Observation of the Earth System from Space

In the summer of 2000 the geo-research satellite CHAMP was launched into orbit. Its innovative payload arrangement and its low injection altitude allow CHAMP to simultaneously collect almost uninterrupted measurement series relating to the Earth gravity and magnetic fields at low altitude. In addition, CHAMP sounds the neutral atmosphere and ionosphere using GPS observations onboard. After 60 months in orbit one arrives at a very positive conclusion for the CHAMP mission. The CHAMP satellite and its instruments have been operated almost uninterruptedly since launch. The great performance of the satellite subsystems and of the mission operation specialists has made it possible to keep CHAMP in the science operation mode for most of the time and in addition to lift its orbit two times. After a series of calibration and validation activities in the course of the mission, which included a number of onboard software updates and parameter adjustments, CHAMP has been providing excellent measurements from its state of the art instruments for now more than 4 years. The effective and steadily functioning of the CHAMP Science Data System and the supporting tracking networks has made it possible to provide large quantities of pre-processed data, precision data products and auxiliary information to hundreds of registered users in an almost uninterrupted manner. This was only possible due to the funding of the project DACH (CHAMP Data Acquisition and Data Use) within the ‘GEOTECHNOLOGIEN’ R+D programme of the BMBF. With the orbit altitude being presently about 60 km higher than originally planned for mid 2005, CHAMP will very likely orbit the Earth for another 3 years at quite low altitude. This mission extension at low altitude will make CHAMP a pioneering long-duration mission for geo-potential research and sounding of the atmosphere.

The GeoForschungsZentrum Potsdam (GFZ) runs an operational system for the CHAMP mission that provides precise orbits on a regular basis. Focus is put on recent analyses and achievements for the Rapid and Ultra-rapid Science Orbits.

Based on very accurate kinematic CHAMP orbits, a new CHAMP gravitational field model was computed by means of a (point-wise) acceleration approach. In order to implement such an acceleration approach, the satellite’s acceleration has to be derived from the kinematic CHAMP orbits by means of interpolation and subsequent numerical differentiation. The iterative method of preconditioned conjugate gradients is implemented to solve the large linear system of equations for the spherical harmonic coefficients. If appropriate preconditioning is applied, convergence can be reached within 7 – 15 iterations. An important topic concerning the accuracy of the gravity field solutions is the detection and filtering or down-weighting of spikes, jumps, outliers and inaccurate data in the kinematic orbits. These problems are adressed by data-preprocessing or robust estimation. Different gravity field solutions up to degree and order 90 were computed, where validation exhibits a signal-to-noise (S/N) ratio per degree of S/N ≥ 1 for coefficients up to degree 80 and S/N ≥ 2 for coefficients up to degree 70. Comparisons to different CHAMP-models, which were obtained by application of alternative algorithms, prove that the acceleration approach can compete with other methods of gravity field determination.

Atmospheric perturbations induced by weather fronts, nuclear explosions, volcano eruptions, and earthquakes can generate signatures in the ionospheric plasma density by atmospheric-ionospheric coupling processes. Because of their sensitivity to the ionospheric ionization, ground and space based GPS measurements offer a unique opportunity for detecting earthquake signatures in the ionosphere. Although numerous case studies and statistical analyzes were made, the GPS radio occultation measurements on CHAMP did not show a clear ionospheric response to earthquakes. On the other hand the retrieved total electron content (TEC) data along numerous ray paths between ground based receivers and GPS satellites has shown clear earthquake related signals for selected earthquakes of magnitudes larger than 6. By using the dense GPS network in North America, earthquake related structures have been found after the Denali earthquake on November 3, 2002 and during the California earthquake on December 22, 2003. Single station observations revealed also typical earthquake signatures after the Sumatra earthquake on December 26, 2004. It is assumed that these significant structures are generated by upward propagating atmospheric acoustic waves which are excited by seismic surface waves. Detection techniques and wave propagation features are discussed.

