Search

The AlpArray Gravity Research Group (AAGRG), as part of the European AlpArray program, focuses on the compilation of a homogeneous surface-based gravity data set across the Alpine area. In 2017 10 European countries in the Alpine realm agreed to contribute with gravity data for a new compilation of the Alpine gravity field in an area spanning from 2 to 23∘ E and from 41 to 51∘ N. This compilation relies on existing national gravity databases and, for the Ligurian and the Adriatic seas, on shipborne data of the Service Hydrographique et Océanographique de la Marine and of the Bureau Gravimétrique International. Furthermore, for the Ivrea zone in the Western Alps, recently acquired data were added to the database. This first pan-Alpine gravity data map is homogeneous regarding input data sets, applied methods and all corrections, as well as reference frames. Here, the AAGRG presents the data set of the recalculated gravity fields on a 4 km × 4 km grid for public release and a 2 km × 2 km grid for special request. The final products also include calculated values for mass and bathymetry corrections of the measured gravity at each grid point, as well as height. This allows users to use later customized densities for their own calculations of mass corrections. Correction densities used are 2670 kg m−3 for landmasses, 1030 kg m−3 for water masses above the ellipsoid and −1640 kg m−3 for those below the ellipsoid and 1000 kg m−3 for lake water masses. The correction radius was set to the Hayford zone O2 (167 km). The new Bouguer anomaly is station completed (CBA) and compiled according to the most modern criteria and reference frames (both positioning and gravity), including atmospheric corrections. Special emphasis was put on the gravity effect of the numerous lakes in the study area, which can have an effect of up to 5 mGal for gravity stations located at shorelines with steep slopes, e.g., for the rather deep reservoirs in the Alps. The results of an error statistic based on cross validations and/or “interpolation residuals” are provided for the entire database. As an example, the interpolation residuals of the Austrian data set range between about −8 and +8 mGal and the cross-validation residuals between −14 and +10 mGal; standard deviations are well below 1 mGal. The accuracy of the newly compiled gravity database is close to ±5 mGal for most areas. A first interpretation of the new map shows that the resolution of the gravity anomalies is suited for applications ranging from intra-crustal- to crustal-scale modeling to interdisciplinary studies on the regional and continental scales, as well as applications as joint inversion with other data sets. The data are published with the DOI https://doi.org/10.5880/fidgeo.2020.045 (Zahorec et al., 2021) via GFZ Data Services., Earth System Science Data, 13 (5), ISSN:1866-3516, ISSN:1866-3508

This chapter describes the application and coverage of gravity and magnetic data for Antarctica with emphasis on airborne and satellite models. Low resolution satellite data help to fill gaps between high-resolution airborne data. Satellite gravity data are best used to study broad-scale lithospheric architecture while airborne data, especially magnetic data, provide finer detail. We review examples of gravity and magnetic analysis and describe the possibilities and pitfalls for estimating the properties of the lithosphere as it relates to the mantle. This is followed by a discussion on geothermal heat flow and possible ways to combine different geophysical and petrological models for a better understanding of the Antarctic mantle.

SUMMARY We present a novel approach for linearized gravity inversion to estimate the Moho depth, which allows the use of any gravitational component instead of the vertical gravity component only. The inverse problem is solved with the Gauss–Newton algorithm and the gravitational field of the undulating Moho depth is calculated with tesseroids. Hereby, the density contrast can be laterally variable by using information from seismological regionalization. Our approach is illustrated with a synthetic example, which we use to explore different regularization parameters. The vertical gravity gradient gzz provides the most reasonable results with appropriate parameters. As a case example, we invert for the Moho depth of the Amazonian Craton and its surroundings. The results are constrained by estimates from active seismic measurements. Our new Moho depth model correlates to tectonic domains and is in agreement with previous models. The estimated density contrasts of the tectonic domains agree well with the lithospheric architecture and show with 300–450 kg m–3 lower density contrasts for continental domains, whereas the oceans reveal a density contrast of 450–500 kg m–3. The wider range of estimated density contrast for the continent reflects uncertainties in Precambrian Fold Belts that arise from its small gravity signal. Our results demonstrate that a variable density contrast at the Moho depth is a valuable enhancement for gravity inversion.

