Error Characteristics Estimated from Champ,GRACE and GOCE Derived Geoids and from Satellite Altimetry Derived Mean Dynamic Topography

This paper presents a review of geoid error characteristics of three satellite gravity missions in view of the general problem of separating scientifically interesting signals from various noise sources. The problem is reviewed from the point of view of two proposed applications of gravity missions, one is the observation of the mean oceanic circulation whereby an improved geoid model is used as a reference surface against the long term mean sea level observed by altimetry. In this case we consider the presence of mesoscale variability during assimilation of derived surface currents in inverse models. The other experiment deals with temporal changes in the gravity field observed by GRACE in which case a proposed experiment is to monitor changes in the geoid in order to detect geophysical interesting signals such as variations in the continental hydrology and non-steric ocean processes. For this experiment we will address the problem of geophysical signal contamination and the way it potentially affects monthly geoid solutions of GRACE. This revised version was published online in August 2006 with corrections to the Cover Date.     

How Operational Oceanography Can Benefit from Dynamic Topography Estimates as Derived from Altimetry and Improved Geoid

With a precise geoid, GOCE will allow an estimation of absolute dynamic topography from altimetry. The projected benefits to operational oceanography and its applications are analyzed herein. After a brief overview of operational oceanography, we explain how the new geoids will be used in the future to improve real time altimeter products and to better constrain modelling and data assimilation systems. A significant impact is expected both for mesoscale (e.g. better estimations and forecasts of currents for pollution monitoring, marine safety, offshore industry) and climate (better initialization of coupled ocean/atmosphere models) applications. This revised version was published online in August 2006 with corrections to the Cover Date.     

Combined Use of Altimetry and In Situ Gravity Data for Coastal Dynamics Studies

Accurate local geoids derived from in situ gravity data will be valuable in the validation of GOCE results. In addition it will be a challenge to use GOCE data in an optimal way, in combination with in situ gravity, to produce better local geoid solutions. This paper discusses the derivation of a new geoid over the NW European shelf, and its comparison with both tide gauge and altimetric sea level data, and with data from ocean models. It is hoped that over the next few years local geoid methods such as these can be extended to cover larger areas and to incorporate both in situ and satellite measured gravity data. This revised version was published online in August 2006 with corrections to the Cover Date.     

Impact of Geoid Improvement on Ocean Mass and Heat Transport Estimates

One long-standing difficulty in estimating the large-scale ocean circulation is the inability to observe absolute current velocities. Both conventional hydrographic measurements and altimetric measurements provide observations of currents relative to an unknown velocity at a reference depth in the case of hydrographic data, and relative to mean currents calculated over some averaging period in the case of altimetric data. Space gravity missions together with altimetric observations have the potential to overcome this difficulty by providing absolute estimates of the velocity of surface oceanic currents. The absolute surface velocity estimates will in turn provide the reference level velocities that are necessary to compute absolute velocities at any depth level from hydrographic data. Several studies have been carried out to quantify the improvements expected from ongoing and future space gravity missions. The results of these studies in terms of volume flux estimates (transport of water masses) and heat flux estimates (transport of heat by the ocean) are reviewed in this paper. The studies are based on ocean inverse modeling techniques that derive impact estimates solely from the geoid error budgets of forthcoming space gravity missions. Despite some differences in the assumptions made, the inverse modeling calculations all point to significant improvements in estimates of oceanic fluxes. These improvements, measured in terms of reductions of uncertainties, are expected to be as large as a factor of 2. New developments in autonomous ocean observing systems will complement the developments expected from space gravity missions. The synergies of in situ and satellite observing systems are considered in the conclusion of this paper. This revised version was published online in August 2006 with corrections to the Cover Date.     