The German CHAMP (CHAllenging Minisatellite Payload) satellite provides continuously GPS radio occultation data since February 2001. The measurements are analyzed by an operational satellite orbit and occultation processing system at GFZ. In total, more than 200,000 precise globally distributed vertical profiles of refractivity, temperature and water vapor are provided as of June 2005. The operational ground infrastructure from GFZ allows for the demonstration of a rapid data analysis since February 2003. A mean delay between measurement and provision of atmospheric excess phase data of ∼4 hours is continuously reached. Results of various validation studies with data from meteorological analyzes of the European Centre for Medium-Range Weather Forecasts (ECMWF) and the global radiosonde network indicate an excellent quality of the CHAMP data. But in the lower troposphere systematic deviations are observed, the known negative refractivity bias of the occultation data. It is most pronounced in the Tropics and is also observed by other GPS occultation instruments. The CHAMP data stimulated a number of studies to investigate the observed bias and to improve the GPS occultation data quality in the lower troposphere. First radio occultation measurements from the GRACE-B (Gravity Recovery And Climate Experiment) satellite are available for a 25 h period on July 28/29, 2004.

The GRACE Information System and Data Center designed, implemented and operated for the management of all scientific GRACE products is one fixed part of the GRACE Science Data infrastructure. The main objectives for the long term archiving and the worldwide dissemination of GRACE data and metadata driven by user requirements as well as the underlying software development work-flow from the business modelling, analysis and design up to the implementation and deployment steps are described in the following article.

GFZ is responsible for routine calculation of atmospheric and oceanic mass variations which have to be considered during GRACE precise orbit determination and calculation of gravity field partial derivatives. This Level-1B Atmosphere and Ocean De-aliasing product (AOD1B) is made available to the GRACE Science Data System and user community in terms of spherical harmonic coefficients with a maximum time delay of about 3–4 days dependent on the availability of required ECMWF meteorological fields. The spatial and time-variable vertical structure of the atmosphere is taken into account by vertical integration of the atmospheric masses. Oceanic mass variations are derived from a barotropic ocean model (PPHA) which was provided by JPL. The individual atmospheric and oceanic contributions as well as the processing strategy to derive the combined AOD1B product are described in the first part of this paper.The PPHA model has some deficiencies such as the exclusion of the Arctic Ocean or reduced level of energy compared to in-situ ocean bottom pressure data. Thus, the influence of different non-tidal ocean models on GRACE gravity field solutions has been investigated for a seasonal cycle. It turned out that the barotropic MOG2D and the baroclinic OMCT models, both providing global output and based on more complex algorithms and parameterization, produce slightly better agreement when compared to NIMA gravity anomalies or to an altimeter-derived geoid. Similar results are obtained when comparing daily times series of 10×10 degrees gravity field models, which have been derived without correcting short-term mass variations, with the candidate non-tidal ocean models.These tests indicate that the PPHA model shall be substituted by OMCT or MOG2D. Nevertheless, a dramatic improvement of the monthly gravity field solutions towards the pre-launch simulated baseline accuracy will not be reached. Instead, future work should primarily concentrate on the improvement of the temporal resolution and the inclusion of short-term (daily) hydrological mass variations.

In this article performance estimates for the accelerometers, the star sensors and the K-Band ranging system onboard the GRACE satellites are given. It is shown that the accelerometers perform slightly worse than specified and that the star sensor and K-Band ranging system performances agree with the specifications. It is also demonstrated that mainly for the accelerometers performance assessment further investigations are needed as effects of unknown origin affect the measurements. For each instrument the results from the highrate L1a to the filtered and downsampled L1b data processing are shown and discussed. Concerning the accelerometer processing, good agreement with the data provided by JPL has been reached, concerning the star sensors and the K-Band system differences remain.

