Fact-checked by Grok 2 weeks ago

Earth Gravitational Model

The Earth Gravitational Model (EGM) is a series of high-resolution mathematical models representing the external gravitational potential of the Earth, expressed as sets of spherical harmonic coefficients expanded to high degrees and orders. These models approximate the irregular Earth's gravity field—deviating from a perfect oblate spheroid due to mass distribution variations—and enable the computation of key geodetic quantities such as geoid undulations (the height of the equipotential surface coinciding with mean sea level above the reference ellipsoid), free-air gravity anomalies, and deflections of the vertical. Jointly developed by the U.S. National Geospatial-Intelligence Agency (NGA) and NASA, EGMs form the foundation for the World Geodetic System 1984 (WGS 84) gravity component, supporting precise applications in satellite navigation, orbit determination, geophysical prospecting, ocean circulation modeling, and global height datum unification.^[1]^[2] The evolution of EGMs reflects advances in satellite technology, gravimetry, and computational methods, beginning with low-degree models in the 1950s and progressing to high-fidelity representations today. Early efforts, such as NASA's Goddard Earth Model (GEM) series in the 1970s and 1980s (e.g., GEM-T1 to degree 36 and GEM-T2 to degree 180), relied primarily on satellite laser ranging and Doppler tracking data to capture long-wavelength features of the gravity field. A milestone was the EGM96 model, released in 1996, which extended to degree and order 360 by integrating over 10 million observations from satellite altimetry (e.g., TOPEX/Poseidon, ERS-1, GEOSAT), surface gravity measurements (including more than 30 million point anomalies from NIMA archives), and satellite tracking (SLR, GPS, DORIS). This model achieved a global root-mean-square (RMS) geoid error of approximately 18 cm over oceans and 28 cm over land, with gravity anomaly accuracy around 29 mGal RMS, marking a tripling of resolution over predecessors like JGM-3.^[3] The current operational standard, EGM2008, released in April 2008, dramatically enhanced spatial resolution and accuracy by reaching degree and order 2159 (complete, with select coefficients to 2190), equivalent to about 5 arc-minute grid spacing or roughly 9.2 km at the equator. It combines a low-degree satellite-only component from the GRACE mission (ITG-GRACE03S to degree 180, based on data from 2002–2007) with a high-degree synthesis from a global 5 arc-minute gravity anomaly grid derived from terrestrial, airborne, and altimetry sources (e.g., over 97% terrestrial coverage at 30 arc-minute resolution, supplemented by fill-in data). Employing least-squares collocation and block-diagonal techniques, EGM2008 yields geoid undulations accurate to 5–10 cm RMS in densely surveyed regions (e.g., North America, Europe) and 11–15 cm globally, with vertical deflections at 1.1–1.3 arc-seconds—improvements by factors of 3–6 over EGM96 in accuracy and 6-fold in resolution.^[2] This model also includes specialized products like the Dynamic Ocean Topography 2008 (DOT2008A) for marine applications and error covariance matrices up to degree 2190 for uncertainty propagation.^[4] EGMs are indispensable for modern geospatial infrastructure, underpinning GPS-derived orthometric heights (via GPS/leveling), inertial navigation systems in aviation and defense, and studies of Earth's interior mass anomalies (e.g., detecting subduction zones or post-glacial rebound). Validation against independent datasets, such as GPS/leveling networks and satellite orbits, confirms their reliability, though limitations persist in data-sparse areas like polar regions or deep oceans, where errors can exceed 20 cm for the geoid. As of 2025, EGM2008 remains the operational standard, with ongoing efforts by NGA and international partners incorporating data from missions like GOCE, GRACE-FO, and planned future gravity satellites to refine these models; NGA tentatively plans to release a new EGM in 2028.^[2]^[1]^[5]

Fundamentals

Definition and Purpose

The Earth Gravitational Model (EGM) is a mathematical representation of the Earth's gravitational field, typically expressed as a spherical harmonic expansion of the geopotential to account for the planet's oblateness, mass anomalies due to density variations, and other irregularities in the gravity field.^[6]^[7] This model describes the geopotential, a scalar function that represents the potential energy per unit mass in the conservative gravitational field, distinguishing it from gravity, which refers to the local acceleration vector derived from the gradient of this potential combined with centrifugal effects.^[8]^[9] The geopotential's equipotential surfaces play a central role, with the geoid defined as the particular equipotential surface that coincides with mean sea level in the absence of currents, tides, or other disturbances, serving as the reference for zero height in geodesy.^[6]^[8] The primary purpose of EGMs is to enable accurate global predictions of gravitational acceleration and geopotential values, which are essential for determining heights above the geoid and performing precise calculations of satellite orbits and trajectories.^[7]^[2] By providing a standardized reference for the non-uniform gravity field, these models support applications in geodetic surveying, where they facilitate the conversion between ellipsoidal and orthometric heights, and in aerospace engineering, where they ensure reliable navigation and mission planning.^[9] For instance, models like EGM2008 integrate satellite, altimetry, and terrestrial gravity data to achieve global coverage with high fidelity.^[2] Historically, the development of EGMs was motivated by the limitations of simpler ellipsoidal approximations, which failed to capture the full complexity of Earth's gravity field as revealed by early satellite observations, such as those from Sputnik in 1957, that showed unexpected orbital perturbations due to mass distributions and density variations.^[7]^[6] These observations underscored the need for more sophisticated models to reflect deviations from rotational equilibrium and subsurface heterogeneities, paving the way for gravity field representations that incorporate data from missions like GRACE and GOCE to refine understanding of Earth's interior structure.^[9]^[6]

