G43A-0967
Relation between the Earth's Spin Rate and Jet Streams
UT1 is the Earth's rotation angle with respect to the celestial reference frame and the time-derivative of UT1 gives the Earth's spin rate. However, the Earth's spin rate is not constant; it fluctuates due to the luni-solar tidal interaction and due to the angular momentum exchange with the geophysical fluids (especially, the atmosphere). The former effects are almost predictable (e.g. see IERS Conventions 2003), however the latter effects are unpredictable. In order to assess the atmospheric effects on the spin rate, we usually evaluate atmospheric angular momentum (AAM) functions. On the other hand, in UT1 data, we observe irregular fluctuations driven by the winds. Since the jet streams have large angular momentum and the jet streams sometimes change into meandering/zonal flow, it is expected that jet streams have large effects on the spin rate. In this paper, we focus on the relation between the spin rate and the meandering flow of the jet streams using daily or sub-daily UT1 data. In order to measure the meandering flow of the jet streams, we use an index called a `zonal index' appeared in meteorological study. We try to modify and optimize the definition of this index such that an index fits the axial AAM function of the jet stream regions best. We compare the time-variation of an index to that of the UT1 excitation calculated from the geodetic observation data and have found that such an index shows a good correlation with the observed UT1 excitation.
G43A-0968
Influence of irregular phase and amplitude variations in the Earth orientation parameters on their prediction errors
The forced annual oscillation in polar motion and UT1-UTC data is related to the seasonal thermal cycle therefore its phase fluctuates around its well-defined expected value. The phases of the free Chandler oscillation in polar motion and Free Core Nutation in precession nutation residuals have not expected values, so their phases may show a drift. The S-transform, the combination of the Fourier transform filter with the Hilbert transform, and complex demodulation approaches were applied to compute phase and amplitude variations of oscillations in the Earth orientation parameters (EOP) as a function of time and frequency. Comparing the time-frequency spectra of the amplitude and phase changes against the EOP prediction errors reveals how such errors are influenced separately by the irregular change of the amplitudes and/or the phases. It was shown that the irregular change of the annual oscillation in polar motion and length of day could be the main cause of the prediction error increase. The EOP data were forecast by the combination of the least-squares extrapolation and the autoregressive prediction.
G43A-0969
Tropospheric and stratospheric wind contributions to Earth's polar motion
The tropopause, a critical factor in climate dynamics studies, represents the boundary between the troposphere and stratosphere. It has been treated simply as an equal-height layer in previous estimates of the tropospheric and stratospheric wind contribution to the Earth's polar motion. However, the height of tropopause is not a constant everywhere. Instead, it changes with location and time. In this study, the NCEP model's variable height of tropopause is employed in computing the tropospheric and stratospheric wind contributions to the Earth's polar motion, from the National Centers for Environmental Prediction-National Center for Atmospheric Research (NCEP/NCAR) reanalysis 6-hourly wind and pressure fields during the period 2000-2005. It is found that the previous approach, which assumes an equal-height tropopause of 100 hpa, significantly underestimates the tropospheric and stratospheric wind contributions and the anti-correlation between them. Because of the considerable cancellation effect between the troposphere and stratosphere, the total winds contribute generally less than the surface pressure to the polar motion excitations at seasonal and intraseasonal time scales.
G43A-0970
Multivariate stochastic prediction of length of day and atmospheric angular momentum time series
Forecasting Length of Day (LOD) and axial component of Atmospheric Angular Momentum (AAM) time series was performed by the multivariate autoregressive models of order p. The optimum autoregressive orders were estimated by the Schwartz Bayesian Criterion (SBC) and the Akaike Information Criterion (AIC). The differences between LOD/AAM data and their multivariate or univariate predictions were computed for different starting prediction epochs for comparison of the unpredictable irregular variations in LOD and AAM data. The need of combining LOD with AAM data in forecasting LOD is indicated in terms of the comparison between the bivariate and univariate models yielding better accuracies of the predictions in the case of the multivariate one. The mean prediction errors of LOD time series were compared with those published on the web page of the Earth Rotation Parameters Prediction Comparison Campaign as well as those obtained by the IERS Rapid Service/Prediction Center.