Based on the GRACE mission data, a new era of static and time-variable gravity models with unprecedented resolution and accuracy have been generated by the GRACE Science Data System teams. In general, the spatial resolution of the field from pre-CHAMP satellite only models of about 1000 km can be increased by a factor of 5 – 6 thanks to the micrometer-precise K-band intersatellite link. The currently obtained gain in accuracy reaches one to two orders of magnitude, compared to the most advanced combination gravity pre-CHAMP models, but is still one order of magnitude away from the projected GRACE baseline accuracy.In this article we highlight the advances in gravity recovery with GRACE, based on recent results from GFZ Potsdam for a new GRACE-only medium-wavelength gravity model, called EIGEN-GRACE03S, a new combined high-resolution model complete up to degree and order 360, called EIGEN-CG03C, and the derivation of time-variable gravity signals from monthly GRACE-only gravity models.Evaluation of EIGEN-GRACE03S and EIGEN-CG03C shows that both models benefit in its long-to-medium wavelength part from an extended data base for GRACE, an augmented processing of the GRACE data as well as a meanwhile more complete and homogeneous compilation of surface data. The progress in resolution and accuracy with respect to earlier GRACE-based gravity models is moderate but visible at the level of 1 – 2 percent for standard comparisons.The derivation of time-variable gravity signals from a time series of 16 monthly GRACE-only gravity solutions reveals the mission’s sensitivity to hydrology-induced surface mass variations. The annual-varying signal on global and regional scales can be resolved down to spatial scales of a few hundred kilometers and the estimates are well above the assumed error level of the GRACE gravity solutions. Observable discrepancies with respect to the signal amplitudes, phases and spatial distribution indicate the potential contributions from GRACE to hydrological modelling, but also reveal systematic errors in the GRACE monthly fields.

The signal content in the low-low SST observables of the gravity field twin-satellite mission GRACE (Gravity Recovery And Climate Experiment) varies in the space domain depending on the roughness of the gravity field features. On the one hand, the maximum degree of the spherical harmonic expansion has to be selected as high as possible to bring out the maximum of gravity field information out of the data. On the other hand, an increasing maximal degree deteriorates the stability of the normal equations to solve for the gravity field parameters. Therefore, a trade-off is necessary between the selection of a maximal degree adequate for representing the signal content in the observables, on the one hand, and a maximal degree which can still be recovered without causing instabilities, on the other hand. We propose to integrate the global gravity field recovery with regional gravity field refinements tailored to the specific gravity field features in these regions: In a first step, the gravity field only up to a moderate safely determinable degree is recovered; the specific analysis features tailored to the individual gravity field characteristics in areas of rough gravity field signal will be modelled subsequently by space localizing base functions in a second step. In a final third step, a spherical harmonic expansion up to an (in principle) arbitrary degree can be derived based on a numerical Gauss — Legendre - quadrature procedure without any stability problems. The procedure will be applied in a first example to observations of a GRACE simulation scenario to test the potential capabilities of the approach. A second application demonstrates the determination of a global gravity field model and regional refinements based on low-low SST data of the GRACE twin satellite mission for the August 2003 observations.

This article describes an approach for global mapping of the Earth’s gravitational field developed, tested and successfully implemented at the Geodetic Institute of the Stuttgart University. The method is based on the Newtonian equation of motion that relates satellite-to-satellite tracking (SST) data observed by the two satellites of the Gravity Recovery And Climate Experiment (GRACE) directly to unknown spherical harmonic coefficients of the Earth’s gravitational potential (geopotential). Observed values include SST data observed both in the low-low (inter-satellite range, velocity and acceleration) and the high-low (satellites’ positions) mode. The low-low SST data specific for the time being to the GRACE mission are available through a very sensitive K-band ranging system. The high-low SST data are then provided by on-board Global Positioning System (GPS) receivers. The article describes how the mathematical model can be modified. The geopotential is approximated by a truncated series of spherical harmonic functions. An alternative approach based on integral inversion of the GRACE data into the geopotential is also formulated and discussed. The article also presents sample numerical results obtained by testing the model using both simulated and observed data. Simulation studies suggest that the model has a potential for recovery of the Stokes coefficients up to degree and order 120. Intermediate results from the analysis of actual data have a lower resolution.

High frequency temporal gravity changes on sub-monthly time scales are caused by Earth’s mass transport primarily originating from tidal and nontidal atmospheric and oceanic motions. Exploitation of precise GRACE satellite-to-satellite ranging measurements now makes it possible to monitor these changes on a global scale with a moderate spatial resolution. Thus, a time series of daily 10 × 10 Earth gravity models has been produced for the time span from July 2 to September 30, 2003. The solid Earth, oceanic and atmospheric tidal forces are accounted for during data processing, while variations in the gravitational potential produced by non-tidal atmosphere and ocean mass transport are omitted in the apriori force models. The recovered gravity changes are then compared to non-tidal model predictions. It is found that the agreement varies with the degree and order of the gravity harmonics. Generally, the recovered harmonics are highly correlated with the models for low order coefficients but the agreement degrades when the order becomes larger. Our results prove that GRACE is able to trace geophysical signals at short time scales, and that GRACE data can be useful to validate model predicted large-scale mass transports. Once we are able to separate tidal and nontidal signals for a longer time span, daily gravity recovery might also prove to be useful to study gravity changes taking place at sub-monthly time scales, such as oceanic tides.