Mathematical Representation

The Earth's gravitational potential in the Earth Gravitational Model (EGM) is mathematically represented using a spherical harmonic expansion, which provides a global description of the geoid and gravity anomalies. This formulation expresses the external gravitational potential V at a point with geocentric radius r, colatitude \theta, and longitude \lambda as a series of harmonic terms that capture deviations from a reference ellipsoid.^[2] The core equation for the geopotential is:

V(r, \theta, \lambda) = \frac{GM}{r} \sum_{n=0}^{n_{\max}} \sum_{m=0}^{n} \left( \frac{a}{r} \right)^n \overline{P}_{nm}(\sin \theta) \left( \overline{C}_{nm} \cos m\lambda + \overline{S}_{nm} \sin m\lambda \right)

where GM is the geocentric gravitational constant (product of the gravitational constant G and Earth's mass M), a is the equatorial radius of the reference ellipsoid (typically 6,378,137 m for WGS84), \overline{P}_{nm} are the fully normalized associated Legendre functions of degree n and order m, and \overline{C}_{nm} and \overline{S}_{nm} are the fully normalized spherical harmonic coefficients for the cosine and sine terms, respectively.^[2]^[10] In this expansion, the degree n represents the highest order of multipole moments included, determining the model's resolution, while the order m (ranging from 0 to n) describes the azimuthal variations; the sum typically starts from n=2 to exclude the monopole term already captured by GM/r. The coefficients \overline{C}_{nm} and \overline{S}_{nm} are derived from satellite and terrestrial gravity measurements and are fully normalized to ensure orthogonality and consistent energy partitioning across harmonics, with the normalization factor involving double factorials for global consistency. For example, in the EGM2008 model, coefficients are provided up to degree n=2190 and order m=2159.^[2]^[10] The model is defined in an Earth-fixed reference frame, aligning with the conventional terrestrial system for geodetic applications, though transformations to inertial frames are possible for orbital dynamics. Options for tidal effects are incorporated, such as tide-free (removing all tidal deformations) or zero-tide (removing variable tides but retaining permanent tide) representations, to account for ocean and solid Earth tidal influences on the potential; EGM2008 provides coefficients in both systems.^[2]^[10] The maximum degree and order dictate the spatial resolution, as higher n resolves shorter-wavelength gravity features; for n=2190 in EGM2008, this corresponds to a grid spacing of approximately 5 arcminutes, enabling representation of features down to about 10 km half-wavelength on Earth's surface.^[2]

Historical Development

Early Gravity Models

Prior to the advent of satellite observations, Earth gravity models relied exclusively on terrestrial gravimetric measurements, which were sparse and predominantly confined to land areas, leading to assumptions of ellipsoidal symmetry for the planet's figure. The foundational International Gravity Formula (IGF), developed by Carlo Somigliana in 1929 and formally adopted by the International Union of Geodesy and Geophysics in 1930, provided a closed-form expression for normal gravity on the International Reference Ellipsoid. This formula approximated gravity as a function of latitude, incorporating the effects of Earth's rotation and oblateness, but it could not account for local anomalies or the full geopotential variations due to the limited global coverage of surface data.^[11]^[12] The launch of early satellites marked the beginning of space-based gravity modeling, with data from the Vanguard 1 satellite in 1958 enabling the first determinations of Earth's gravitational harmonics beyond simple ellipsoidal approximations. Analysis of Vanguard 1's orbital perturbations, combined with observations from Sputnik and Explorer satellites, yielded spherical harmonic expansions of the geopotential up to degree and order 8, revealing significant zonal asymmetries such as the J2 oblateness term. These initial models improved upon terrestrial efforts by providing global insights, yet they remained constrained by the low orbital inclinations and short data arcs available from the few early satellites.^[7]^[13] In the 1970s, the Goddard Earth Model (GEM) series advanced this work through integrated satellite tracking and surface gravimetry, with GEM 10 (developed around 1979) achieving a complete expansion to degree and order 22, supplemented by selected terms up to degree 30 for a total of 592 coefficients. This model incorporated data from missions like GEOS-3, enhancing resolution for mid-latitude features, but higher-degree terms were still limited by data sparsity and computational constraints. Key limitations of these early models included low spatial resolution—often equivalent to wavelengths greater than 1,000 km—due to insufficient global coverage, particularly over oceans, and an inability to resolve short-wavelength gravity anomalies without dense terrestrial measurements.^[14]^[15] The transition to more sophisticated representations was driven by improved tracking technologies, exemplified by the 1976 launch of the Laser Geodynamics Satellite (LAGEOS), which provided high-precision laser ranging data to constrain low-degree harmonics and enable subsequent higher-degree models. LAGEOS's passive reflector design facilitated long-term orbital stability, allowing detection of subtle perturbations that informed refinements in geopotential modeling and paved the way for combined satellite-terrestrial solutions.^[16]^[17]

