G33B-0686
Reanalysis of CORS and Global GPS Data at the National Geodetic Survey
We present current results and preliminary interpretations from an ongoing project at the National Geodetic Survey (NGS) to reanalyze Global Positioning System (GPS) data collected since 1994 at stations of the International GNSS Service (IGS) global tracking network and at stations of the U.S. Continuously Operating Reference Stations (CORS) network. Reanalysis of the global data is expected to be complete by January 2009, while the reanalysis of the CORS data will take much longer, perhaps until late 2011. The reanalysis is accomplished in two stages. The first stage is designed to obtain a consistent set of GPS satellite orbits, Earth Rotation Parameters (ERPs) and a time series of global station coordinates expressed in the current IGS Terrestrial Reference Frame, i.e., IGS05. The second stage is designed to obtain a time- series of station coordinates for the much denser CORS network. An important aspect to this two-stage approach is that a relatively uniform global network is used to determine the satellite orbits, ERPs and terrestrial frame, which are then fixed and used to position the CORS network accurately and consistently within the same framework. Both stages of the analysis use a) the PAGES software from NGS to preprocess and reduce the RINEX observations, and b) the CATREF software from Institut Géographique National to obtain regularized station coordinates and secular velocities. The CATREF software provides the added benefit of allowing weekly Helmert frame alignments to the long-term frame and detection of station discontinuities and other frame distortions in both the global and CORS solutions.
G33B-0687
The CORS Network Products and Services
The National Geodetic Survey (NGS) is responsible for defining, providing access and maintaining the
National Spatial Reference System (NSRS) which is the legally binding reference frame for the civilian sector
of the US Federal Government, and most of the public. With the increasing access to space geodetic data
sources, especially GPS, NGS has been transitioning to a space based NSRS. In 1994 NGS created the
Continuously Operating Reference Station (CORS) network as a partnership with site operators of
continuously operating GNSS sites who were willing to share their data with the public. The network currently
contains ~1,200 stations, located mostly in the United States, and is growing at ~ 200 stations a year. The
CORS network is particularly dense in the eastern United States, and its GPS observations are used
extensively by researchers.
NGS is taking a number of steps to improve its CORS related products and services including: 1) More
stringent requirements for constructing new CORS as well as operating existing stations, 2) Investigating
biases affecting GPS observations collected at CORS, in particular the effects of multipath, and 3) Improving
processing software of GPS observables to compute orbits and coordinate computation. This effort has been
emphasized because NGS is both an IGS analysis center and the current IGS analysis center coordinator.
NGS is reprocessing all CORS data collected since 1994 with the goal of updating the current legally binding
geometric reference frame known as NAD83. NGS has also improved and expanded its post-processing web
service tools, including the Online Positioning User Service (OPUS). Lastly NGS is now operating a prototype
real time GNSS data-delivery service broadcasting in RTCM format via NTRIP.
http://www.ngs.noaa.gov/CORS
G33B-0688
Extending HTDP to address crustal motion across Alaska
Deformation in the western United States, due to tectonic forces associated with the Pacific-North American
plate boundary, causes ongoing changes of the positions of points on the Earth's surface . As a result,
accurate surveying in the western US requires an equally accurate description of this deformation to allow
survey measurements conducted at different epochs to be corrected for such movement. NOAA's National
Geodetic Survey (NGS) has developed the HTDP (horizontal time dependent positioning) software that
enables its users to make these corrections for the horizontal component of motion. HTDP contains a model
of the secular (continuous) velocity field for the contiguous United States (from 125º to 100ºW longitude and
31º to 49ºN latitude) and separate models for the displacements associated with 28 earthquakes. The
software is updated periodically to address the displacements associated with new earthquakes, most
recently in June 2008 with the release version 3.0.
This presentation describes a major effort to update HTDP to accurately model earth deformation in Alaska.