V: SEA LEVEL: Benefits of GRACE and GOCE to sea level studies

The recently published Third Assessment Reports of the Intergovernmental Panel on Climate Change have underlined the scientific interest in, and practical importance of past and potential future sea level changes. Space gravity missions will provide major benefits to the understanding of the past, and, thereby, in the prediction of future, sea level changes in many ways. The proposal for the GOCE mission described well the improvements to be expected from improved gravity field and geoid models in oceanography (for example, in the measurement of the time-averaged, or ‘steady state’, ocean surface circulation and better estimation of ocean transports), in geophysics (in the improvement of geodynamic models for vertical land movements), in geodesy (in positioning of tide gauge data into the same reference frame as altimeter data, and in improvement of altimeter satellite orbits), and possibly in glaciology (in improved knowledge of bedrock topography and ice sheet mass fluxes). GRACE will make many important steps towards these ‘steady state’ aims. However, its main purpose is the provision of oceanographic (and hydrological and meteorological) temporally-varying gravity information, and should in effect function as a global ‘bottom pressure recorder’, providing further insight into the 3-D temporal variation of the ocean circulation, and of the global water budget in general. This paper summaries several of these issues, pointing the way towards improved accuracy of prediction of future sea level change. This revised version was published online in August 2006 with corrections to the Cover Date.     

Tidal Models in a New Era of Satellite Gravimetry

The high precision gravity measurements to be made by recently launched (and recently approved) satellites place new demands on models of Earth, atmospheric, and oceanic tides. The latter is the most problematic. The ocean tides induce variations in the Earth's geoid by amounts that far exceed the new satellite sensitivities, and tidal models must be used to correct for this. Two methods are used here to determine the standard errors in current ocean tide models. At long wavelengths these errors exceed the sensitivity of the GRACE mission. Tidal errors will not prevent the new satellite missions from improving our knowledge of the geopotential by orders of magnitude, but the errors may well contaminate GRACE estimates of temporal variations in gravity. Solar tides are especially problematic because of their long alias periods. The satellite data may be used to improve tidal models once a sufficiently long time series is obtained. Improvements in the long-wavelength components of lunar tides are especially promising. This revised version was published online in August 2006 with corrections to the Cover Date.     

III: OCEAN CIRCULATION: Global Ocean Data Assimilation and Geoid Measurements

Parts of geodesy and physical oceanography are about to mature into a single modeling problem involving the simultaneous estimation of the marine geoid and the general circulation. Both fields will benefit. To this end, we present an ocean state estimation (data assimilation) framework which is designed to obtain a dynamically consistent picture of the changing ocean circulation by combining global ocean data sets of arbitrary type with a general circulation model (GCM). The impact of geoid measurements on such estimates of the ocean circulation are numerous. For the mean circulation, a precise geoid describes the reference frame for dynamical signals in altimetric sea surface height observations. For the time-varying ocean signal, changing geoid information might be a valuable new information about correcting the changing flow field on time scales from a few month to a year, but the quantitative utility of such information has not yet been demonstrated. For a consistent estimate, some knowledge of the prior error covariances of all data fields is required. The final result must be consistent with prior error estimates for the data. State estimation is thus one of the few quantitative consistency checks for new geoid measurements anticipated from forthcoming space missions. Practical quantitative methods will yield a best possible estimate of the dynamical sea surface which, when combined with satellite altimetric surfaces, will produce a best-estimate marine geoid. The anticipated accuracy and precision of such estimates raises some novel modeling error issues which have not conventionally been of concern (the Boussinesq approximation, self-attraction and loading). Model skill at very high frequencies is a major concern because of the need to de-alias the data obtained by the inevitable oceanic temporal undersampling dictated by realistic satellite orbit configurations. This revised version was published online in August 2006 with corrections to the Cover Date.     

Geodetic Methods for Calibration of GRACE and GOCE

It is beyond doubt that calibration and validation are essential tools in the process of reaching the goals of gravity missions like GRACE and GOCE and to obtain results of the highest possible quality. Both tools, although general and obvious instruments for any mission, have specific features for gravity missions. Therefore, it is necessary to define exactly what is expected (and what cannot be expected) from calibration and what from validation and how these tools should work in our case. The general calibration and validation schemes for GRACE and GOCE are outlined. Calibration will be linked directly to the instrument and the measurements whereas validation will be linked to data derived from the original measurements. Calibration includes on-ground, internal, and external calibration as well as error assessment. The calibration phase results in corrected measurements along with an a posteriori error model. Validation of e.g. calibrated measurements or geoid heights means checking against independent data to assess whether there are no systematic errors left and/or whether the error model describes the true error reasonably well. Geodetic methods for calibration typically refer to external calibration and error assessment, and will be illustrated with an example. This revised version was published online in August 2006 with corrections to the Cover Date.     