EGM Series Evolution

The Earth Gravitational Model series, developed under the auspices of the U.S. National Geospatial-Intelligence Agency (NGA) and its predecessors, represents a standardized lineage of global geopotential models that have progressively enhanced the representation of Earth's gravity field since the mid-1980s. The inaugural model in this series, EGM84, was released in 1984 by the Defense Mapping Agency (DMA, later NIMA and now NGA), providing normalized spherical harmonic coefficients complete to degree and order 180. It combined satellite tracking data from missions such as Landsat and Seasat with surface gravimetry measurements, achieving gravity anomaly errors of approximately 20-30 mGal in many regions, though with notable limitations in data-sparse areas like oceans and polar zones.^[1]^[18] Building on EGM84's foundation, EGM96 marked a significant update released in 1996 (with full documentation in 1998) through collaboration between NASA, NIMA, and Ohio State University. This model extended to degree and order 360, incorporating block-mean gravity anomalies resolved to 30 arc-minutes and integrating altimetry data from the TOPEX/POSEIDON mission alongside enhanced surface gravity datasets and satellite tracking observations. The development employed least-squares collocation and block-diagonal techniques for coefficient estimation, resulting in improved global geoid accuracies better than 1 meter in well-observed areas, with root-mean-square gravity anomaly fits around 5-12 mGal depending on resolution.^[3] The series advanced further with EGM2008, publicly released by NGA in 2008 to degree 2190 (corresponding to 5 arc-minute resolution), merging Gravity Recovery and Climate Experiment (GRACE) satellite gravimetry data up to degree 180 with a global grid of 5 arc-minute mean free-air gravity anomalies derived from terrestrial, airborne, and altimetry sources. This least-squares spectral combination approach, supplemented by topographic corrections, yielded a global geoid accuracy of approximately 15 cm when validated against GPS/leveling data.^[2] EGM2020 was planned for release by NGA around 2020 and intended to maintain the degree 2190 resolution while integrating post-2008 datasets, including observations from the GRACE Follow-On (GRACE-FO) mission, updated altimetry models like DTU17, and enhanced surface gravimetry with a focus on underrepresented regions. However, as of 2025, EGM2020 has not been released, and EGM2008 remains the current operational standard, with NGA tentatively planning a new model around 2028.^[19]^[20]^[1] Across the EGM series, coefficient estimation has relied on the integration of satellite altimetry for oceanic gravity anomalies, terrestrial and airborne gravimetry for land-based measurements, and GPS-derived positions for validation and height references, all processed via least-squares adjustment techniques to harmonize disparate data types into a unified spherical (or ellipsoidal) harmonic expansion.^[2]^[3]

Recent and Experimental Models

The Gravity Recovery and Climate Experiment (GRACE) mission produced the GGM05 series of models in 2007, which advanced the characterization of time-variable gravity fields up to spherical harmonic degree and order 160, derived primarily from satellite-to-satellite tracking data.^[21] These models, including the satellite-only GGM05S and GRACE-specific GGM05G variants, incorporated monthly solutions that revealed continental-scale mass transport processes, such as hydrological cycles and ice mass changes.^[21] GRACE's monthly gravity fields, processed to capture temporal variations, demonstrated significant signals from global mass redistributions, enabling quantitative assessments of phenomena like groundwater depletion and ocean mass gain.^[22] The Gravity Field and Steady-State Ocean Circulation Explorer (GOCE) mission yielded the TIMR series of models in the 2010s through the time-wise approach, which processed satellite gravity gradiometry observations to derive high-resolution static gravity fields complete to degree 280.^[23] These models emphasized the recovery of short-wavelength gravity features associated with crustal density variations, leveraging GOCE's electrostatic gradiometer to measure second derivatives of the gravity potential with high sensitivity.^[24] The TIMR5 release, for instance, provided enhanced resolution for regional geoid undulations and gravity anomalies over areas with complex topography, surpassing prior satellite-only models in the medium- to high-degree spectrum.^[25] Experimental efforts toward a successor to EGM2008, including prototypes for the planned EGM2020, were pursued during 2018–2020, incorporating innovative gravimetry techniques such as altimetry-derived gravity from satellite radar systems to test high-resolution global coverage.^[20] Complementing this, the XGM2019e model combined satellite gravimetry from the GOCO06s solution with extensive terrestrial gravity data, yielding a global field represented by spheroidal harmonics up to degree and order 5399 for a spatial resolution of approximately 4 km.^[26] This experimental model integrated over 20 million terrestrial measurements to refine high-frequency components, particularly in data-sparse regions.^[26] Alternative approaches include the European Space Agency's DIR-R5 model from 2014, updated in subsequent releases through 2021, which employed a direct least-squares method on satellite data to model dynamic gravity variations up to degree 300, focusing on low- to medium-frequency signals from GRACE and GOCE.^[27] For local anomalies, non-spherical harmonic techniques such as mass concentration (mascon) modeling represent the gravity field using discrete point or disc masses, effectively capturing isolated density contrasts without full global parameterization.^[28] Mascon methods, applied regionally, improve resolution for features like sedimentary basins or volcanic structures by optimizing mass parameters against local gravity data.^[28] A key innovation in these recent models is the explicit incorporation of time-dependency, which has enabled detailed studies of climate-related mass changes, such as Antarctic and Greenland ice melt contributing to sea-level rise.^[29] GRACE-derived time-variable solutions, for example, quantified ice sheet mass loss at rates exceeding 400 gigatons per year in the early 2000s, linking gravitational signals to global warming impacts.^[30] This temporal dimension extends beyond static fields, providing essential data for monitoring Earth's mass budget in response to anthropogenic and natural forcings. Ongoing data from GRACE-FO continues to support refinements in time-variable gravity modeling as of 2025.^[30]

