Site-specific multipath characteristics of global IGS and CORS GPS sites

The site-specific multipath characteristics of 217 Global Positioning System (GPS) sites worldwide were analyzed using the variability of the post-fit phase residuals. Because the GPS satellite constellation returns to the same configuration in a sidereal day (23 h 56 min 4 s), the multipath repeats on that period. However, daily GPS position estimates are usually based on the solar day. When several days of GPS data are processed, this steady change in the orientation of the satellite constellation with respect to the station manifests itself in the form of patterns in the post-fit phase residuals which shift by 3 min 56 s per day. It was found that the mean root mean square of the time-shifted post-fit phase residuals is highly dependent on the GPS antenna type. The conclusions derived from the analysis of the time-shifted post-fit residuals were verified by performing a cross-correlation of the post-fit residuals across many days for selected sites.     

Duration Magnitude Scale and Site Residuals for Northern Morocco

— The first empirical duration magnitude (M_D) formula is developed and tested for the Northern Morocco Seismic Network (NMSNET). This relationship is obtained by relating the IGN (Instituto Geografico National, Madrid) body-waves mbLg^IGN to the duration (), and the epicentral distance (), at 25 analogue stations of the NMSNET for 479 earthquakes with 2.5 mb 5.4, from March 1992 to February 2001. M_D estimates are significantly more precise while introducing a correction term for each of these stations, cSta_j. The magnitude for the i^th event (M_D)_i is the mean value of individual M_Dij=–0.14+1.63log₁₀(_ij)+0.031(_ij)+cSta_j. The cSta_j corrections reduce considerably the local site effects which influence the recorded durations and cause stations to either overestimate, or underestimate M_D up to 0.5 magnitude units. Average station M_D residuals (–cSta_j) are found to be independent of the distance from the epicenter to at least 10 degrees. It seems evident that regional geological features in the immediate behavior of stations have a systematic effect on the corresponding obtained residuals: older well-consolidated Precambrian crystalline rocks produce high negative residuals (shorter durations), younger unconsolidated sediments produce high positive residuals (longer durations), whereas, intermediate M_D site residuals appear to be the result of the effect of various factors, principally age and state of consolidation of the bedrock, combined with the local tectonic.     

A procedure for testing the assumption of homoscedasticity in least squares residuals: a case study of GPS carrier-phase observations

Due to increased demands on the quality of the results of Global Positioning System (GPS) evaluations, various authors have studied improvements of the stochastic model of GPS carrier-phase observations. These improvements are based on the reasonable assumption that the commonly used stochastic model with independent and homoscedastic (i.e. equal variance) errors is unrealistic. However, this has not been proved rigorously so far. A statistical test procedure based on uncorrelated least–squares residuals, which allows verification of the hypothesis of a heterogeneous variance, is provided. The statistical test procedure is of interest in its own right, and is independent of the practical problem considered. The presented technique is applied to GPS carrier-phase observations. Results show that the variances of the investigated observations are far from homogeneous. It is indicated that the error variances of the presented data increase with decreasing GPS satellite elevation. These results confirm the assumption that the commonly used stochastic model of GPS observations is inadequate and has to be improved.     

GPS快速精密定位的随机模型研究

综合与归纳了GPS精密定位的三种比较实用的随机模型,提出了将这三种随机模型用于GPS快速精密定位中,通过两个实例分析了三种随机模型对GPS快速精密定位的精度影响,得出几点有益的结论。     

A Procedure for Estimating the Variance Function of Linear Models and for Checking the Appropriateness of Estimated Variances: A Case Study of GPS Carrier-phase Observations

In many applications of linear model theory, homogeneous variances are assumed. In practice, however, the variances are frequently heterogeneous. Therefore, to improve the results, the unknown variances have to be estimated. The appropriateness of the estimated variances has then to be checked by a suitable statistical test procedure. Such a procedure is also useful to study models of global positioning system (GPS) carrier-phase observations. While the functional model of GPS carrier-phase observations is widely accepted, the stochastic model is still under development. As well as the neglected correlations of GPS observations, a homogenous variance function is frequently assumed. In Bischoff et al. (J Geod 78:397–404, 2005), we showed by statistical testing that the assumption of constant variances is not appropriate. In this paper, we give a procedure to estimate an individual variance function for a pair of satellites and a procedure to check the appropriateness of the estimated variances. As an example, the approach is applied to double-differenced carrier-phase GPS observations.     