After an overview of approaches and results in precise orbit determination (POD) for the CHAMP satellite in the Low Earth Orbit (LEO) we focus on the relations between kinematic POD and gravity field determination. We discuss determination of kinematic velocities out of kinematic positions that enter the gravity field determination in the form of kinetic orbital energy. After testing several numerical differentiation techniques, we selected conceptually two alternative methods, the Newton-Gregory interpolation and the smoothing cubic spline function. Finally, performance of numerical differentiation techniques for the CHAMP orbit is presented based on the gravity field determination.

GOCE will be the first satellite mission equipped with a gravity gradiometer. In order to achieve maximum precision and spatial resolution, the instrument is guided around the Earth in an extremely low orbit, employing active along track drag-free control and angular control by magnetic torquers. Furthermore, the orbit trajectory is determined very accurately by continuous and three-dimensional GPS satellite-to-satellite tracking. These mission characteristics are modelled by a system of two sequential simulators. The sensor system simulator computes the interaction of the complete sensor system and provides time series or power spectral densities of the gradiometer components. The mission simulator derives the geoid and gravity model performance. It takes as input mission and orbit parameters, expected GPS performance as well as the gradiometer error spectral densities derived from the sensor system simulator.The computational effort of the actual data analysis can only be managed by powerful computer systems, in principle, due to the large number of observations and unknown gravity field parameters. In this article a Semi-Analytical Approach is presented; it is a simple and fast alternative to a direct solution. It is based on simplifying assumptions, which allow to use FFT-techniques. It can be divided in two approaches: the 1D-FFT approach, and the 2D-FFT approach or torus-approach. In several case studies, the basic properties of the two approaches are shown and a comparison to the direct solution is carried out. The Semi-Analytical Approach will be used as Quick-Look Tool in the official ESA GOCE gravity field processing.

The modelling of the Earth’s gravity field by means of a high-resolving spherical harmonic analysis is a numerically demanding task, especially when realistic (non gridded) data sets are analysed. The free kite numbering scheme, presented in the current article, allows a flexible combination of models. It is focussed, in particular, on the combination of a model containing rotation-symmetrical, high-resolving data with a second model comprising fully correlated data, which allows the determination of the lower degrees. This kite scheme may, depending on the degree of conformance with rotation symmetry, be used both with a direct solver and to improve the convergence rate of an iterative solver.

GOCE (Gravity Field and Steady-State Ocean Circulation Explorer) is a dedicated satellite gravity field mission to be launched in the year 2006. The payload of GOCE will consist of a GPS receiver for a precise orbit determination and for recovering the long and medium spectral part of the gravity field. The high resolution spectral part of the gravity field will be derived by in-orbit gravity gradients in three spatial directions measured by a gravity gradiometer consisting of six three-axis accelerometers. In this article an integrated gravity field recovery procedure is presented that allows to determine a global gravity field solution with high long and medium wavelength accuracy and to improve this global solution in regions with characteristic gravity field features by an adapted regional recovery procedure. If necessary, several regional solutions with global coverage can be merged by means of quadrature methods to obtain an improved global solution. Simulation results are presented to demonstrate this approach. Due to the improved regionally adapted gravity field solutions this technique provides better global gravity field recovery results than calculating a spherical harmonics solution by recovering the potential coefficients directly.

The GOCE mission, planned to be launched in autumn 2006, will allow to determine the static Earth gravity field down to features of 100 km-70 km (half wavelength) in terms of spatial resolution. Since satellite gradiometry is restricted to the medium- to short-wavelength part of the gravitational spectrum, only its combination with satellite-to-satellite measurements in the high-low mode will meet the mission requirements as demanded by the ESA, namely a high-accurate GOCE-only terrestrial gravity field modeling. Here we apply the acceleration approach which is predominantly characterized by numerical differentiation of the kinematic GOCE orbit. Gradiometry is treated by analysis of the fundamental invariants of the gravitational tensor. These quantities neither depend on reference frame rotations nor on the orientation of the gradiometer frame in space. Linearization, computational effort and amalgamation of tensor elements provided with different levels of accuracy make this approach hard to handle. The use of high performance computing facilities, parallel programming standards and optimized numerical libraries are the key to accomplish efficient gravity field recovery.

