results of displacement and stress field. The results show that our model clearly reflects the situation of pushing forces from Qinghai-Tibet plateau in the northwest and prevention from Yangzi block in the east. The displacement direction rotates clockwise on eastern Himalayan syntaxis inside rhombic block and Lhasa block. These results conform with reality. Simulation results of stress field indicate that the stress distribution is obviously uneven in some areas, such as the border region of Longmanshan, Xianshuihe and Anninghe fault zone, the border region of Xiaojinhe-Lijiang and Lancangjiang fault zone, the south and north flanks of Nantinghe fault zone, as well as the east and west flanks of Red River fault zone. Seismic risk in these areas is larger than other regions.
In order to calculate motion acceleration of the airborne gravity measurement, double difference GPS (DDGPS) carrier phase pseudo-range observations are analyzed considering ionospheric and tropospheric delay effects. Then, through the kinematic modeling of the carrier, a new Kalman filtering algorithm based on PVA model is proposed. The algorithm can directly calculate kinematic acceleration from the phase observations, relative to traditional difference method. Validated by the simulation analysis, when adopting the adaptive Kalman filter, the standard deviation of kinematic acceleration without low-pass filtering is 7.26×10-3 m/s2.
In view of the Gutenberg-Bullen average earth model and the integral Green’s function, load effects of ocean tide models on displacement at the GNSS continuous operation reference station along China’s coastline are calculated with the two China regional ocean tide models and global ocean tide model EOT11a. Furthermore, the root mean square (RMS) and root sum square (RSS) are comprehensive analyzed. The results show that most of the differences of horizontal displacement loading for the two models are in sub-millimeter, with millimeter magnitude for almost all the vertical displacement loading with a maximum of 5.8 mm. Meanwhile, the coverage area of Chinasea 2010 in the China Sea is larger than Naoregional 1999. There is a great difference between the two models in the Yellow Sea and the East China Sea, but in the Bohai Sea and the South China Sea the difference is small. As the differences among models are relevant to the station location and tidal wave frequency, we should compare the load correction of data and adopt the preferable model suitable for the particular studied areas.