New seismological constraints on differential rotation of the inner core from Novaya Zemlya events recorded at DRV, Antarctica

Novaya Zemlya nuclear test records at the seismic station DRV, Antarctica, are analysed in order to obtain further constraints on a possible differential rotation of the inner core with respect to the mantle. These data allow the sampling of the inner core along a nearly polar path in very stable conditions over more than two decades, from 1966 to 1990. The PKP (BC)– PKP (DF) traveltime residuals, which reflect the inner-core anisotropy and/or heterogeneities sampled along the path, exhibit a great stability through time. A computation of the residuals that are expected for various differential rotation rates and the same rotation axis as the mantle has been performed using the worldwide residual catalogue of Engdahl et al . (1997) for summary rays that include the time as an additional parameter in data stacking. Comparison of data and predictions shows that an eastward differential rotation with a rate as large as 3°  yr⁻¹, as suggested by some authors, is not possible, but an eastward rotation at 1°  yr⁻¹ or lower cannot be rejected.     

选权迭代定位粗差时权函数参数之功能

本文研究了幂函数型权函数的参数与定位粗差能力的关系。提出了权函数基本变量的幂取值范围为-2.5～-4.0;计算权的基本变量对于第1,2次迭代用“放大余差”,以后则用余差;对余差进行统计判断的统计量采用“标准化余差”。     

High velocity anomaly beneath the Deccan volcanic province: Evidence from seismic tomography

Analysis of teleseismicP-wave residuals observed at 15 seismograph stations operated in the Deccan volcanic province (DVP) in west central India points to the existence of a large, deep anomalous region in the upper mantle where the velocity is a few per cent higher than in the surrounding region. The seismic stations were operated in three deployments together with a reference station on precambrian granite at Hyderabad and another common station at Poona. The first group of stations lay along a west-northwesterly profile from Hyderabad through Poona to Bhatsa. The second group roughly formed an L-shaped profile from Poona to Hyderabad through Dharwar and Hospet. The third group of stations lay along a northwesterly profile from Hyderabad to Dhule through Aurangabad and Latur. Relative residuals computed with respect to Hyderabad at all the stations showed two basic features: a large almost linear variation from approximately +1s for teleseisms from the north to—1s for those from the southeast at the western stations, and persistance of the pattern with diminishing magnitudes towards the east. Preliminary ray-plotting and three-dimensional inversion of theP-wave residual data delineate the presence of a 600 km long approximately N−S trending anomalous region of high velocity (1–4% contrast) from a depth of about 100 km in the upper mantle encompassing almost the whole width of the DVP. Inversion ofP-wave relative residuals reveal the existence of two prominent features beneath the DVP. The first is a thick high velocity zone (1–4% faster) extending from a depth of about 100 km directly beneath most of the DVP. The second feature is a prominent low velocity region which coincides with the westernmost part of the DVP. A possible explanation for the observed coherent high velocity anomaly is that it forms the root of the lithosphere which coherently translates with the continents during plate motions, an architecture characteristic of precambrian shields. The low velocity zone appears to be related to the rift systems (anomaly 28, 65 Ma) which provided the channel for the outpouring of Deccan basalts at the close of the Cretaceous period.     

Core mantle boundary topography from short period PcP, PKP, and PKKP data

We use a total of 839,369 PcP, PKPab, PKPbc, PKPdf, PKKPab, and PKKPbc residual travel times from [Bull. Seism. Soc. Am. 88 (1998) 722] grouped in 29,837 summary rays to constrain lateral variation in the depth to the core-mantle boundary (CMB). We assumed a homogeneous outer core, and the data were corrected for mantle structure and inner-core anisotropy. Inversions of separate data sets yield amplitude variations of up to 5 km for PcP, PKPab, PKPbc, and PKKP and 13 km for PKPdf. This is larger than the CMB undulations inferred in geodetic studies and, moreover, the PcP results are not readily consistent with the inferences from PKP and PKKP. Although the source-receiver ambiguity for the core-refracted phases can explain some of it, this discrepancy suggest that the travel-time residuals cannot be explained by topography alone. The wavespeed perturbations in the tomographic model used for the mantle corrections might be too small to fully account for the trade off between volumetric heterogeneity and CMB topography. In a second experiment we therefore re-applied corrections for mantle structure outside a basal 290 km-thick layer and inverted all data jointly for both CMB topography and volumetric heterogeneity within this layer. The resultant CMB model can explain PcP, PKP, and PKKP residuals and has approximately 0.2 km excess core ellipticity, which is in good agreement with inferences from free core nutation observations. Joint inversion yields a peak-to-peak amplitude of CMB topography of about 3 km, and the inversion yields velocity variations of ±5% in the basal layer. The latter suggests a strong trade-off between topography and volumetric heterogeneity, but uncertainty analyses suggest that the variation in core radius can be resolved. The spherical averages of all inverted topographic models suggest that the data are best fit if the actual CMB radius is 1.5 km less than in the Earth reference model used (i.e. the average outer core radius would be 3478 km).