As a start in this process, the recent release of HTDP introduced a model for the Denali Earthquake of 2002
to accompany its pre-existing model for the Prince William Sound earthquake of 1964. The Denali
Earthquake produced displacements of several meters near the fault and measurable displacements
throughout most of central Alaska. The dislocation model we used was developed by Elliott et al. (2007). We
tested the predicted movement associated with the Denali earthquake using a set of 126 measured
displacement vectors distributed over most of the interior of Alaska. The agreement between predicted and
observed displacements was good with a RMS misfit of 0.1m in both the north and east components. A few
large (> 1m) discrepancies were found in the vicinity of the fault trace. These discrepancies probably
indicate that complex fault geometry--that no simple dislocation model can match-- introduces significant
errors in predicting the deformation near the rupture plane.
In the future, NGS plans to partner with the earth science community to develop a model for Alaska's secular
velocity field and models for additional M>7 earthquakes that have occurred in Alaska since 2000.
Elliott, J. L., J. T. Freymueller, and B. Rabus (2007), Coseismic deformation of the 2002 Denali fault
earthquake: Contributions from synthetic aperture radar range offsets, J. Geophys. Res., 112, B06421,
doi:10.1029/2006JB004428.
http://www.ngs.noaa.gov/TOOLS/Htdp/Htdp.shtml
G33B-0689
IGS Working Group "Regional Dense Velocity Fields": Objectives and Work Plan
The IAG Working Group (WG) on "Regional Dense Velocity Fields" was created within IAG sub-commission
1.3 "Regional Reference Frames" at the IUGG General Assembly in Perugia in 2007. The goal of the
Working Group is to densify the latest realization of the ITRS and provide regional dense velocity information
in a common global reference frame.
For that purpose, working group members join efforts with the regional sub-commissions (AFREF, NAREF,
SIRGAS, EUREF, ·s ) and analysis groups processing data from local/regional continuous and episodic
GNSS stations.
In a first step, dedicated region coordinators will gather as many as possible velocity solutions for their region
(in accordance with the WG requirements) and combine these solutions with the sub-commission regional
solutions to produce a regional cumulative combined solution in the SINEX format.
In a second step, combination coordinators will perform combinations of the regional SINEX submissions and
SINEX solutions from global GNSS networks like e.g. TIGA. The purpose of multiple combination coordinators
is to evaluate both the results and different approaches. To assist in this task regional coordinators will solicit
discontinuity tables in addition to the weekly SINEX solutions.
At the same time, the WG will also study the strengths and shortcomings of local/regional and
continuous/episodic GNSS solutions to determine site velocities, and define optimal strategies for the
combination of regional and global SINEX solutions.
http://epncb.oma.be/IAG/
G33B-0690
A Software Development for an Alternative GPS Processing Strategy
The evolution of the layers on the earth's surface are realized according to crustal movements in geological time dimension. The values of the crustal movements vary depending on the structure of the plates in the regions such as Anatolia. Anatolia is located at an earthquake zone in the intersection of major plates Africa, Arabia and Eurasia, which causes plate movements continuously, and approximately 2-3 cm per year. If coordinate of any station point on the Earth liked to be defined accurately in an instant time, the velocities of the points must be regarded with coordinate values. Because the point coordinates change according to the time, most suitable approach should be determined to process and adjust the GPS measurements. The point movements due to tectonic activities may affect the coordinate solutions in long GPS campaigns. This is important for the projects carried out in tectonically active countries like Turkey. Through the use of new strategy presented in this paper, such uncertainties will be removed and more realistic solutions will be obtained. In the enclosure of the project of the renovation and densification of Istanbul GPS Network (IGNA); long period GPS measurements (21 March 2005 and 8 September 2005) are adjusted depending on mean weighted observation epoch. However, in this study, baseline components were shifted to the reference epoch (2005.0) using the velocities of the first order points. Velocities of the other control stations are calculated through interpolation using TNFGN (Turkish National Fundamental GPS Network) station velocity vectors. This strategy takes the time factor into account. The adjustment computations are carried out at the reference epoch. Therefore the effects of crustal movements in the long GPS campaigns can be minimised. A computer programme has been developed to change all baseline values without doing any editing manually. Baseline components and velocities are in two different input files. In this java based programme, using these values and identifying the reference epoch, corrected baseline components are computed. Then, the corrected ASCII file (output file) is imported to the commercial software and adjustment computations are done. It has been reached more realistic coordinate values using this strategy. Moreover, these coordinate values are compared with IGNA 2005 project coordinate outputs, and differences are estimated and interpreted.