Applications and Uses

Geodesy and Mapping

Earth Gravitational Models (EGMs) play a central role in geodesy by enabling the computation of the geoid, which represents the equipotential surface of Earth's gravity field approximating mean sea level. The geoid undulation, denoted as N, quantifies the separation between the geoid and a reference ellipsoid, and is calculated using spherical harmonic coefficients from EGMs to convert ellipsoidal heights (h, measured via GNSS) to orthometric heights (H, referenced to the geoid) through the relation N = h - H. This process involves evaluating the disturbing potential from the EGM coefficients up to a specified degree and order, typically combined with local gravity data for higher accuracy. For instance, the EGM2008 model, complete to degree and order 2159, provides global geoid undulations with a resolution of about 5 arcminutes, supporting precise height transformations essential for surveying and vertical datum realization.^[31]^[2] In global height systems, EGMs are integral to frameworks like the World Geodetic System 1984 (WGS84) and the International Terrestrial Reference Frame (ITRF), where they facilitate the removal of gravitational effects to achieve consistent orthometric heights. Within WGS84, the EGM component—such as EGM2008—computes geoid heights and gravity anomalies to align ellipsoidal coordinates with physical heights, ensuring compatibility across GNSS applications and vertical datums. Similarly, in ITRF realizations, EGMs contribute to the International Height Reference Frame (IHRF) by providing geopotential values at stations, enabling the unification of national height systems through precise leveling corrections that account for spatial variations in gravity. This gravitational adjustment is critical for eliminating inconsistencies in spirit leveling surveys, where cumulative errors from terrain and Earth curvature are mitigated to sub-centimeter precision over long distances.^[1]^[32] EGMs enhance mapping applications by integrating with Digital Elevation Models (DEMs) to perform terrain corrections, refining gravity field estimates for topographic surveys. In gravimetric geoid modeling, residual terrain effects are computed by subtracting the attraction of a reference DEM from observed gravity data, with EGMs supplying the long-wavelength component; high-resolution DEMs, such as those from SRTM, extend this correction to short wavelengths for accurate mass redistribution modeling. Furthermore, satellite gravimetry missions like GRACE and GOCE, whose data underpin modern EGMs, aid ocean bathymetry by deriving gravity anomalies that correlate with seafloor topography, allowing inference of underwater features where direct measurements are sparse—such as abyssal plains—through inverse modeling of the marine geoid. This synergy supports global topographic mapping, including combined land-ocean DEMs like ETOPO, by incorporating gravitational signals to validate and interpolate bathymetric grids.^[33]^[34] A notable case study is the incorporation of EGM2008 into the United States Gravimetric Geoid 2012 (USGG2012), which served as the reference for the hybrid GEOID12B model. More recently, EGM2008 has been integrated into GEOID18 (released in 2018), the current hybrid model used by the National Geodetic Survey (NGS) for national surveying, combining global coefficients with airborne and terrestrial gravity data over North America and achieving gravimetric geoid accuracies of approximately 1 cm relative to GPS/leveling benchmarks in many areas. This enables orthometric height determinations for infrastructure projects and coastal management with minimal distortion from the North American Vertical Datum of 1988 (NAVD88). With the NSRS modernization planned for 2025-2026, GEOID2022—a purely gravimetric model—will replace GEOID18, further enhancing precision. This exemplifies how EGMs anchor local refinements, reducing systematic biases in height transformations to support centimeter-level precision in geodetic control networks.^[35]^[36]^[37]^[38] Earth Gravitational Models (EGMs) play a critical role in orbital perturbation modeling within numerical propagators, enabling accurate simulation of satellite trajectories by accounting for Earth's non-spherical mass distribution. The J2 term, representing the planet's oblateness, dominates these perturbations, inducing secular variations in orbital elements such as the right ascension of the ascending node, argument of perigee, and mean anomaly, which manifest as along-track drifts of up to several kilometers per day if unmodeled. Higher-degree harmonics, including zonal and tesseral terms, capture finer gravitational irregularities that exacerbate these drifts over extended missions, allowing propagators to mitigate cumulative position errors through precise force integration. For instance, models like EGM2008 incorporate spherical harmonic expansions up to degree 2190, reducing prediction inaccuracies in low Earth orbit simulations to centimeters.^[39]^[2] In Global Positioning System (GPS) and Global Navigation Satellite System (GNSS) applications, EGMs provide essential gravitational corrections for precise point positioning (PPP), where dual-frequency observations are processed to achieve sub-centimeter accuracy. EGM96 and EGM2008 are integrated into broadcast ephemerides to model Earth-fixed gravitational effects, facilitating ionospheric-free linear combinations that eliminate range delays from the ionosphere while preserving geopotential influences on satellite clocks and orbits. These corrections ensure reliable user positioning for aviation and maritime navigation, with EGM2008's enhanced resolution improving long-wavelength accuracy derived from GRACE satellite data. Without such models, uncompensated gravitational perturbations would degrade PPP solutions by meters.^[40]^[2] Aerospace engineering leverages EGMs for trajectory design in interplanetary missions, where precise modeling of Earth escape dynamics informs launch windows and hyperbolic trajectories, as utilized in NASA's Deep Space Network (DSN) for Doppler and ranging tracking of outbound spacecraft. High-degree terms in these models are particularly vital for re-entry predictions of returning capsules, refining atmospheric interface timelines by simulating gravitational torques and drag onset influenced by Earth's irregular potential. For example, EGM-based propagators support DSN operations by minimizing tracking errors during critical maneuvers, ensuring safe de-orbiting with predicted landing ellipses reduced from kilometers to hundreds of meters.^[7] The practical impacts of EGMs on error budgets are profound: neglecting these models can lead to orbit errors exceeding 1 km per day from unaccounted J2 and higher-order effects, severely compromising mission longevity and safety. Advancements from GRACE-derived data in EGM2008 have dramatically improved this, shrinking radial and along-track errors from tens of meters in early 1980s models to centimeters today, thereby enabling more reliable aerospace operations and reducing fuel margins for orbit maintenance.^[41]^[2]