Airborne gravity gradiometry measures several gravity gradient components unlike conventional gravimetry taking only the vertical gravity component into account. However, processing of multicomponent airborne gravity gradient (AGG) data without corrupting their internal consistency is often challenging. Therefore, we have developed an equivalent source technique to tackle this challenge. With a combination of Gauss-fast Fourier transform and the Landweber iteration, we have developed an efficient way to compute equivalent sources for AGG data. This method can handle two components simultaneously. We first examined its viability by applying this approach to a synthetic example. Afterward, we applied our method to real AGG data collected in the area of Karasjok, Norway. Our result is almost the same as the results that meet the industry standard but with great efficiency.

A 3D crustal density model for Egypt was compiled. It is constrained by available deep seismic refraction, receiver functions analysis, borehole, and geological data. In Egypt, seismic data are sparsely and irregularly distributed. Consequently, we developed the crustal thickness model by integrating seismic and gravity data. Satellite gravity data was inverted to build an initial model, which was followed by a detailed 3D forward gravity modelling. The initial crustal thickness is determined by applying seismically constrained non-linear inversion, based on the modified Bott's method and Tikhonov regularization assuming spherical Earth approximation. Moreover, the gravity inversion-based Moho depth estimates are in good agreement with results of seismic studies and are exploited for the 3D forward modelling. Crustal thicknesses range from 25 to 30 km along the rifted margins of the Red Sea, which thin toward the Mediterranean Sea. Thicknesses in southern Egypt reach values between 35 and 40 km. A maximum crustal thickness of 45 km is found in the southwestern part of Egypt. Within the Sinai Peninsula, the thickness varies from the shallow southern edge (∼ 31 km) and increases toward the North (∼ 36 km). Our model revealed a thick lower crust beneath the southern part of Egypt, which can be associated with crustal modification that occurred during the collision of East Gondwana and the Saharan Metacraton along the Keraf suture zone during the final assembly of Gondwana in the Neoproterozoic. Finally, the isostatic implications of the differences between the seismic and gravity-derived Mohos are thoroughly discussed. In conclusion, the developed 3D crustal thickness model provides high-resolution Moho depth estimates that closely resembles the major geological and tectonic features. Also, the existing correlation between the topography, Bouguer anomalies, and Moho depths indicates that the investigated area is close to its isostatic equilibrium.

The AlpArray Gravity Research Group (AAGRG), as part of the European AlpArray program, focuses on the compilation of a homogeneous surface-based gravity dataset across the Alpine area. From this data set, Bouguer- and Free Air anomalies are calculated and presented here. In 2016/17 ten European countries in the Alpine realm have agreed to contribute with gravity data for a new compilation of the Alpine gravity field in an area from 2° to 23° East and from 41° to 51° North. This compilation relies on existing national gravity databases and, for the Ligurian and the Adriatic seas, on ship-borne data of the Bureau Gravimétrique International. Furthermore, for the Ivrea zone in the Western Alps, recently acquired data were added to the database. This first pan-Alpine gravity data map is homogeneous regarding input data sets, applied methods and all corrections as well as reference frames. Here, the AAGRG presents the data set of the recalculated gravity fields on a 4 km × 4 km grid for public release, 2 km × 2 km for special request. The final products also include calculated values for mass/bathymetry corrections of the measured gravity at each grid point, as well as height. This allows users to use later customized densities for their own calculations of mass corrections. Correction densities used are 2670 kg m−3 for landmasses, 1030 kg m−3 for water masses above and −1640 kg m−3 below the ellipsoid. The correction radius was set to the Hayford zone O2 (167 km). The new Bouguer anomaly is station completed (CBA) and compiled according to the most modern criteria and reference frames (both positioning and gravity), including atmospheric corrections. Special emphasis was put on the gravity effect of the numerous lakes in the study area, which can have an effect of up to 5 mGal for gravity stations located at shorelines with steep slopes, e.g., for the rather deep reservoirs in the Alps. The results of an error statistic based on cross validations and/or interpolations residuals is provided for the entire database. As an example, the interpolation residuals of the Austrian data set range between about −8 and +8 mGal, the cross-validation residuals between −14 mGal and +10 mGal; standard deviations are well below 1 mGal. The accuracy of the newly compiled gravity database is close to ±5 mGal for most areas. A first interpretation of the new map shows that the resolution of the gravity anomalies is suited for applications ranging from intra-crustal to crustal scale modelling to interdisciplinary studies on the regional and continental scales as well as applications as joint inversion with other datasets. The data will be published with the DOI https://doi.org/10.5880/fidgeo.2020.045 (Zahorec et al., 2020) when the final paper is accepted. In the meantime, the data is accessible via this temporary review link: https://dataservices.gfz-potsdam.de/panmetaworks/review/fdc35a9f6551b01b6152ee1af7b91a5a0c3de5341d067644522c192ad7f25e7f.