G43A-0971
Virtual Polar Motion and Universal Time Variations in Space Geodetic Techniques due to Atmospheric Pressure Loading
Earth rotation variations, in polar motion and universal time (ERP), appear as a response due to the sum of solid Earth displacements, fluid and gaseous mass transports. In finite networks, e.g., the network of eleven operational VLBI stations during the CONT05 VLBI experiment, horizontal displacements due to atmospheric pressure loading (APL) may accidentally introduce a net rotation. Generally, a no-net-rotation is expected, hypothesizing a surface normal stress due to APL upon a radially symmetric Earth. However, the horizontal crustal deformations due to APL given on a 2.5x2.5 degrees grid provided by the Goddard VLBI Group show systematic temporal net rotations. We compared the change of the eleven station network of CONT05 with and without APL, every six hours, by a three Helmert parameter transformation (three rotations). The �virtual� predicted ERP variations were validated w.r.t. the estimated ones, obtained from CONT05 (using the OCCAM 61E VLBI software). These tiny ERP variations, representing about 2 mm on Earth's surface, could statistically be detected if more VLBI sessions were processed. Even the inverted and non-inverted barometric assumptions of the response of the oceans to atmospheric pressure variations could potentially be verified.
G43A-0972
Hourly angular momentum functions for excitation of Earth orientation parameters
Here we have used the NASA GEOS4 model � data assimilation system to calculate the excitations of the Earth orientation parameters, namely polar motion and length of day on an hourly basis based on fields of winds and surface pressure, representing the motion and mass terms, respectively. The resulting excitation functions agree reasonably with those from other systems reporting 6-hourly. The atmospheric system is updated by assimilating new observational data every six hours; however, in between these assimilation times, the numerical model forecasts the state of the atmospheric fields, which we extracted hourly for our calculations. The degree to which the forecasts connect smoothly to the assimilation state may be taken as a measure of success of the atmospheric model. We chose October 2002 for this study as it encompasses the CONT 2002 special observing period that was held for Earth rotation observations. The polar motion excitation terms based on winds, which are known to have a strong diurnal signal due in part to tidal fluctuations, appear to be relatively successful in this regard. Time-spectra of these wind-based, and the pressure-based terms reveal diurnal and semidiurnal signals in the prograde and retrograde senses, with other spectral peaks around 8 hours. Other terms may have some jumps between the forecast and analysis times at the six-hour mark, and we are considering a posteriori methods of reducing these discontinuities to derive useful signals. Newer models, including one from a new NASA initiative, are becoming available that would inherently reduce such differences between forecast and assimilation epochs, and we plan to analyze the atmospheric excitations from them in the future, focusing also on more recent intensive study periods for Earth rotation. Also, characteristics of the combination of such high frequency atmospheric excitations with those from an ocean model will be important to compare with spectra of the Earth orientation series.
G43A-0973
Seasonal terms in GPS coordinate time series: artifact or real signal?
Permanent GNSS networks on global, regional and local scales are serving both for reference frame maintenance and geophysical studies. The long term station coordinate time series allows the estimation of the linear station velocities, which then turn to be a basic input for tectonic interpretation. A more detailed time series analysis however can reveal further vital information. The analysis of the noise characteristics and the apparent seasonal signal in the time series development allows a deeper insight into some processes in the network and at the stations environment as well. An appropriate statistical analysis of the noise may provide information on the presence of the colored noise in the time series. This allows a more reliable velocity uncertainty estimation and may provide information on the monument stability. The seasonal coordinate variation may come from real physical sources (e.g. atmospheric and hydrologic effects), but may also introduced through the GPS data analysis (mismodeling) and some local environmental effects (e.g. multipath, thermal effects) may also contribute to the annual coordinate variations. In this paper, through case studies we try to distinguish the real and artificial agents, which contribute to the seasonal periodicity observed in the GPS time series. We use the coordinate time series of the EPN (EUREF Permanent Network), which has been installed in 1996 and now includes more than 180 permanent GNSS sites. The weekly combined sinex solutions are stacked with the CATREF software using the minimum constraint approach. As this combination step removes the common mode signal, we analyzed the station specific residual time series. The noise and harmonic analysis of the coordinate time series was done by the CATS software, which is based on the MLE approach. The results of the noise analysis, in agreement with the studies of other networks confirmed the presence of the colored noise. The harmonic analysis has shown moderate seasonal variation of the station coordinates. The distribution of the amplitudes and phases are fairly random and hard to interpret as a real geophysical signal. Rather they reflect the modeling shortcuts of antenna phase centre variations (PCV) and environmental impacts (multipath, thermal effects). The results of the study are largely supports the importance of the introduction of the absolute PCV models and then the general re-processing of the whole GPS observation material using the new ITRF2005 frame and the new PCV models.