G33B-0691
An Experimental Approach to Improve the International Terrestrial Reference Frame
To monitor global geophysical changes with realistic uncertainties using millimeter-precision geodesy, it is essential to define, realize and maintain the International Terrestrial Reference Frame (ITRF) consistently and accurately. For example, reliable determinations of geocentric site coordinates and motions, satellite orbits, geocenter motion, Earth orientation and its variations, mean sea level rise, and polar ice mass changes at various time scales all depend critically on the accuracy and stability of the ITRF. The most recent ITRF2005 realization uses time series from all space geodetic techniques and is a major step forward in the transition to fully time-variable geodesy. However, ITRF2005 still assumes linear motion models for geodetic sites, and the combination of different technique data is done only for their separate secular solutions. The lack of non-linear motion in the model implies that the frame is only averagely defined, which complicates near-instantaneous geophysical monitoring facilitated by revolutionary global geodetic data. The neglected centimeter-level non-linear motion can alias into errors in the linear model, which can be significantly compounded by the heterogeneity of data distribution in space and time. We define a nearly instantaneous reference frame by specifying steady frame parameters and combining different technique data weekly. A Kalman filter and smoother algorithm has been developed and coupled to the ITRF/CATREF software to solve for time-variable weekly coordinates, as well as a model of secular, periodical and stochastic motion components. With significant progress being made, we will present preliminary results of the new experimental approach to ITRF realization.
G33B-0692
Seasonal signals in the reprocessed GPS coordinate time series
The global (IGS) and regional (EPN) CGPS time series have already been studied in detail by several authors to analyze the periodic signals and noise present in the long term displacement series. The comparisons indicated that the amplitude and phase of the CGPS derived seasonal signals mostly disagree with the surface mass redistribution models. The CGPS results are highly overestimating the seasonal term, only about 40% of the observed annual amplitude can be explained with the joint contribution of the geophysical models (Dong et al. 2002). Additionally the estimated amplitudes or phases are poorly coherent with the models, especially at sites close to coastal areas (van Dam et al, 2007). The conclusion of the studies was that the GPS results are distorted by analysis artifacts (e.g. ocean tide loading, aliasing of unmodeled short periodic tidal signals, antenna PCV models), monument thermal effects and multipath. Additionally, the GPS series available so far are inhomogeneous in terms of processing strategy, applied models and reference frames. The introduction of the absolute phase center variation (PCV) models for the satellite and ground antennae in 2006 and the related reprocessing of the GPS precise orbits made a perfect ground and strong argument for the complete re-analysis of the GPS observations from global to local level of networks. This enormous work is in progress within the IGS and a pilot analysis was already done for the complete EPN observations from 1996 to 2007 by the MUT group (Military University of Warsaw). The quick analysis of the results proved the expectations and the superiority of the reprocessed data. The noise level (weekly coordinate repeatability) was highly reduced making ground for the later analysis on the daily solution level. We also observed the significant decrease of the seasonal term in the residual coordinate time series, which called our attention to perform a repeated comparison of the GPS derived annual periodicity and the surface mass redistribution models. We expect that using the reprocessed EPN data we can exclude several analysis related artifacts and we get a more clear view on the real physical information content of the data. In this paper we present a general overview and results of the EPN reprocessing and we show the detailed results of the harmonic analysis.