Geothermal heat flux under the Antarctic ice is one of the least known parameters. Different methods (based on e.g., magnetic or seismic data) have been applied in recent years to quantify the thermal structure and the geothermal heat flux, resulting in vastly different estimates. In this study, we use a Bayesian Monte-Carlo-Markov-Chain approach to explore the consistency of such models and to which degree lateral variations of the thermal parameters are required. Hereby, we evaluate the input from different lithospheric models and how they influence surface heat flux. We demonstrate that both Curie isotherm and heat production are dominating parameters for the thermal calculation and that use of incorrect models or sparsely available data lead to unreliable results. As an alternative approach, geological information should be coupled with geophysical data analysis, as we demonstrate for the Antarctic Peninsula.

In this study, we present a novel approach combining the advantages of tesseroids in representing geophysical structures though their voxel-like discretization features with a spherical harmonic representation of the magnetic field. Modelling of the Earth lithospheric magnetic field is challenging since part of the spectra is hidden by the core field and the forward modeled field of a lithospheric magnetization is always biased by the spectral range used. In our approach, a spherical harmonic representation of the magnetic field of spherical prisms (tesseroids) is used for high-resolution magnetic inversion of lithospheric field models. The use of filtered spherical harmonic models of the magnetic field of each tesseroid ensures that the resulting field matches the spectral range of the input data. For the inversion, we use the projected gradient method. The projected gradient method easily allows us to assign an initial guess (i.e., a-priori assumption) for the inversion and avoids negative values of susceptibilities. The latter is providing more plausible models since induced magnetization is assumed to be dominant over the continents and, for the oceans, a remanence model can be subtracted. We show an application of the technique to a synthetic dataset and a satellite-derived lithospheric field model where the model geometry is based on seismic information. We also demonstrate a proof-of-concept for high-resolution tile-wise inversion for the Bangui anomaly in Africa.

The nature and origin of the two Large Low Velocity Provinces in the lowest part of the mantle remain controversial. They have been interpreted as a purely thermal feature, accumulation of subducted oceanic lithosphere or a primordial zone of iron enrichment. Information regarding the density of the LLVPs would help to constrain a possible explanation. In this work, we perform a density inversion for the entire mantle, by constraining the geometry of potential density anomalies using tomographic vote maps. Vote maps describe the geometry of potential density anomalies according to their agreement of multiple seismic tomographies, hence not depending on a single representation. Therefore, the geometries used for inversion are features observed in most tomographies. We use linear inversion and determine the regularization parameters using cross-validation. Two different input fields are used to study the sensitivity of the mantle density results to the treatment of the lithosphere. We find the best data fit is achieved if we assume that the lithosphere is in isostatic balance. The estimated densities obtained for the LLVPs are systematically positive density anomalies for the LLVPs in the lower 800–1000 km of the mantle, which would indicate a chemical component for the origin of the LLVPs. Both iron-enrichment and a MORB contribution are in accordance with our data, but the required super-adiabatic temperature anomalies for MORB would be close to 1000 K.