G43A-0974
Real Time GPS- Satellite Clock Estimation � Development of a RTIGS Web Service
Since 3 years the IGS (International GNSS Service) Real-Time Working Group disseminates via Internet raw observation data of a subset of stations of the IGS network. This observation data can be used to establish a real- time integrity monitoring of the IGS predicted orbits (Ultra Rapid (IGU-) Orbits) and clocks, according to the recommendations of the IGS Workshop 2004 in Bern. The Institute for "Geodesy and Geophysics" of the TU-Vienna develops in cooperation with the IGS Real-Time Working Group the software "RTR- Control", which currently provides a real-time integrity monitoring of predicted IGU Clock Corrections to GPS Time. Our poster presents the results of a prototype version which is in operation since August this year. Besides RTR-Control allows for the comparison of pseudoranges measured at any permanent station in the global network with theoretical pseudoranges calculated on basis of the IGU- orbits. Thus, the programme can diagnose incorrectly predicted satellite orbits and clocks as well as detect multi-path distorted pseudoranges in real- time. RTR- Control calculates every 15 seconds Satellite Clock Corrections with respect to the most recent IGU- clocks (updated in a 6 hours interval). The clock estimations are referenced to a stable station clock (H-maser) with a small offset to GPS- time. This real-time Satellite Clocks are corrected for individual outliers and modelling errors. The most recent GPS- Satellite Clock Corrections (updated every 60 seconds) are published in Real Time via the Internet. The user group interested in a rigorous integrity monitoring comprises on the one hand the components of IGS itself to qualify the issued orbital data and on the other hand all users of the IGS Ultra Rapid Products (e.g. for PPP in Real Time).
G43A-0975
European Vertical Height Datums: Unification and Time variation
Estimation of the fundamental geodetic parameter, w$_{0}$, the potential value of the Gauss-Listing geoid, is required to define the geoid, the best fitting ellipsoid to the geoid and as a means of unification of the vertical datums in use by various mapping agencies throughout the world. Consideration should also be given to the time derivative of w$_{0}$ given sea-level change etc. Methodologies to quantify w$_{0}$ include (i) the use of tide gauge time series and collocated GPS and (ii) satellite altimetry. The first methodology enables global isostatic adjustment and other local deformation to be removed from the tide gauge time series. The tide gauge time series, levelling heights and the GPS coordinates provide a measure of the normal height at the tide gauge benchmark. By reference to a global geopotential model or high accuracy and high resolution local geoid model the equipotential value corresponding to mean sea-level can be recovered for each of the tide gauge locations. In this manner the equipotential surface that provides the best fit to mean sea-level is used as a reference for the unification purposes. Through annual investigations the methodology has the potential to estimate the rate of change of w$_{0}$. In particular this study is used to connect UK and European datums through analysis of seven tide gauges distributed around the UK coastline. GPS positioning is quantified by using the GIPSY OASIS II software. In the second methodology, the geodetic parameter w$_{0}$ can be recovered from altimetric mean sea-level models from TOPEX/Poseidon and ERS. This facilitates unification of datums by estimating local differences from the global altimetric value. This study will report and inter-compare results from the two methodologies.