G33B-0693
Earth System Center to TRF Origin Weekly Series From SLR Analysis
Since the first realization of the International Terrestrial Reference Frame (ITRF), its origin, defined to coincide with the geocenter, has been realized through the estimated coordinates of its defining set of positions and velocities at epoch. Satellite Laser Ranging (SLR) contributes to the ITRF realization this unique information along with that for its absolute scale, for over two decades. Over the past decade, the focus extended beyond the accuracy at epoch to include the stability of these realizations, given the increasingly more accurate observations of geophysical mass redistribution within the Earth system. Driven by numerous geophysical processes, the continuous mass redistribution within the Earth system causes concomitant changes in the long-wavelength terrestrial gravity field that result in geometric changes in the figure described by the tracking station network. The newly adopted ITRF development approach allows the simultaneous estimation of origin variations at weekly intervals through a geometric approach during the stacking step, and for the first time in the history of the ITRF accounts to some extent for these effects. Our dynamic approach has been used since the mid-90s, delivering, initially biweekly and later on, weekly variations of an "origin-to-geocenter" vector, simultaneously with an SLR-only TRF realization. Over the past year, the International Laser Ranging Service's (ILRS) Analysis Working Group adopted significant modeling improvements for the SLR data reduction of future as well as the historical SLR data. Based on this new standards, we have reanalyzed the LAGEOS 1 & 2 SLR data set up to present, and developed a uniformly consistent set of weekly variations with respect to a frame realized simultaneously by the ensemble of the data, closely approximating the current (scaled) ITRF2005. These series can complement the precise application of the ITRF when used as Cartesian offsets or the GRACE-derived monthly gravitational models, when converted to degree-1 harmonics. A simple model based on the dominant frequencies decomposition of the series can be easily used to account for the most significant part of the signal in various applications (examples).
G33B-0694
Can we measure Geocenter Motion accurately?
It is ten years since the IERS Campaign to investigate motion of the geocenter. One conclusion from that campaign was "It appears that even if Space Geodesy geocenter estimates are sensitive to seasonal variations, the determinations are not yet accurate and reliable enough to adopt an empirical model that would represent a real signal". So where are we 10 years later? Even putting aside the fact that we have ten years of new data, we have seen relatively dramatic improvements in data processing methodologies and models (e.g. atmospheric mapping and propagation functions, satellite force models and sub-daily ocean loading corrections) and the spatial expansion of geodetic networks. This period has also seen the development of a number of differing "deformation" estimation approaches, where geocenter motion can be inferred from purely geometric techniques (even VLBI) via the Earth's elastic properties. Even more recently, geocenter motion estimates are starting to appear from ranging to the GRACE satellites and via the combination of GRACE gravity fields with assumptions on land-ocean interactions. We compare estimates of geocenter motion in an attempt to gauge accuracy. We also present new estimates from re-processed GPS and SLR data using various direct and "deformation" inversion methods. We try as much as possible to investigate the effects of specific data processing changes upon the geocenter motion estimates.
G33B-0695
The Effect of Future Thermal Sea Level Changes on the Excitation of Earth Orientation Variations
We present a mechanism by which steric sea level rise leads to changes of the Earth's orientation parameters, e.g., polar motion (PM) and length-of-day (LOD) on decadal to centennial timescales. Steric sea level change is commonly considered unable to excite changes in Earth's orientation parameters. We show, however, that a causal link exists since thermal sea level rise leads to horizontal mass redistribution within ocean basins and hence to ocean bottom pressure changes. The projected changes of ocean angular momentum in the 21st and 22nd century are derived from simulations with the coupled climate models ECHAM5/MPI-OM and GFDL-CM2.1, forced with the IPCC-A1B emission scenario. The net effect is a mass transfer from the southern to the northern hemisphere, and also a net movement of mass closer towards Earth's axis of rotation. This, in turn, leads to a small negative LOD trend of about -0.002 ms per 10 millimeter of steric sea level rise. For polar motion, we project that ocean warming excits a movement of about 1.7 milli-arcseconds per 10 millimeter of steric sea level rise, nearly linearly polarized towards 150 degrees West. Because steric sea level rise was too small over the last 50 years, this mechanism cannot account for unexplained observed decadal fluctuations of PM and LOD during this period. However, as ocean warming and steric sea level rise are expected to accelerate in the decades to come, we conjecture that steric sea level rise could contribute increasingly more to Earth orientation variations in the near future.