This study presents a recent combined regional gravity field model over Egypt, developed by integrating satellite and terrestrial data via applying the remove-compute-restore (RCR) principle and the least-squares collocation (LSC) procedure. A high-resolution digital terrain model was exploited for the computation of the terrain and residual terrain corrections. Hereby, all the signals that can be modelled or deterministically computed are considered known and then removed in order to reduce the order of magnitude of the input gravity data prior to applying the LSC. Several GOCE-only and combined global geopotential models (GGMs) have been thoroughly investigated with respect to the EGM2008, in which the space-wise (SPW) solution, namely the SPW-R5 model, demonstrated the best performance. For the development of the combined model, the SPW-R5 GGM has been integrated with both the EGM2008 GGM and the terrestrial data retrieved from 56,250 gravity stations of the Getech data, acquired in the framework of the African Gravity Project. The combined regional gravity model was compared to the state-of-art XGM2016 global gravity model. The standard deviation of the differences is 18.0 mGal in terms of Bouguer anomalies. The combined regional model fits well with the terrestrial gravity data along the chosen North–South oriented profile through the Nile Delta region. The improvements of the developed combined regional model over the XGM2016 are due to the use of a more extensive terrestrial dataset. In conclusion, our model is more suitable than solely using the ground data or GGMs for regional density modelling over Egypt. As an example, the comparison of using a global or regionally defined gravity model with the forward gravity modelling based on Saleh (Acta Geodaetica et Geophysica Hungarica 47(4):402–429, 2012) density model is performed.

We present a study on the depth to basement and magnetic crustal domains beneath the Iranian Plateau by modeling aeromagnetic and gravity data. First, field processing of the aeromagnetic data was undertaken to estimate the general characteristics of the magnetic basement. Afterwards, inverse modeling of aeromagnetic data was carried out to estimate the depth to basement. The obtained model of basement was refined using combined gravity and magnetic forward modeling. Hereby, we were able to distinguish different magnetic domains in the uppermost crust (10–20 km depths) influencing the medium to long wavelength trends of the magnetic anomalies. Magnetic basement mapping shows that prominent shallow magnetic features are furthermore located in the volcanic areas, e.g. the Urumieh Dokhtar Magmatic Assemblage. The presence of ophiolite outcrops in SE Iran implies that shallow oceanic crust (with high magnetization) is the main source of one of the biggest magnetic anomalies in entire Iran area located north of the Makran.

The Earth masses reside in a near-hydrostatic equilibrium, while the deviations are, for example, manifested in the geoid, which is nowadays well determined by satellite gravimetry. Recent progress in estimating the density distribution of the Earth allows us to examine individual Earth layers and to directly see how the sum approaches the observed anomalous gravitational field. This study evaluates contributions from the crust and the upper mantle taken from the LITHO1.0 model and quantifies the gravitational spectra of the density structure to the depth of 435 km. This is done without isostatic adjustments to see what can be revealed with models like LITHO1.0 alone. At the resolution of 290 km (spherical harmonic degree 70), the crustal contribution starts to dominate over the upper mantle and at about 150 km (degree 130) the upper mantle contribution is nearly negligible. At the spatial resolution $$