G43A-0976
High-rate GPS accuracy: its spatio-temporal dependence
A device that we built to simulate the ground motions caused by transient geophysical signals such as earthquakes, volcanoes, and glacier flow, enables us to assess the technique of high-rate GPS. We have made extensive measurements with this system while inducing simulated motions on a GPS antenna with a temporal spectrum similar to real seismic events. The difference between the antenna position inferred from GPS observations and the known ��true'' antenna position yields directly the time-dependent error in the GPS measurements. We have determined that, for a particular test event, 96% of the 1-Hz GPS position estimates were in error by less than 5 mm, and the root-mean-square error of those estimates over the 15-min duration of the simulated seismic event was 2.5 mm. This accuracy was reached when using as a reference a secondary stationary GPS antenna at a distance of just ~10 m, therefore attaining a high degree of error mitigation, and an optimal satellite geometry. Such short baselines are typically not available in realistic geophysical situations. Similarly, the time of an event, and thus satellite geometry, cannot be controlled.� We will present an assessment of the dependence of accuracy on baseline and satellite geometry for several simulated seismic motions.
G43A-0977
The Fixed Point Theorem of Ambiguity Resolution for Precise Point Positioning of GPS Networks: Theory and Applications
The properties of ambiguity resolution are explored both theoretically and by experiment when applied to GPS network solutions that have first been derived by precise point positioning (PPP). Since its invention by {\it Zumberge et al}. [1997], PPP has become popular for regional network processing, because formal covariances between stations are zero, and so processing time scales linearly with the number of stations (unlike traditional processing models that scale quadratically). PPP network solutions can be further improved by the application of ambiguity resolution, however, the processing time for this step generally scales quadratically [{\it Blewitt et al}., 1989]. Furthermore, ambiguity resolution introduces inter-station correlations that cause subsequent network kinematic analysis to scale quadratically rather than linearly. Thus some of the practical advantages of PPP are subsequently lost. A theoretical understanding of how PPP networks respond to ambiguity resolution may point the way to faster, linear processing schemes, and may help to assess from a theoretical perspective current ad hoc schemes for producing solutions that are experimentally almost identical to optimal solutions. A reasonable condition for implementing such schemes is that the differences between optimal and sub-optimal solutions should be statistically insignificant ("near-optimal"). Here a fixed point theorem is derived, which identifies linear combinations of network parameters that are theoretically invariant under ambiguity resolution. This theorem is useful to assist the design and justification of near-optimal network processing schemes that scale linearly with the number of stations. In addition to reducing processing time, linear schemes readily lend themselves to parallel processor implementation, thus there is the potential to reduce real processing time by several of orders of magnitude for extremely large networks. Such schemes would allow for very rapid, multiple reanalysis of extremely large networks to assess various models.
G43A-0978
Total Least Squares in Geodetic Coordinate Transformation
Technological developments have made it possible to implement complicated algorithms into the geodetic applications. In fact these developments have opened up new perspectives for the solution of geodetic problems in different areas. One of these mathematical algorithms is called the Total Least Squares (TLS). In general, the Least-Squares (LS) estimation is used in geodetic problems. The LS estimation is based on the probability theory, adjusting a set of observations to make the sum of the squares of the residuals minimum. In this case, it is considered that the observations are only erroneous. Contrary to the LS estimation, TLS approach assumes that all or a portion of the variables can have error or noise. Therefore, it can be said that the TLS can provide more realistic results and can be applied in many fields of geodesy. Coordinate transformation and deformation analysis can be given as the examples of such applications. The studies show that, the TLS approach can be successfully applied in such studies. In this study, basic algorithm of the TLS approach is given and the results obtained through the TLS and LS approaches have been compared.
G43A-0979
Determination of a Global Reference Frame Based on a Reprocessed GPS Network
In a joint effort, TU Dresden and TU Munich/GFZ Potsdam reprocessed a global GPS network of more than 200 stations covering the time span from January 1994 to December 2005. As one final product, a global GPS-only reference frame is determined using a rigorous combination of daily normal equations. Our concept of the realization of the global terrestrial reference system follows the center of mass approach considering changing surface loads in a consistent way. The stability of our reference frame will be evaluated based on the obtained long term trends of station coordinates, load-induced deformation estimates and homogeneous time series of station positions. The geophysical implications of these results will be addressed. Finally, we will compare our solution with other recent terrestrial reference system realizations.