Comparisons and Future Directions

Model Accuracies and Limitations

The Earth Gravitational Models (EGMs) achieve varying levels of precision depending on the specific model and the gravitational functional considered. For instance, the EGM2008 model attains a global root-mean-square (RMS) commission error of approximately 11 cm for geoid undulations, meeting or exceeding its development goal of 15 cm, with regional improvements such as 5-6 cm over continental United States and ocean areas below 66° latitude.^[2] Gravity anomaly residuals from EGM2008 typically range from 5 to 10 mGal globally, with lower values (around 2-5 mGal) in well-surveyed regions and higher in data-sparse areas, representing a three- to six-fold improvement over prior models like EGM96.^[2] These metrics establish the scale of model performance but highlight that absolute accuracies remain sub-meter for geoid heights and sub-10 mGal for anomalies in most applications. Key limitations in EGMs stem from two primary error sources: omission and commission errors. Omission errors arise from the truncation of spherical harmonic expansions at a finite degree (e.g., degree 2190 for EGM2008), which excludes high-frequency gravitational signals associated with fine-scale topography and subsurface density variations below the model's resolution limit of about 5 arcminutes.^[2] Commission errors, on the other hand, result from inaccuracies in the estimated harmonic coefficients due to noise in input data such as satellite tracking, altimetry, and terrestrial gravity measurements, leading to propagated uncertainties in the modeled potential.^[2] These errors are particularly pronounced in the high-degree terms, where signal-to-noise ratios diminish. Regional variations in model accuracy reflect disparities in data density and quality. Over continents with extensive terrestrial gravity surveys, such as North America and Europe, EGMs exhibit superior performance with geoid errors often below 10 cm, owing to dense ground data integration.^[42] In contrast, oceanic and polar regions suffer from sparser coverage, relying more on satellite altimetry and gravimetry, resulting in elevated errors—up to 20-30 cm for geoid undulations and 10-15 mGal for gravity anomalies.^[2] Additionally, EGMs represent a static, mean gravitational field and thus do not incorporate dynamic effects like tidal deformations or variations in Earth's rotation, which can introduce short-term perturbations on the order of centimeters in height and milligals in gravity.^[2] Validation of EGMs typically involves comparisons with independent datasets to quantify these errors. Geoid accuracies are assessed against GPS/leveling networks, where differences yield RMS residuals that benchmark model performance, such as the 4.8 cm observed over the continental United States for EGM2008.^[2] For gravity anomalies and broader field consistency, forward modeling of satellite orbits—comparing predicted perturbations from the model against observed tracking data—provides global validation, often confirming residuals below 10 mGal when integrated with missions like GRACE.^[2]