The main purpose of this study is to construct a new 3-D model of the Central Asian Orogenic Belt (CAOB) crust, which can be used as a starting point for future lithospheric studies. The CAOB is a Paleozoic accretionary orogen surrounded by the Siberian Craton to the north and the North China and Tarim Cratons to the south. This area is of great interest due to its enigmatic and still not completely understood geodynamic evolution. First, we estimate an initial crustal thickness by inversion of the vertical gravity component of the Gravity Field and Steady-State Ocean Circulation Explorer (GOCE) and DTU10 models. Second, 3-D forward modeling of the GOCE gravity gradients is performed, which determines the topography of the Moho, the geometry, and the density distribution of the deeper parts of the CAOB and its surroundings, taking into account the lateral and vertical density variations of the crust. The model is constrained by seismic refraction, reflection, and receiver function studies and geological studies. In addition, we discuss the isostatic implications of the differences between the seismic Moho and the resulting 3-D gravity Moho, complemented by the analysis of the lithostatic load distribution at the upper mantle level. Finally, the correlation between the contrasting tectonic domains and the thickness of the crust reveals the inheritance of Paleozoic and Mesozoic geodynamics, particularly the magmatic provinces and the orocline which preserve their crustal features.

The pseudo-spectral form of the sea level equation (SLE) requires the approximation of a radially-symmetric visco-elastic Earth. Thus, the resulting predictions of sea level change (SLC) and glacial isostatic adjustment (GIA) often ignore lateral variations in the Earth structure. Here, we assess the capabilities of a Multiple 1D Earth Approach (M1DEA) applied to large-scale ice load components with different Earth structures to account for these variations. In this approach the total SLC and GIA responses result from the superposition of individual responses from each load component, each computed globally assuming locally-appropriate 1D Earth structures. We apply the M1DEA to three separate regions (East Antarctica, West Antarctica, and outside Antarctica) to analyze uplift rates for a range of Earth structures and different ice loads at various distances. We find that the uplift response is mostly sensitive to the local Earth structure, which supports the usefulness of the M1DEA. However, stresses transmitted across rheological boundaries (e.g., producing peripheral bulges) present challenges for the M1DEA, but can be minimized under two conditions: (1) If the considered time period of ice loading for each component is consistent with the relaxation time of the local Earth structure. (2) If the load components can be subdivided according to the scale of the lateral variations in Earth structure. Overall, our results indicate that M1DEA could be a computationally much cheaper alternative to 3D finite element models, but further work is needed to quantify the relative accuracy of both methods for different resolutions, loads, and Earth structure variations.

In this study we combine seismological and petrological models with satellite gravity gradient data to obtain the thermal and compositional structure of the Antarctic lithosphere. Our results indicate that Antarctica is largely in isostatic equilibrium, although notable anomalies exist. A new Antarctic Moho depth map is derived that fits the satellite gravity gradient anomaly field and is in good agreement with independent seismic estimates. It exhibits detailed crustal thickness variations also in areas of East Antarctica that are poorly explored due to sparse seismic station coverage. The thickness of the lithosphere in our model is in general agreement with seismological estimates, confirming the marked contrast between West Antarctica (

The lithospheric structure of Antarctica is still under‐explored. Moho depth estimate studies are in disagreement by more than 10 km in several regions, including for example the hinterland of the Transantarctic Mountains. Taking account the sparseness of seismological stations and the non‐uniqueness of potential field methods, inversions of Moho depth are performed here based on satellite gravity data in combination with currently available seismically constrained Moho depth estimates. Our results confirm that a lower density contrast at the Moho is present under East Antarctica than beneath West Antarctica. A comparison between the Moho depth derived from our inversion and an Airy‐isostatic Moho model also reveals a spatially variable buoyancy contribution from the lithospheric mantle beneath contrasting sectors of East Antarctica. Finally, to test the plausibility of different Moho depths scenarios for the Transantarctic Mountains–Wilkes Subglacial Basin system, we present 2‐D lithospheric models along the TAMSEIS/GAMSEIS seismic profile. Our models show that if a moderately depleted lithospheric mantle of inferred Proterozoic age underlies the region, then a shallower Moho is more likely beneath the Wilkes Subglacial Basin. If however, re‐fertilisation processes occurred in the upper mantle, for example in response to Ross‐age subduction, then a deeper Moho scenario is preferred. We conclude that 3D lithospheric modeling, coupled with the availability of new seismic information in the hinterland of the Transantarctic Mountains is required to help resolve this controversy, thereby also reducing the ambiguities in geothermal heat flux estimation beneath this key part of the East Antarctic Ice Sheet.