G43A-0980
A Comparison of GPS, VLBI and Model Estimates of Ocean Tide Loading Displacements
In recent years Ocean Tide Loading Displacements (OTLD) have been measured using the Global Positioning System (GPS) and Very Long Baseline Interferometry (VLBI). We assess the accuracy of GPS measurements of OTLD by comparison with VLBI measurements and estimates derived from numerical ocean tide models. A daily precise point positioning (PPP) analysis was carried out on ~11 years of GPS data for each of 25 sites that have OTLD estimates using data from co-located VLBI sites. Carrier phase ambiguities were fixed to integer values where possible. The resulting daily estimates of OTLD, at eight principal diurnal and semi-diurnal tidal frequencies, were combined in a Kalman Filter to give GPS measurements of OTLD at each site. The three-dimensional GPS and VLBI measurements of OTLD were compared with estimates computed (by convolution with Green's functions) from five modern ocean tide models. The GPS / model agreement is shown to be similar to the VLBI / model agreement. In the important radial direction, the GPS / model misfit is shown to be smaller than the VLBI / model misfit for six of the eight tidal constituents; exceptions being the K1 and K2 constituents. Fixing of GPS carrier phase ambiguities to integer values resulted in marginal improvement to the GPS / model agreement. Statistically, we show there to be no significance to the difference between the fit of the GPS and VLBI measurements of OTLD to modelled values. Equally, differences in fit of either GPS or VLBI to the five sets of model derived values cannot be identified with statistical significance. It is thus concluded that, overall, we cannot distinguish between GPS and VLBI measurements of OTLD, and that at the global scale, present ocean tide models are accurate to within the current measurement noise of these techniques. We go on to apply our refined GPS measurement technique to 19 sites in Antarctica with upto 10 years of data, comparing the resulting OTLD estimates to modelled values computed using regional and global numerical ocean tide models.
G43A-0981
Noise Characterization of the Four Techniques Entering in the Realization of the International Terrestrial Reference Frame
The residual signal in VLBI, SLR, DORIS and GPS station motion after a linear trend and seasonal components have been removed is analysed to investigate technique-specific error spectra. The study concentrates on 60 sites with dense observation history by two or more space geodetic techniques. Statistical methods used include Allan variance analysis and the three-cornered hat method. Technique-specific noise characteristics are estimated in several different ways, leading to a white noise diagnostic for VLBI, SLR, and DORIS, and flicker noise for GPS. The three-dimensional noise levels are 2 mm for VLBI, 3 mm for SLR, and 5 mm for DORIS and GPS.
G43A-0982
CALREF, a stable reference frame for the Northern California
The San Francisco Bay area (SFBA) is one of the most actively deformed areas in California. A large part of the deformation ($75 %$) between the north American plate and the Pacific plate is accommodated along faults lying in a land stripe of about 50 km width. At least two major events ($M_w\geq6.5$) are expected along two major faults: the San Andreas and the Hayward fault. Possible triggering between the two events is not excluded. Since 1994, the Berkeley Seismological Laboratory has been involved in collecting and processing data from the cooperative regional BARD network of permanent GPS receivers (Bay Area Regional Deformation network, reference: Murray et al., 1998).The Berkeley Seismological Laboratory (BSL) currently maintains 29 BARD sites and is in the process of upgrading its infrastructure so that it better responds to the needs of earthquake hazards related research, and in particular the use of GPS data in real-time. Most of the BARD stations operated by BSL are collocated with seismic instruments (broadband seismometers or geophones) or strain-meters. We present here for the first time the BARD GPS sites velocities computed in the ITRF2000 (Altamimi et al, 2002) from the data collected from 1994 to present. The velocity field highlights the deformation across the two main active faults in the SFBA. We have compared our results with the velocity field obtained from previous studies and obtained good agreement. In particular the ITRF2000 and the BAVU (d'Alessio et al., 2005) velocities will be discussed. Additionally, we discuss the velocities of sites belonging to the recently installed �mini-PBO� network which contribute to constraining the extent of the locked section on the northern part of the Hayward fault.