G33B-0696
Monitoring the ITRF/EOP consistency
Space techniques (VLBI, GPS, SLR, DORIS) are contributing to the determination of both terrestrial frame and Earth Orientation Parameters (EOP). The ITRF2005 is the first rigorous combination ensuring TRF and EOP consistency. However the current EOP reference (IERS 05C04) is so far derived from the combination of independent time series of the various techniques. In order to monitor the evolution with time of the consistency between both the ITRF2005 and IERS 05 C04, two procedures are used: (1) using CATREF combination model involving station positions and EOPs, and (2) using EOP-only combination method as it is currently performed at the EOP product centre. Comparisons of results obtained by the two procedures are discussed in an attempt to assess the current accuracy of EOP determination by space geodesy techniques. First results seem to indicate that the consistency between the two methods is at the level of 30 microarcseconds
G33B-0697
IERS Working Group on Prediction: Preliminary Analysis
The International Earth Rotation and Reference Systems Service (IERS) has established a Working Group on Prediction (WGP). The goals are to investigate the needs of the user community regarding IERS Earth orientation prediction products and to examine in detail the fundamental properties of possible input data sets and prediction algorithms. The first task is complete. The user community needs increased accuracy and there seems to be a growing interest in daily and sub-daily prediction. The task to understand fundamental properties of input data sets and algorithms is in progress. A repository for data sets and results was established at the University of Luxembourg, input data sets were identified and placed in the repository, algorithms were identified, and information on various algorithms was gathered. Preliminary results are presented.
G33B-0698
Earth-Atmosphere interaction at geological timescale
The Earth-atmosphere interaction is known to be one of the major source of variation of the length-of-day change up to 10 years, when the core is supposed to be the dominant effect. Nevertheless, major climate change occurs, the change in atmospheric circulation might affect the Earth rotation variation to an extend comparable or larger than the core. We investigate a broad range of Earth-atmosphere interaction situation, by using climate model from the past, at time where the circulation was very different, due for instance to ocean-continent repartition very different from now. We show that the effect on the LOD is at the level of a few millisecond, and how this change can be interpreted in terms of the geological/climatic properties of the epoch.
G33B-0699
Tidal Dissipation and Dynamical History of the Earth Moon System - Revisited
Solar tidal torque significantly affects the deceleration of the Earth spin rotation, although the angular momentum transfer to the Earth orbital rotation can not be easily detected. Often the total angular momentum of the Earth Moon System has been regarded as constant in time. We are trying to calculate again the past rotational state of the Earth and Moon with including the angular momentum leakage and other factors.
G33B-0700
Free Core Resonance Parameters From Strain Tides Recorded by Paired Extensometers at Gran Sasso, Italy
Two crossed 90-m long laser extensometers are operating in the Gran Sasso underground observatory (Central Italy) since several years. Because of their stability, sensitivity (at the picostrain level) and underground location (1400 m under the free surface) these instruments provide strain records showing high S/N ratio in the tidal bands and allow detailed analysis of the Free Core Nutation (FCN), a rotational eigenmode giving a resonance in the tidal response of the Earth in the diurnal band. It is well-known that tidal strain data are affected by ocean loading, local distortion of the strain field, and environmental parameter effects. The main contribution to ocean loading at the instrument site comes from the Mediterranean Sea, which is badly represented in global ocean models. We compute ocean loading effects using Agnew's SPOTL, after including a new Mediterranean local model from the Oregon State University, at 1/30 degree resolution. We estimate local effects using reference strains due to Earth tidal components in the diurnal (far from the FCR frequency) and semi-diurnal bands and a matrix representation of the relationship between instrument and remote strains. Tidal analysis of 4.5 years of strain, pressure and temperature data is carried out by using VAV/2003 code. Inversions for resonance parameters is performed using NA (Neighbourhood Algorithm); probability density distributions of parameters is obtained using NAB (NA-Bayes). Robustness of the results, e. g. with respect to changes in the adopted solid Earth and global ocean models, is checked for. We obtain values of the FCN period in agreement with the most recent determinations. As regards the quality factor, we show that values lower than about 20000 are inconsistent with our data.