Gravity gradients of the GOCE (Gravity field and steady state Ocean Circulation Explorer) mission are available and provide free available data with global homogeneous coverage. The spatial resolution is about 80 km which makes it possible to construct regional models of large or remote areas. We study the benefit of using this data for compiling a 3D regional density model of the lithosphere with emphasis on the crustal structure for the Arabian Peninsula. We keep the data at satellite level to avoid noise amplification. The initial density model was set up with only a few seismic constraints and modified by forward modeling. Forward models benefits from the use of the full gravity gradient tensor and the invariants. The misfits between the measured and calculated field are more easily recognized in the gravity gradient tensor than in the gravity field alone. Different shape and location of anomalies are visible in the different components. Hence, modeling is better controlled as geometries which fit one component do not necessarily fit all other components. Regional models are important for a general understanding of any area and as a background for local interpretations. Here, we explain how to construct density models from satellite gravity gradients and show the advantage of such an approach by comparison to earlier studies based on either seismic or gravity data.

Airborne gravity gradient (AGG) measurements offer an increased resolution and accuracy compared to terrestrial measurements. But interpretation and processing of AGG data are often challenging as levelling errors and survey noise affect the data, and these effects are not easily recognised in the gradient components. We adopted the classic method of upward continuation in the noise reduction using the noise level estimates by the AGG system. By iteratively projecting the survey data to a lower level and upward continuing the data back to the survey height, parts of the high-frequency signal are suppressed. The filter, which is defined by this approach, is directly dependent on the noise level of the AGG data, the maximum number of iterations and the iterative step. We demonstrate the method by applying it to both synthetic data and real AGG data over Karasjok, Norway, and compare the results to the directional filtering method. The results show that the iterative filter can effectively reduce high-frequency noise in the data.

A newly developed software package to calculate the magnetic field in a spherical coordinate system near the Earth's surface and on satellite height is shown to produce reliable modeling results for global and regional applications. The discretization cells of the model are uniformly magnetized spherical prisms, so called tesseroids. The presented algorithm extends an existing code for gravity calculations by applying Poisson's relation to identify the magnetic potential with the sum over pseudogravity fields of tesseroids.By testing different lithosphere discretization grids it is possible to determine the optimal size of tesseroids for field calculations on satellite altitude within realistic measurement error bounds. Also the influence of the Earth's ellipticity upon the modeling result is estimated and global examples are studied.The new software calculates induced and remanent magnetic fields for models at global and regional scale. For regional models far-field effects are evaluated and discussed. This provides bounds for the minimal size of a regional model that is necessary to predict meaningful satellite total field anomalies over the corresponding area. Tool for computation of the magnetic field of spherical prisms was developed.Appropriate width of spherical prisms for modeling is 1 or less degrees.Elliptic and spherical lithospheric models have approximately 2nT difference.15° Extension of a regional model is sufficient to avoid big edge effects.

We present a 3D regional crustal model for the North Atlantic, which is based on the integration of seismic constraints and gravity data. The model addresses the crustal thickness geometry, and includes information on sedimentary thickness, the presence of high-velocity zones in the lower crust, and information on the crustal density distribution in the continental and oceanic domains. Using an iterative forward- and inverse-modelling approach, we adhere to the seismic constraints within their uncertainty, but manage to enhance the crustal geometry in areas where seismic data are sparse or absent. A number of basins are resolved with more detail. Recently released seismic reflection data beneath the NE Greenland Shelf allowed for a major improvement of the crustal thickness estimates. Estimated Moho depths beneath the basins there vary between 15 and 25 km, which is compatible with the conjugate Norwegian margin. A major lower-crustal seismic velocity anomaly in the vicinity of the Greenland–Iceland–Faroe Ridge complex is supported by density modelling. We discuss the validity and uncertainties of our model assumptions and discuss the correlation with the main structural elements of the North Atlantic.