Ongoing Improvements

Ongoing research in Earth gravitational modeling emphasizes the development of next-generation satellite missions to enhance spatial and temporal resolution of gravity field measurements. The European Space Agency's (ESA) Next Generation Gravity Mission (NGGM), part of the Mass Change and Geosciences International Constellation (MAGIC), proposes a constellation of four satellites operating in pairs to monitor temporal variations in Earth's gravity field with unprecedented detail, achieving a geoid accuracy of 1 mm at 500 km spatial resolution every 3 days and at 150 km every 10 days. This mission builds on predecessors like GRACE-FO by incorporating advanced laser interferometry for inter-satellite ranging, enabling detection of mass changes in over 80% of significant river basins and aquifers that were previously unresolvable. Additionally, ESA's GOCE mission underwent a second phase in 2012–2013 by lowering its orbit to 225 km, which improved gravity gradient data resolution and contributed to refined static field models, paving the way for similar enhancements in future gravimetry concepts.^[43]^[44]^[45] Methodological advancements are incorporating machine learning techniques to address limitations in traditional spherical harmonic coefficient estimation, particularly for interpolating sparse datasets. Neural networks, such as convolutional models, have demonstrated superior performance in predicting gravity anomalies in data-void regions by learning spatial patterns from multi-source inputs like satellite altimetry and terrestrial gravimetry, achieving prediction accuracies up to 20% higher than classical kriging methods in tested areas. Furthermore, integration of relativistic effects into gravity models is gaining traction for applications requiring sub-nanosecond timing precision, such as GNSS and deep-space navigation; general relativistic corrections for gravitational time dilation, including post-Newtonian terms in the geopotential, ensure clock synchronization errors remain below 1 ns over global scales.^[46]^[47]^[48] Key challenges persist in filling data gaps, especially in remote or inaccessible regions like Antarctica and parts of South America, where terrestrial measurements are sparse and satellite coverage is limited, leading to uncertainties exceeding 10 mGal in local gravity anomalies. Temporal variations driven by climate processes, such as groundwater depletion and glacier melt, further complicate modeling; for instance, GRACE-derived data reveal annual groundwater storage losses of up to 20 cm equivalent water height in arid basins, necessitating dynamic updates to static models to capture these signals accurately.^[49]^[50]^[51] The U.S. National Geospatial-Intelligence Agency (NGA) and ESA are collaborating on roadmaps for future models, including the completed but unreleased EGM2020 to supersede EGM2008 with enhanced data integration from recent missions, through multi-sensor fusion of gravimetry, altimetry, and GNSS data. These efforts, informed by NGGM simulations, prioritize ensemble approaches to propagate uncertainties and integrate time-variable components for decadal forecasting of mass redistribution.^[52]^[53]^[54]