Since the main goal of the GOCE mission is the derivation of a static gravity field, significant temporal gravity changes from mass redistributions in the System Earth have to be removed from the measurement data in a dealiasing step. Furthermore, a method for gravity field recovery has to be developed, which is capable to process different kinds of data simultaneously. The effects of different mass redistribution systems, like atmosphere, oceans or hydrology, are investigated in terms of geoid and gravity gradients. Main focus is laid on hydrology effects, since global models of the continental water storage turned out to be rather inconsistent, compared to models of the other systems. However, they may benefit from the newly available GRACE gravity field models. It is shown that all time variable gravity effects are small compared with the gradiometer performance; nevertheless it is recommended to use the data from geophysical models and from monthly GRACE gravity field solutions to diminish aliasing effects in the GOCE measurements. In order to simplify the assimilation of gradiometric and satellite-to-satellite-tracking data (e.g. also from GRACE), a method for gravity field recovery has been developed, which is capable to handle the gradiometric data directly in the gradiometer reference frame. It benefits from a filter algorithm based on colored noise for the decorrelation of the gradients and applies powerful parallelization techniques. A high degree gravity field is recovered from simulated SGG data by this approach.

To meet the accuracy requirements of the GOCE mission, the gradiometer has to be calibrated and validated internally as well as externally. An internal quality assessment of the observed GOCE data is possible by comparisons of observations at the same satellite position, i.e. at satellite track cross-overs. Due to the orbit characteristics of the mission, satellite ground track cross-overs have to be used instead of identical repeat positions. Therefore, an appropriate reduction concept has to be applied to consider the differences caused by different satellite altitudes and orientations. It is shown here, that present global gravity field models meet the accuracy and resolution requirements for the reduction concept, and hence for the relative validation of GOCE gradients.For an external calibration or validation based on regional data sets, terrestrial gravity anomalies are upward continued to gravitational gradients at GOCE altitude. The computations are done with synthetic data in a closed-loop simulation. Two upward continuation methods are considered, namely least-squares collocation and integral formulas based on the spectral combination technique. Both methods are described and the results are compared numerically with the ground-truth data. Finally, the results of a regional calibration experiment with simulated noisy GOCE gradients are described.

In the framework of the Geotechnologien project “Integration of space geodetic techniques and development of a user centre for the International Earth rotation and Reference systems Service (IERS)” the IERS Data and Information System1 has been developed at BKG. The system allows the proper maintenance of all data and products of the IERS and provides user-friendly interfaces to browse the data, to search for specific data and to download the data. The system is databasedriven to guarantee the timeliness and consistency of the contents of the information system. Additionally, meta data of all products and publications are modelled in a database to allow the users to search for specific data or topics with respect to space, time and content. In order to be able to link related data the heterogeneous formats of the products are transformed into a common format. The eXtensible Markup Language (XML) will be used, to perform this ambitious task. The usage of XML not only links related data but also supports the exchange of data and the output in different formats like html, pdf, etc. The system is being completed by an administration tool to manage and coordinate the tasks of the Central Bureau and by a general information system with respect to IERS-related topics. The IERS Data and Information System is the basis for the development and implementation of a German contribution to a “Global Geodetic Observing System (GGOS)2” realising a central interface to transfer information between the highly complicated system of measurement and analysis procedures and the users. The standardised database tools will promote an easy exchange of information with and links to other databases within the GGOS project to realise a powerful instrument to serve the Earth observing system.