The interpretation protocols for defining offshore rifted margin architecture normally include seismic-reflection analysis supplemented by refraction and/or potential field modelling to help constrain sedimentary, basement and Moho geometries at depth and/or the presence of magmatic material. Interpretation of modern high-resolution long-offset reflection profiles shows that significant mismatches may arise between the structural observations made from these data and the common translation of density, magnetic or velocity values into specific rock types made by geophysical models. We illustrate this problem with three examples from the Mid-Norwegian rifted system, and discuss the implications with respect to the geological interpretation.

Curvature components derived from satellite gravity gradients provide new global views of Earth’s structure. The satellite gravity gradients are based on the GOCE satellite mission and we illustrate by curvature images how the Earth is seen differently compared to seismic imaging. Tectonic domains with similar seismic characteristic can exhibit distinct differences in satellite gravity gradients maps, which points to differences in the lithospheric build-up. This is particularly apparent for the cratonic regions of the Earth. The comparisons demonstrate that the combination of seismological, and satellite gravity gradient imaging has significant potential to enhance our knowledge of Earth’s structure. In remote frontiers like the Antarctic continent, where even basic knowledge of lithospheric scale features remains incomplete, the curvature images help unveil the heterogeneity in lithospheric structure, e.g. between the composite East Antarctic Craton and the West Antarctic Rift System.

Teleseismic receiver function analysis imaged a complex upper mantle structure in the Central Fjord Region of East Greenland, including an east-dipping high velocity layer and a mantle wedge of high crustal or low mantle velocities. This was interpreted as a fossil Caledonian subduction complex, including a slab of eclogitised mafic crust and an overlying wedge of serpentinised mantle. In this paper, we use a multi-disciplinary geophysical and petrological modelling approach to test this proposed fossil subduction model. The consistency of the obtained velocity model with the regional gravity field is tested by forward density modelling and isostatic calculations. The models show that the sub-crustal structure, given by the more buoyant mantle wedge and the dipping high velocity/density layer, yield in a markedly better fit as compared to a homogeneous mantle lithosphere. Petrological-geophysical modelling is performed by testing different upper mantle compositions with regard to topography, gravity and seismic velocities using Litmod2D. This suggests that the observed lower crustal/uppermost mantle bodies could be a combination of mafic intrusions, serpentinised peridotite and metamorphosed mafic crust. The preferred composition for the dipping structure is eclogitised mafic crust, and hydrated peridotite filling the overlying mantle wedge. Models lacking an eclogite layer or a hydrated upper mantle composition show an inferior fit and, therefore, are not favoured representatives. This supports the interpretation as a fossil subduction zone complex. The spatial relations with Caledonian structures suggest an early Caledonian origin.

Lithospheric density structure can be constructed from seismic tomography, gravity modelling, or using both data sets. The different approaches have their own uncertainties and limitations. This study aims to characterize and quantify some of the uncertainties in gravity modelling of lithosphere densities. To evaluate the gravity modelling we compare gravity-based and seismic velocity-based approaches to estimating lithosphere densities. In this study, we use a crustal model together with lithospheric isostasy and gravity field observations to estimate lithosphere densities. To quantify the effect of uncertainty in the crustal model, three models are implemented in this study: CRUST1.0, EuCrust-07 and a high-resolution P-wave velocity model of the British Isles and surrounding areas. Different P-wave velocity-to-density conversions are used to study the uncertainty in these conversion methods. The crustal density models are forward modelled into gravity field quantities using a method that is able to produce spherical harmonic coefficients. Deep mantle signal is assumed to be removed by removing spherical harmonic coefficients of degree 0–10 in the observed gravity field. The uncertainty in the resulting lithosphere densities due to the different crustal models is ±110 kg m−3, which is the largest uncertainty in gravity modelling. Other sources of uncertainty, such as the VP to density conversion (±10 kg m−3), long-wavelength truncation (±5 kg m−3), choice of reference model (