References

[1]
World Geodetic System 1984 (WGS 84) - NGA - Office of Geomatics
Earth Gravitational Model 2008 (EGM2008). The EGM84 and EGM96 are legacy products. Users are encouraged to use the latest EGM from NGA, currently EGM2008 ...
[2]
The development and evaluation of the Earth Gravitational Model 2008 (EGM2008)
Summary of each segment:
[3]
[PDF] NASA/TP—1998–206861
... Earth Gravitational Model 1996 (EGM96), incorporates improved surface gravity data, altimeter- derived gravity anomalies from ERS–1 and from the GEOSAT ...
[4]
https://earth-info.nga.mil/index.php?dir=gravity&action=egm2008
[5]
GEOID18 Technical Details - National Geodetic Survey
For global needs, the National Geospatial-Intelligence Agency (NGA) produces EGM2008 and the soon to be released EGM2020. Other Hybrid Geoid Models. The ...
[6]
[PDF] Global Models - ICGEM
It is a mathematical function which describes the gravity field of the Earth in the 3-dimensional space. The determination of the Earth's global gravity field ...
[7]
[PDF] THE EVOLUTION OF EARTH GRAVITATIONAL MODELS USED IN ...
The ultimate goal in the de- velopment of Earth gravitational field models is to allow construction of a very accurate satellite-based global model to very high ...
[8]
Earth's gravity field | McGraw Hill's AccessScience
The gravitational potential V at a point P is the potential energy, due to the Earth's gravitational attraction, of a unit mass situated at P.
[9]
Earth's Gravitational Field - an overview | ScienceDirect Topics
The Earth's gravitational field is defined as the gravitational influence exerted by the Earth, which is modeled using classical potential theory and Newton's ...
[10]
NGA Geomatics - WGS 84
### Summary of Earth Gravitational Models, Especially EGM2008
[11]
[PDF] Geodetic Reference System 1980
This theory had already served as a base for the International Gravity Formula adopted at the General Assembly in Stockholm in 1930.Missing: history | Show results with:history
[12]
International Gravity Formula
It was further elaborated by C. Somigliana in 1929 and served already as the basis for the International Gravity. Formula adopted at the General Assembly of the ...
[13]
The gravitational field of the earth derived from motions ... - NASA ADS
... harmonics of the earth's gravitational field by analyzing the motion of the first Vanguard artificial satellite, 1958 ~32. Some observational data of other ...Missing: gravity model expansion
[14]
Gravity model improvement using Geos 3 (GEM 9 and 10) - Lerch
Jul 10, 1979 · Geocentric station positions were determined for approximately 150 stations in GEM 10. These station coordinates, their mean sea level heights, ...
[15]
Gravity model improvement using GEOS-3 (GEM 9 and 10)
GEM 10 is complete to 22 x 22 with selected higher degree terms out to degree and order 30 amounting to a total of 592 coefficients. Comparisons with surface ...Missing: 1970s Earth
[16]
Now 40, NASA's LAGEOS Set the Bar for Studies of Earth
May 4, 2016 · On May 4, 1976, NASA launched a cannonball-shaped satellite that transformed studies of Earth's shape, rotation and gravity field.Missing: transition | Show results with:transition
[17]
LAGEOS (Laser Geodynamics Satellite-I) / LAGEOS-II - eoPortal
May 31, 2012 · Launch date, 04 May 1976 ; Measurement domain, Gravity and Magnetic Fields ; Measurement category, Gravity, Magnetic and Geodynamic measurements.Missing: transition | Show results with:transition
[18]
The World Geodetic System 1984 Earth Gravitational Model. - DTIC
The World Geodetic System 1984 WGS 84 Earth Gravitational Model EGM consists of a set of normalized geopotential coefficients complete through degree n and ...Missing: NGA | Show results with:NGA
[19]
https://ui.adsabs.harvard.edu/abs/2020AGUFMG016...08B/abstract
[20]
[PDF] EGM2020: Updates - Sirgas
Sep 18, 2019 · NGA Hybrid Model. Approved for Public Release #19-987. Page 16. 16. UNCLASSIFIED. UNCLASSIFIED. Approved for Public Release #19-987. Page 17. 17.
[21]
GRACE Gravity Model - Center for Space Research
The global Earth gravity model, GRACE Gravity Model 05 (GGM05) is a significant improvement upon, and supersedes, the previously released models.
[22]
Overview - Monthly Mass Grids | Data - GRACE Tellus - NASA
Monthly mass grids show changes in gravity caused by mass changes, measured in equivalent water thickness, relative to a time-mean baseline.Missing: GGM05 160
[23]
[PDF] GOCE gravity field models following the time-wise approach
... GOCE gravity field models was released. One of the series of models follows the so called time-wise approach (EGM_TIM). They are purely based on GOCE obser-.
[24]
GOCE (Gravity field and steady-state Ocean Circulation Explorer)
GOCE's four years in orbit resulted in a series of four gravity models, each more accurate than the last. These models have been used to generate corresponding ...Missing: TIMR | Show results with:TIMR
[25]
Processing aspects of level 2 GOCE gradiometer data for regional ...
Nov 16, 2018 · ... model TIMR5 (Brockmann et al. 2014), a GOCE-only gravity model complete up to degree and order 280. The comparisons and the evaluated ...2 Data Acquisition And... · 3 Data Filtering And... · 4 Projection Onto A Mean...Missing: TIMR | Show results with:TIMR
[26]
The combined global gravity field model XGM2019e
Jul 20, 2020 · XGM2019e is a combined global gravity field model represented by spheroidal harmonics up to degree and order (d/o) 5399, corresponding to a spatial resolution ...
[27]
ESA's satellite‐only gravity field model via the direct approach ...
Oct 22, 2014 · The model contains signal over the Northern Hemisphere midlatitudes (i.e., lowest part of the GOCE orbit) to d/o 300. Evaluations from degrees ...
[28]
[PDF] A review of different mascon approaches for regional gravity field ...