Until today the products of the International Earth Rotation and Reference Systems Service (IERS), like International Terrestrial Reference Frame (ITRF), International Celestial Reference Frame (ICRF) and Earth Orientation Parameters (EOP), are combined independently, neither intra-technique nor inter-technique combinations, including the full variance-covariance information are performed. To overcome this deficiencies in the present IERS product generation the IERS implemented a new structure in January 2001. This includes the new IERS Combination Research Centres (CRC) and the IERS Analysis Coordinator (AC). He is responsible for the long-term and internal consistency of the IERS products. To achieve the highest accuracy and consistency, it is crucial to proceed towards a fully rigorous combination of all the parameters common to more than one space geodetic technique. Since 2001 the IERS AC initiated and coordinated many different projects and campaigns towards this overall goal. Now the results of the last three years build the theoretical and practical base for the latest project, the Combination Pilot Project (CPP). This project will prepare the generation of a combined IERS product on a routine basis.

The International Terrestrial Reference Frame (ITRF) is realized by epoch positions and linear velocities of a set of geodetic points on the Earth’s surface. Up to the present, i.e. the ITRF2000, the computation is done by a 14 parameter similarity transformation (7 for the stationary and 7 for the kinematic coordinates) of individual solutions from the different space geodetic observations (VLBI, SLR, GPS, DORIS) and the simultaneous adjustment of the position and velocity coordinates. The analysis of the ITRF2000 shows some problems resulting from this transformation procedure. The refined TRF computations done by DGFI use unconstrained normal equations from the solutions of the individual techniques. After a thorough analysis and editing they are combined in a first step internally by accumulation per technique. In the second step the normal equations of the unique techniques are accumulated for an inter-technique combination. The datum of the final TRF solution is attained by no net rotation w.r.t. ITRF2000. Detailed comparisons show a generally good agreement between the DGFI TRF and the ITRF2000. Some outliers are discussed.

The quality and reliability of products of geodetic Very Long Baseline Interferometry (VLBI) observations, as those of any other space geodetic technique, are greatly enhanced if results of different analysis software packages with their individual models are brought together using a suitable combination process. In the framework of the Geotechnologien-project “Integration of space geodetic techniques - IERS”, tools and procedures have been developed for the combination of VLBI results taking into account the peculiarities of the VLBI observing technique. They have found their use in the routine operation of the International VLBI Service for Geodesy and Astrometry (IVS) which produces earth orientation parameters (EOP) and terrestrial reference frame (TRF) realisations from session-wise geodetic VLBI observations. The operational combination itself has several levels of latency requirements. Therefore, different types of combined EOP series based on EOP input series alone or on the full variance/covariance information of the EOP and the underlying TRF are generated.

This paper focusses on some relevant aspects when proceeding towards a rigourous combination of space geodetic observations that are addressed by the IERS Combination Research Centers at DGFI and FESG in the frame of the IERSGEOTECHNOLOGIEN project. These are in particular investigations related to the datum realization for the space geodetic solutions, the analysis and combination of SLR and VLBI data, and research activities within the IERS SINEX Combination Campaign and the IERS Combination Pilot Project.

This report is divided into two parts: the first part gives an overview of the combination studies performed by the Forschungseinrichtung Satellitengeodäsie TU München (FESG) and the Deutsches Geodätisches Forschungsinstitut (DGFI) based on the data of the continuous IVS campaign CONT02. The close cooperation of the two institutions established the basis for a detailed adaption of the GPS and the VLBI software concerning models and parameterization to avoid systematic differences between the technique contributions. Special attention was payed to parameters with a high temporal resolution; in this study tropospheric parameters and Earth rotation parameters (ERP) are considered. Including the troposphere parameters offered a good possibility to study the correlation between troposphere parameters and station coordinates. It was found that this interaction can deliver a very important contribution to validate the available local tie information. For comparison of the troposphere results derived for the 14-days campaign CONT02, long time series for VLBI and GPS were used as well, and it turned out that the results are in good agreement. Regarding the sub-daily Earth rotation parameters it can be shown that a combination of the space techniques improves the results compared to single-technique solutions. Furthermore, it is illustrated that UT1-UTC can be combined from VLBI together with the satellite techniques. All in all, the presented results demonstrate the high potential of a combination of VLBI, GPS and SLR data. The second part is devoted to the combination of long sub-daily EOP time series from VLBI and GPS. Space geodetic techniques like the Global Positioning System (GPS) and Very Long Baseline Interferometry (VLBI) can provide Earth Orientation Parameter (EOP) time series with very high sampling rates. This offers the opportunity to study sub-daily tidal excitations and the influences of high-frequency or episodic geophysical effects on Earth rotation. Therefore we need sub-daily time series as consistent and homogeneous as possible. Based on the Combined Smoothing method of Vondrak and Cepek (2000), we developed a new combination scheme for sub-daily EOPs to obtain a new sub-daily time series which benefits from the longterm stability of VLBI and the continuity of GPS. Furthermore we can remove the weakness of UT1 estimations of the satellite techniques. We analysed the combination using spectral imaging methods and the results from sub-daily tidal harmonic estimation.