G43A-0983
Removing Grid Effect of 3-D Euler Deconvolution using Rotating Coordinate
A new method for removing grid effect of Euler deconvolution is suggested. Gridding is inevitable process for calculating Euler deconvolution, but in the process of gridding and calculating derivative using FFT, errors are generated like Gibb's phenomenon, edge effect, and circular convolution problem. Solutions which are locations and depths of anomaly bodies are calculated at all window locations, therefore, undesirable, scattered, and too many solutions are achieved and looks like meaningless. In this paper, rotating coordinate technique is used for removing grid effect of Euler deconvolution. Once Euler deconvolution is calculated for locations and depths of anomaly bodies at all window positions, the coordinate is rotated until 90 degree by 15 degree, and re-calculated for locations and depths. The process for removing grid effect is the follows: (1) Rotate the gridded potential data from 0 to 90 degree (2)Calculate the conventional 3-D Euler deconvolution for each data (3)Re-rotate Euler solution(locations and depths) to original coordinate (4) Collocation each solution and remove the solution which is not shown in others. Total 7 solutions can be achieved from these procedures, and common locations and depths results which are shown in all 5 solutions are accepted. The effectiveness of rotating technique is evaluated by using rectangular prism model which has 1km thickness at 5km depth from ground. Random noise also added for verifying the rotating technique. In case of adding noise, the depth accuracy become lower, but location still shows good estimate. This method is also implemented to analysis the tectonic interpretation of the Eastern Asia including Korea, China, and Japan using GRACE and CHAMP satellite gravity and magnetic data
G43A-0984
Seasonal Signal Variations in GPS Time Series in the Central Himalaya of Nepal
Continuous GPS stations (cGPS) have been deployed in Nepal, by the DMG and the CEA/DASE, and operated since 1997. Daily positions were processed using the Bernese V4.2 software (Beutler et al., 2001). Further, Quasi-Observation Combination Analysis (QOCA) software (Dong et al., 1998) modeling of the time series data was performed through sequential Kalman filtering, allowing adjustment for global translation and rotation of each daily solution. It appears that the dominant signal depicted by the time series is a secular pattern of interseismic strain which can be modeled from 19mm/yr creep along the MHT, assumed locked from the surface to beneath the front of the high range over a width of about 100km (Bettinelli et al., 2006). In addition ,the time series encompass strong seasonal variations visible on the vertical as well as on the horizontal components (bettinelli et al., 2006), respectively up to 50 mm and 30 mm. The purpose of this presentation is to analyse these time variations to try to understand its origin. First of all, we use meteorological data, acquired nearby each station, to improve estimation of the tropospheric delay time variations. Then, we estimate possible surface displacement variations due to seasonal loadings such as atmospheric, aquifer or snow loadings, using both elastic analytical approaches and finite element modeling encompassing a realistic topography. Finally, we document possible slow slip events and further discuss the magnitude of each contributions to the measured signal.
G43A-0985
Anomalous Harmonics in the Spectra of GPS Position Estimates
Prior studies of the power spectra of GPS position time series have found pervasive seasonal signals against a power-law background of white noise plus flicker noise. Dong et al. [2002] estimated that less than half the observed GPS seasonal power can be explained by redistributions of geophysical fluid mass loads. Much of the residual variation is probably caused by unidentified GPS technique errors and analysis artifacts. Among possible mechanisms, Penna and Stewart [2003] have shown how unmodeled analysis errors at tidal frequencies (near 12- and 24-hour periods) can be aliased to longer periods very efficiently. Signals near fortnightly, semi-annual, and annual periods are expected to be most seriously affected. We have examined spectra of the 167 sites of the International GNSS Service (IGS) network having at least 200 weekly measurements during 1996.0-2006.0. The non-linear residuals of the weekly IGS solutions that were included in ITRF2005 have been used. In order to improve the detection of common-mode signals, the normalized spectra of all sites have been stacked, then smoothed with a boxcar filter (0.03 cpy window), for each local N, E, and U component. The stacked, smoothed spectra are similar for all three components. Peaks are evident at harmonics of about 1 cpy up to at least 6 cpy, but the peaks are not at strictly 1.0 cpy intervals. Based on the 6th tone of the N spectrum, which is among the sharpest peaks, and assuming a linear overtone model, then a common fundamental of 1.043 cpy can explain all peaks well. A flicker noise power-law continuum describes the background spectrum down to periods of a few months, after which the residuals become whiter. Similiar tones are not apparent in the residuals of available SLR and VLBI sites, which are both 10 times less numerous and dominated by white noise. There is weak evidence for 1 cpy overtones in the spectra of atmospheric loadings but these are much noisier than for GPS positions. Alternatively, as pointed out to us by U. Hugentobler (U. Berne), the period of the 1.043 cpy frequency, about 341.5 days, is close to a "GPS year"; i.e., the interval required for the constellation to repeat its inertial orientation with respect to the sun. This could indicate that the harmonics are a type of systematic error related to the modeling of the GPS satellite orbits.