Measuring the Distribution of Ocean Mass Using GRACE

The Gravity Recovery and Climate Experiment (GRACE), which was successfully launched March 17, 2002, has the potential to create a new paradigm in satellite oceanography with an impact perhaps as large as was observed with the arrival of precision satellite altimetry via TOPEX/Poseidon (T/P) in 1992. The simulations presented here suggest that GRACE will be able to monitor non-secular changes in ocean mass on a global basis with a spatial resolution of ≈500 km and an accuracy of ≈3 mm water equivalent. It should be possible to recover global mean ocean mass variations to an accuracy of ≈1 mm, possibly much better if the atmospheric pressure modeling errors can be reduced. We have not considered the possibly significant errors that may arise due to temporal aliasing and secular gravity variations. Secular signals from glacial isostatic adjustment and the melting of polar ice mass are expected to be quite large, and will complicate the recovery of secular ocean mass variations. Nevertheless, GRACE will provide unprecedented insight into the mass components of sea level change, especially when combined with coincident satellite altimeter measurements. Progress on these issues would provide new insight into the response of sea level to climate change. This revised version was published online in August 2006 with corrections to the Cover Date.     

Estimating the High-Resolution Mean Sea-Surface Velocity Field by Combined Use of Altimeter and Drifter Data for Geoid Model Improvement

The mean sea-surface height obtained from satellite altimeters is different from the geoid by the amount of mean sea-surface dynamic topography associated with ocean currents. Assuming geostrophy at the sea surface, the mean sea-surface dynamic topography can be obtained from the mean sea-surface velocity field. This field is derived by combining anomalies (i.e., deviations from the mean) of sea-surface velocity obtained from altimeter data and in situ surface velocities estimated from trajectories of surface drifting-buoys (hereafter, drifters). Where a drifter measured the surface velocity, the temporal mean velocity can be estimated by subtracting the altimeter-derived velocity anomaly at that time from the drifter-measured surface velocity. The method is applied to the surface flow field of the North Pacific, using TOPEX/POSEIDON and ERS-1/2 altimeter data, and WOCE-TOGA surface drifter data obtained from October 1992 through December 2000. The temporal mean velocity field is estimated with a resolution of quarter degrees in both latitude and longitude. The obtained mean velocity field clearly shows the Kuroshio and Kuroshio Extension, which are narrower and stronger than the climatological mean features derived from historical hydrographic data averaged over several decades. Instantaneous velocities are estimated by summing up these temporal mean velocities and anomalies, every ten days during the eight years. They compare well with in situ velocities measured by the surface drifters. The instantaneous velocity field shows energetic fluctuation of the Kuroshio Extension vividly. This revised version was published online in August 2006 with corrections to the Cover Date.     

A geopause satellite system concept

The forthcoming 10 cm range tracking accuracy capability holds much promise in connection with a number of Earth and ocean dynamics investigations. These include a set of earthquake-related studies of fault motions and the Earth's tidal, polar and rotational motions, as well as studies of the gravity field and the sea surface topography which should furnish basic information about mass and heat flow in the oceans. The state of the orbit analysis art is presently at about the 10 m level, or about two orders of magnitude away from the 10 cm range accuracy capability expected in the next couple of years or so. The realization of a 10 cm orbit analysis capability awaits the solution of four kinds of problems, namely, those involving orbit determination and the lack of sufficient knowledge of tracking system biases, the gravity field, and tracking station locations. The Geopause satellite system concept offers promising approaches in connection with all of these areas. A typical Geopause satellite orbit has a 14 hour period, a mean height of about 4.6 Earth radii, and is nearly circular, polar, and normal to the ecliptic. At this height only a relatively few gravity terms have uncertainties corresponding to orbital perturbations above the decimeter level. The orbit s, in this sense, at the geopotential boundary, i.e., the geopause. The few remaining environmental quantities which may be significant can be determined by means of orbit analyses and accelerometers. The Geopause satellite system also provides the tracking geometery and coverage needed for determining the orbit, the tracking system biases and the station locations. Studies indicate that the Geopause satellite, tracked with a 2 cm ranging system from nine NASA affiliated sites, can yield decimeter station location accuracies. Five or more fundamental stations well distributed in longitude can view Geopause over the North Pole. This means not only that redundant data are available for determining tracking system biases, but also that both components of the polar motion can be observed frequently. When tracking Geopause, the NASA sites become a two-hemisphere configuration which is ideal for a number of Earth physics applications such as the observation of the polar motion with a time resolution of a fraction of a day. Geopause also provides the basic capability for satellite-to-satellite tracking of drag-free satellites for mapping the gravity field and altimeter satellites for surveying the sea surface topography. Geopause tracking a coplanar, drag-free satellite for two months to 0.03 mm per second accuracy can yield the geoid over the entire Earth to decimeter accuracy with 2.5° spatial resolution. Two Geopause satellites tracking a coplanar altimeter satellite can then yield ocean surface heights above the geoid with 7° spatial resolution every two weeks. These data will furnish basic boundary condition information about mass and heat flows in the oceans which are important in shaping weather and climate.     

