[1] Apparent seasonal site position variations are derived from 4.5 years of global continuous GPS time series and are explored through the ''peering'' approach. Peering is a way to depict the contributions of the comparatively well-known seasonal sources to garner insight into the relatively poorly known contributors. Contributions from pole tide effects, ocean tide loading, atmospheric loading, nontidal oceanic mass, and groundwater loading are evaluated. Our results show that $40% of the power of the observed annual vertical variations in site positions can be explained by the joint contribution of these seasonal surface mass redistributions. After removing these seasonal effects from the observations the potential contributions from unmodeled wet troposphere effects, bedrock thermal expansion, errors in phase center variation models, and errors in orbital modeling are also investigated. A scaled sensitivity matrix analysis is proposed to assess the contributions from highly correlated parameters. The effects of employing different analysis strategies are investigated by comparing the solutions from different GPS data analysis centers. Comparison results indicate that current solutions of several analysis centers are able to detect the seasonal signals but that the differences among these solutions are the main cause for residual seasonal effects. Potential implications for modeling seasonal variations in global site positions are explored, in particular, as a way to improve the stability of the terrestrial reference frame on seasonal timescales.

[1] Spatial filtering is an effective way to improve the precision of coordinate time series for regional GPS networks by reducing so-called common mode errors, thereby providing better resolution for detecting weak or transient deformation signals. The commonly used approach to regional filtering assumes that the common mode error is spatially uniform, which is a good approximation for networks of hundreds of kilometers extent, but breaks down as the spatial extent increases. A more rigorous approach should remove the assumption of spatially uniform distribution and let the data themselves reveal the spatial distribution of the common mode error. The principal component analysis (PCA) and the Karhunen-Loeve expansion (KLE) both decompose network time series into a set of temporally varying modes and their spatial responses. Therefore they provide a mathematical framework to perform spatiotemporal filtering. We apply the combination of PCA and KLE to daily station coordinate time series of the Southern California Integrated GPS Network (SCIGN) for the period 2000 to 2004. We demonstrate that spatially and temporally correlated common mode errors are the dominant error source in daily GPS solutions. The spatial characteristics of the common mode errors are close to uniform for all east, north, and vertical components, which implies a very long wavelength source for the common mode errors, compared to the spatial extent of the GPS network in southern California. Furthermore, the common mode errors exhibit temporally nonrandom patterns.

[1] We use 2.5 to 14 years long position time series from >800 continuous Global Positioning System (GPS) stations to study vertical deformation rates in the Euro-Mediterranean region. We estimate and remove common mode errors in position time series using a principal component analysis, obtaining a significant gain in the signal-to-noise ratio of the displacements data. Following the results of a maximum likelihood estimation analysis, which gives a mean spectral index~À0.7, we adopt a power law + white noise stochastic model in estimating the final vertical rates and find 95% of the velocities within ±2 mm/yr, with uncertainties from filtered time series~40% smaller than from the unfiltered ones. We highlight the presence of statistically significant velocity gradients where the stations density is higher. We find undulations of the vertical velocity field at different spatial scales both in tectonically active regions, like eastern Alps, Apennines, and eastern Mediterranean, and in regions characterized by a low or negligible tectonic activity, like central Iberia and western Alps. A correlation between smooth vertical velocities and topographic features is apparent in many sectors of the study area. Glacial isostatic adjustment and weathering processes do not completely explain the measured rates, and a combination of active tectonics and deep-seated geodynamic processes must be invoked. Excluding areas where localized processes are likely, or where subduction processes may be active, mantle dynamics is the most likely process, but regional mantle modeling is required for a better understanding.

No abstract

Abstract. We collected GPS data from the southern Tarim basin, the Qaidam basin, and the western Kunlun Shan region between 1993 and 1998 to determine crustal deformation along the Altyn Tagh fault system at the northern margin of the Tibetan plateau. We conclude from these data that the Altyn Tagh is a left-lateral strike slip fault with a current slip rate of •9 mm/yr, in sharp contrast with geological estimates of 20-30 mm/yr. This contrast poses an enigma: because the GPS data cover a wider region than the geologic data, they might be expected to reveal somewhat more slip. We also find that the Tarim and Qaidam basins behave as rigid blocks within the uncertainty of our measurements, rotating clockwise at a rate of •11 and •4.5 nrad/yr, respectively, with respect to the Eurasia plate. The rotation of the Tarim basin causes convergence across the Tian Shan, increasing progressively westward from •6 mm/yr at 87øE to •18 mm/yr at 77øE. Our data and other GPS data suggest that the Indo-Asia collision is mainly accommodated by crustal shortening along the main Himalayan thrust system (•53%) and the Tian Shan contractional belt (•19%). Eastward extrusion of the Tibetan plateau along the Altyn Tagh and Kunlun faults accommodates only •23% of the Indo-Asia convergence.