Sep 29, 2022 · Mascon approaches are used for modeling gravity fields, including point mass and planar disc mascons, and are used for both the Moon and Earth.
[29]
Time‐variable gravity observations of ice sheet mass balance ...
May 2, 2013 · We analyze current progress and uncertainties in GRACE estimates of ice sheet mass balance. We discuss the impacts of errors associated with ...
[30]
Contributions of GRACE to understanding climate change - PMC
Time-resolved satellite gravimetry has revolutionized understanding of mass transport in the Earth system. Since 2002, the Gravity Recovery and Climate ...
[31]
Evaluation of Global Gravitational Models Based on DGPS/leveling ...
The EGM84 is a gravitational model of the Earth that gave us a lot of geopotential coefficients to degree and request 180. A 30-minute overall geoid tallness ...
[32]
Strategy for the realisation of the International Height Reference ...
Feb 22, 2021 · This study focuses on the strategy for the realisation of the IHRS; ie the establishment of the International Height Reference Frame (IHRF).
[33]
[PDF] Gravity and Bathymetry modelling from satellite altimetry - CORE
Sep 17, 2019 · Using a high resolu- tion Digital Elevation Model (DEM), the RTM effects can be computed by numerical integration of mass prisms (alternatively ...
[34]
Bathymetry from space: Rationale and requirements for a new, high ...
A new satellite altimeter mission, optimized to map the deep ocean bathymetry and gravity field, will provide a global map of the world's deep oceans at a ...1. Mapping The Ocean Floor · 1.1. Sensing Gravity And... · 2. New Science
[35]
[PDF] Technical Details USGG2012 Models - National Geodetic Survey
All three GEOID models are computed based on the gravimetric geoid USGG2012 . ... Both USGG2009 and USGG2012 were based on the EGM2008 reference model.
[36]
Establishing Orthometric Heights Using GNSS — Part 3 - GPS World
Oct 7, 2015 · The relative accuracy of GEOID12B to NAVD88 is characterized by a misfit of +/-1.7 centimeters nationwide. It should be noted that other ...
[37]
[PDF] Gravity Modeling for Variable Fidelity Environments
• EGM96 (Earth Gravitational Model) developed by NASA, OSU, and the National Imagery and Mapping Agency. (NIMA). Published geopotential models list normalized ...
[38]
NGA - Office of Geomatics - National Geospatial-Intelligence Agency
Earth Gravitational Model 2008 (EGM2008) Data and Apps. The following links contain documentation, source code, data, and supporting inputs and tools for ...Missing: specification | Show results with:specification
[39]
Improvement in the radial accuracy of altimeter-satellite orbits due to ...
We demonstrate the improvement in the accuracy of the orbit radius due to Earth gravity models, from the early 1980s (order of tens of meters), to the present ...
[40]
[PDF] Assessment of EGM2008 over Germany Using Accurate Quasigeoid ...
EGM2008 is most accurate (i. e., low- est commission errors) in regions with high-quality ter- restrial gravity data sets (i. e., dense coverage, sufficient.<|control11|><|separator|>
[41]
ESA - The future of gravity is MAGIC - European Space Agency
Apr 25, 2023 · “MAGIC will give us data on more than 90% of Earth's major hydrological units such as river basins and aquifers with 260 km spatial resolution ...Missing: models 2030
[42]
Next Generation Gravity Mission design activities within the MAGIC
The objective of ESA's Next Generation Gravity Mission (NGGM) is long-term monitoring of the temporal variations of Earth's gravity field.Missing: MAGNET | Show results with:MAGNET
[43]
GOCE's second mission improving gravity map - ESA
Nov 16, 2012 · ESA's GOCE gravity satellite has already delivered the most accurate gravity map of Earth, but its orbit is now being lowered in order to obtain even better ...
[44]
Gravity Predictions in Data-Missing Areas Using Machine Learning ...
Nov 8, 2024 · The results indicate that machine learning methods exhibit a marked advantage in gravity data prediction, significantly enhancing the predictive accuracy.
[45]
Application of Machine Learning Methods for Gravity Anomaly ...
The use of machine learning offers an advance in gravity data prediction by improving the accuracy of interpolation in regions with sparse or irregular ...2. Materials And Methods · 2.5. Gaussian Process... · 3. Results
[46]
Relativistic geoid: Gravity potential and relativistic effects
Mar 17, 2020 · In this work, we define a relativistic gravity potential, which generalizes the Newtonian one. It exists for any stationary configuration1 with ...
[47]
High-resolution regional gravity field modeling in data-challenging ...
Feb 26, 2024 · This paper addresses this issue in a case study of Colombia, where eight decades of historical terrestrial and airborne gravity measurements are available.
[48]
New Antarctic Gravity Anomaly Grid for Enhanced Geodetic and ...
We present the first modern Antarctic-wide gravity data compilation derived from 13 million data points covering an area of 10 million km 2.
[49]
Identifying Climate-Induced Groundwater Depletion in GRACE ...
Mar 11, 2019 · We explore relations between natural and human drivers and spatiotemporal changes in groundwater storage derived from the Gravity Recovery and Climate ...
[50]
Short‐Period Mass Variations and the Next Generation Gravity Mission
Jan 21, 2025 · A simplified comparison between two proposed designs for a future earth gravity mission with and without strongly reduced non-gravitational ...
[51]
Next Generation Gravity Mission Elements of the Mass Change and ...
The mission aims at enabling long-term monitoring of the temporal variations of Earth's gravity field at relatively high temporal (down to 3 days) and ...<|separator|>