IV: GEODESY: Remarks on the Role of Height Datum in Altimetry-Gravimetry Boundary-Value Problems

The problem of global geoid determination is usually solved using satellite altimetry data on the oceans, together with an oceanographic model of sea surface topography, and gravity anomaly data on the continents. Such data, however, enable to obtain only potential differences with respect to a reference surface whose absolute potential is unknown. This situation suggests to modify the classical mixed boundary-value problem of physical geodesy by inserting into the boundary conditions an unknown additive constant, that must be determined by imposing a suitable additional constraint. Yet, such formulation of the boundary-value problem, from the point of view of its mathematical properties, is not unconditionally well-posed, and, furthermore, does not reflect faithfully the available physical model, as the present knowledge of ocean circulation does not allow to connect along coastlines the reference surfaces defined on the oceans and on the continents. The introduction of two different unknown additive constants, one for the oceans and one for the earth, to be determined by imposing two additional constraints, gives rise to a more faithful picture of the present physical knowledge, and, at the same time, to a new well-posed formulation of the boundary-value problem. This revised version was published online in August 2006 with corrections to the Cover Date.     

Aiming at a 1-cm Orbit for Low Earth Orbiters: Reduced-Dynamic and Kinematic Precise Orbit Determination

The computation of high-accuracy orbits is a prerequisite for the success of Low Earth Orbiter (LEO) missions such as CHAMP, GRACE and GOCE. The mission objectives of these satellites cannot be reached without computing orbits with an accuracy at the few cm level. Such a level of accuracy might be achieved with the techniques of reduced-dynamic and kinematic precise orbit determination (POD) assuming continuous Satellite-to-Satellite Tracking (SST) by the Global Positioning System (GPS). Both techniques have reached a high level of maturity and have been successfully applied to missions in the past, for example to TOPEX/POSEIDON (T/P), leading to (sub-)decimeter orbit accuracy. New LEO gravity missions are (to be) equipped with advanced GPS receivers promising to provide very high quality SST observations thereby opening the possibility for computing cm-level accuracy orbits. The computation of orbits at this accuracy level does not only require high-quality GPS receivers, but also advanced and demanding observation preprocessing and correction algorithms. Moreover, sophisticated parameter estimation schemes need to be adapted and extended to allow the computation of such orbits. Finally, reliable methods need to be employed for assessing the orbit quality and providing feedback to the different processing steps in the orbit computation process. This revised version was published online in August 2006 with corrections to the Cover Date.     

What Might GRACE Contribute to Studies of Post Glacial Rebound?

The NASA/DLR satellite gravity mission GRACE, launched in March, 2002, will map the Earth's gravity field at scales of a few hundred km and greater, every 30 days for five years. These data can be used to solve for time-variations in the gravity field with unprecedented accuracy and resolution. One of the many scientific problems that can be addressed with these time-variable gravity estimates, is post glacial rebound (PGR): the viscous adjustment of the solid Earth in response to the deglaciation of the Earth's surface following the last ice age. In this paper we examine the expected sensitivity of the GRACE measurements to the PGR signal, and explore the accuracy with which the PGR signal can be separated from other secular gravity signals. We do this by constructing synthetic GRACE data that include contributions from a PGR model as well as from a number of other geophysical processes, and then looking to see how well the PGR model can be recovered from those synthetic data. We conclude that the availability of GRACE data should result in improved estimates of the Earth's viscosity profile. This revised version was published online in August 2006 with corrections to the Cover Date.     

How to Climb the Gravity Wall

Space Science Reviews - What type of gravity satellite mission is required for the time after GRACE and GOCE? Essentially, the variables at our disposal are experiment altitude, compensation of...     