The combination of different space geodetic or satellite techniques offers the possibility to determine more complete, reliable, and accurate reference frame parameters. Two new approaches are proposed: the onboard collocation and the integrated method. The onboard collocation utilizes the availability of different techniques onboard a Low Earth Orbiting (LEO) satellite. For the CHAMP and GRACE satellites, and for the GPS-35 and -36 satellites, the radial distances between the Global Positioning System (GPS) phase centers and the Satellite Laser Ranging (SLR) retroreflectors are examined. The integrated method treats a multi-satellite configuration with measurements by various tracking techniques on the observation level. From three months of observations from the GRACE/GPS constellation, a series of low degree harmonics of the Earth’s gravity field, representing dynamical geocenter location and axes orientation, is determined with high accuracy and daily resolution.

Over the last 40 years ring laser gyroscopes became one of the most important instruments in the field of inertial navigation and precise rotation measurements. They have a high resolution for angular velocities, a very good scale factor stability and a wide dynamic range. These properties made them suitable for aircraft and autonomous submarine navigation. Over the last decade we have developed several very large perimeter ring laser gyroscopes for the application in geodesy and geophysics (Schreiber et al., 2001). Because of a substantial upscaling of these ring lasers, their sensitivity to rotations has been increased by at least 5 orders of magnitudes. At the same time the instrumental drift was reduced by about the same amount. This progress in rotational sensor technology led to the successful detection of rotational signals caused by earthquakes (Pancha et al., 2000) several thousands kilometers away. These observations stimulated the development of a highly sensitive ring laser gyro for specific seismological applications. The GEOsensor provides rotational motions along with the usual translational motions at a high data acquisition rate of at least 20 Hz. Observations of seismic induced rotations show that they are consistent in phase and amplitude with the collocated recordings of transverse accelerations obtained from a standard seismometer over a wide range of distances and frequencies.

Airborne gravimetry systems provide the most economical way to improve the spatial resolution of gravity data measured by satellite missions. So the paper deals with the presentation of a modern airborne gravitymeter designed, developed and tested at the University FAF Munich. The specific forces are measured by a high precision strapdown INS and the kinematical accelerations are derived using numerous differential GNSS observations.The first part of the paper describes the system architecture and the aircraft installation. Then the data processing methods are mentioned including the filtering and derivation procedures, the computation of aircraft accelerations and different algorithms for providing the gravity profiles. Static tests in a laboratory environment approve the error budget of two sensor types on acceleration level.The main part of the article contains the description of practical test flights carried out in the middle of Germany and the corresponding results. An evaluation of the current system performance is possible. Final remarks refer to the planned improvements until the end of the project.

A strapdown airborne gravimeter of a peculiar configuration has been developed and is nearly operational for observing total acceleration. Precision high sampling rate GPS receivers provide the kinematic acceleration. The overall system and its hardware is analysed with focus on signal flow.

The Institute of Flight Guidance (IFF) of the Technical University of Braunschweig (TU BS) is involved in the development of airborne gravimetry since 1985. Fundamental examinations of airborne gravimeters were carried out between 1991 and 1993. In 1998 a high-precision two-frame inertial platform and a gravimeter sensor were purchased and modified for airborne application in cooperation with the Russian manufacturer Elektropribor.Successful flight tests have been executed in the recent years. So far the resolution achieved is 2 km with a standard deviation of 3mGal1 (resp. 5km, 1mGal). The two-frame inertial platform was extended to a three-frame INS by mounting a ring laser gyro on top of the platform. The gyro provides an additional degree of freedom (yaw) around the vertical axis.Since 2001 the IFF was involved in the programme GEOTECHNOLOGIEN funded by the German Federal Ministry of Education and Research. The goal of the project, which ended in 2005 was to develop an airborne gravimetry system with a resolution of 1mGal for wavelengths of 1 km.