G43A-0986
Detecting geophysical signals in station position time series of ITRF2005 data
ITRF 2005 is the first terrestrial reference frame combined from weekly or sessionwise data sets of the geodetic space techniques VLBI, GPS, SLR and DORIS. This allows to detect time variable effects in station positions such as discontinuities or periodic signals and to account for them in the combination process. The paper discusses visible effects in the ITRF2005 data sets, caused by changes in instrumentation or to be interpreted as geophysical processes. We use the official data sets for ITRF2005 as provided by the IAG Services. At first a TRF solution was computed with these data sets including station positions and earth orientation parameters. The individual weekly or sessionwise station coordinates were then transformed by a similarity transformation to the TRF solution in order to obtain consistent time series. We analyse the various effects in the time series, try to detect their causes and compare them with information derived from geophyiscal models. Detected discontinuities are mainly caused by earthquakes, seasonal variations by geophysical effects such as atmospheric or hydrological loading. As a consequence future versions of the TRF should provide seasonal variations of station positions in addition to the linear velocities.
G43A-0987
Impact of Atmosphere-Ocean-Land Interactions on Short-Term Variations of Earth Rotation Parameters
The Earth's variable rotation is precisely observed by modern space geodetic techniques down to subdiurnal timescales. For the interpretation of observed rotational parameters a reliable and consistent representation of mass redistributions within and mass exchanges among the Earth's subsystems is crucial. Here, causative physical processes of observed rotational variations are investigated by means of a numerical model approach allowing mass and impulse fluxes among the subsystems atmosphere, ocean, and continental hydrology. Operational analysis data from ECMWF are used to consistently force a hydrological discharge model and a global model for the ocean's baroclinic circulation and ephemeral tides. Effects from sea ice, pressure tides as well as gravitational tides, loading, self-attraction and nonlinear interactions between the various dynamic components are considered. The unconstrained hydrology and ocean models are coupled via continental discharge in order to close the hydrological cycle. Thus, resulting estimates of excitations of rotational variations implicitely take into account effects of continental discharge on near-shore ocean dynamics as well as the impact of variations in total ocean mass due to time-varying atmospheric and continental freshwater fluxes. Focussing on the period 2001 - 2005, individual angular momentum contributions of the atmosphere, the oceans, and continental hydrology are separated and typical variation patterns of underlying physical processes are identified. Agreement of simulated and observed excitations are significantly improved when water mass fluxes from the continents into the oceans are applied as additional boundary condition in the ocean model, since accompanying modifications of oceanic angular momentum partly compensate hydrological angular momentum contributions. As atmospheric analyses from ECMWF are routinely available on a near real-time basis, numerical estimates of Earth rotation excitations as presented here are principally suitable for operational purposes.
G43A-0988
IERS Working Group on Prediction Plans and Activities
The International Earth Rotation and Reference Systems Service (IERS) has established a Working Group on Prediction to investigate what IERS prediction products are useful to the user community in addition to making a detailed examination of the fundamental properties of the different input data sets and algorithms. The major goals and objectives of the Working Group are to determine the desired Earth orientation prediction products, the importance of observational accuracy, which types of input data provide an optimal prediction, the strengths and weaknesses of various prediction algorithms, and the interactions between series and algorithms that are beneficial or harmful. The plans and activities of the Working Group will be summarized.