Ocean Tides in GRACE Monthly Averaged Gravity Fields

The GRACE mission will map the Earth's gravity fields and its variations with unprecedented accuracy during its 5-year lifetime. Unless ocean tide signals and their load upon the solid earth are removed from the GRACE data, their long period aliases obscure more subtle climate signals which GRACE aims at. In this analysis the results of Knudsen and Andersen (2002) have been verified using actual post-launch orbit parameter of the GRACE mission. The current ocean tide models are not accurate enough to correct GRACE data at harmonic degrees lower than 47. The accumulated tidal errors may affect the GRACE data up to harmonic degree 60. A study of the revised alias frequencies confirm that the ocean tide errors will not cancel in the GRACE monthly averaged temporal gravity fields. The S₂ and the K₂ terms have alias frequencies much longer than 30 days, so they remain almost unreduced in the monthly averages. Those results have been verified using a simulated 30 days GRACE orbit. The results show that the magnitudes of the monthly averaged values are slightly higher than the previous values. This may be caused by insufficient sampling to fully resolve and reduce the tidal signals at short wavelengths and close to the poles. This revised version was published online in August 2006 with corrections to the Cover Date.     

VII: CLOSING SESSION: GOCE: ESA's First Earth Explorer Core Mission

This paper introduces the first ESA Core Earth Explorer mission, GOCE, in the context of ESA's Living Planet programme. GOCE will measure highly accurate, high spatial resolution differential accelerations in three dimensions along a well characterised orbit: the mission is planned for launch in early 2006. The mission objectives are to obtain gravity gradient data such that new global and regional models of the static Earth's gravity field and of the geoid can be deduced at length scales down to 100 km. These products will have broad application in the fields of geodesy, oceanography, solid-earth physics and glaciology. This revised version was published online in August 2006 with corrections to the Cover Date.     

Space-Wise,Time-Wise,Torus and Rosborough Representations in Gravity Field Modelling

The decade of the geopotentials started July 2000 with the launch of the German high-low SST mission CHAMP. Together with the joint NASA-DLR low-low SST mission GRACE and the ESA gradiometry mission GOCE an unprecedented wealth of geopotential data becomes available over the next few years. Due to the sheer number of unknown gravity field parameters (up to 100 000) and of observations (millions), especially the latter two missions are highly demanding in terms of computational requirements. In this paper several modelling strategies are presented that are based on a semi-analytical approach. In this approach the set of normal equations becomes block-diagonal with maximum block-sizes smaller than the spherical harmonic degree of resolution. The block-diagonality leads to a rapid and powerful gravity field analysis tool. Beyond the more-or-less conventional space-wise and time-wise formulations, the torus approach and Rosborough's representation are discussed. A trade-off between pros and cons of each of the modelling strategies will be given. This revised version was published online in August 2006 with corrections to the Cover Date.     

The Elusive Stationary Geoid

We discuss the various problems occurring when trying to fix a geoid or geopotential model using sea level observations sampled during a limited time span from a bounded geographical domain. Such problems are on the one hand aliasing and spectral leakage, and on the other, the non-conservation of matter over only part of the world ocean. In the light of these issues we discuss whether it is sensible to include in a definition of the global geoid the radially symmetric part of either the mean sea level field itself, or its linear or nonlinear time dependence, arriving at a negative conclusion. This revised version was published online in August 2006 with corrections to the Cover Date.     

Autonomous orbit determination using epoch-differenced gravity gradients and starlight refraction

Autonomous orbit determination via integration of epoch-differenced gravity gradients and starlight refraction is proposed in this paper for low-Earth-orbiting satellites operating in GPS-denied environments.Starlight refraction compensates for the significant along-track position error that occurs from only using gravity gradients and benefits from integration in terms of improved accuracy in radial and cross-track position estimates.The between-epoch differencing of gravity gradients is employed to eliminate slowly varying measurement biases and noise near the orbit revolution frequency.The refraction angle measurements are directly used and its Jacobian matrix derived from an implicit observation equation.An information fusion filter based on a sequential extended Kalman filter is developed for the orbit determination.Truth-model simulations are used to test the performance of the algorithm,and the effects of differencing intervals and orbital heights are analyzed.A semi-simulation study using actual gravity gradient data from the Gravity field and steady-state Ocean Circulation Explorer (GOCE) combined with simulated starlight refraction measurements is further conducted,and a three-dimensional position accuracy of better than 100 m is achieved.