The national research program GéoFrance 3D is aimed to expand the knowledge of France's subsurface by combining geophysical methods based on geological objects defined by their geometry and physical properties. The first project achieved in the framework of the program concerns the northern part of the Armorican Massif. This massif preserves the most extended Precambrian terrane of France, the Cadomian Neoproterozoic arc-continent collision, a piece of the pan-African belt. GeoFrance 3D has revealed the anatomy of this NE-SW continental margin on which the Variscan belt has been built by combined geological and geophysical methods (seismic profiles and tomography, gravimetry, aeromagnetics). The differences between the eastern and western margin of the Cadomian belt can be explained by the original Neoproterozoic thrust system and by the location of the major strike slip faults.The Variscan belt is a metallogenical province at the scale of Europe. The Massif Central project is focused on the hydrothermal paleofield which was active at 315-310 Ma, responsible of the deposition of epigenetic gold and sulfide. Gravimetrical survey and seismic profiles across the Limousin revealed the layered structure of the crust and its decoupling by large transverse extensional faults like the Argentat fault. The onset and development of the hydrothermalism (source, channelling and deposition) correspond to a peculiar stage of the orogenic development i.e. the change from the Neovariscan collision to the late Variscan extension. A project is focused on the present day tectonics of the Alps and its crustal structure. The combination of neotectonic and sismotectonic studies, new constrains on the age of the high pressure metamorphism and seismic tomography imaging of the crust are indicative of the recent partitioning of the deformation at the scale of the lithosphere. Implications for kinematics of the collision are discussed. From the Cadomian belt of northern Brittany to the present day tectonics of the Alps, GeoFrance 3D contributes to the definition of a new crustal-scale section through France. A 3D geological editor has been developed for the needs of this program, connected to an information system using Internet and Java classes, aimed for management and exchanging of data and models.
The Sudbury Basin and the Matagami anticline of the Canadian Shield were selected for conducting the world's first 3D seismic surveys for deep base-metal exploration. The 3D seismic experiments confirm that in a hardrock environment, massive sulfide bodies cause a characteristic seismic scattering response (Milkereit and Eaton, 1998). This provides an excellent basis for the direct detection of deep massive sulfides in the crystalline crust by high-resolution seismic methods. We review the physical rock properties of massive sulfides, discuss the seismic reflection response of deep seated ore bodies, and present borehole geophysical data from the site selection process. Examples will illustrate various stages of the ambitious hardrock exploration project: from survey design to heliportable data acquisition and from special processing considerations to final interpretation of the 3D seismic data volume. Special attention must be paid to improve static corrections, in order to enhance images of seismic scatterers. In addition, velocity and density logs must be obtained for detailed interpretation of 3D seismic images. A comprehensive calibration and ground truthing program has been initiated to further guide the development of this new exploration technology for the crystalline crust.
Milkereit B & Eaton D, Tectonophysics, 286, 5-18, (1998).
The use of underground space for infrastructures and the exploitation of the mineral resources require the control of realities of the physical environment. To envisage the in-depth extension of the structures and the geological phenomena in order to reduce uncertainties is thus the challenge with which is confronted each day the community of Earth's Sciences. The Géofrance 3D program, dedicated to the knowledge of the Lithosphere was defined to achieve this goal. It is also a framework research topic, implementing the most advanced methods in each discipline. This program targeted to become a reference for storage, availability and 3D valorization of the data geological and geophysics, gathered or treated within the framework of the projects. The final objective of the program is to realize a "3D observatory" which will gather the means of scanning of the earth's crust. GéoFrance 3D envisages:- to collect the existing data on the ground and the underground of the french territory; - to carry out new exploration by deploying the measuring instruments geophysics necessary to the imagery of the underground;- to constitute multi-source data bases and to provide the means of calculation necessary to the implementation of 3D models and methods of generalized inversions; - to produce representations in maps and 3D solids volumes of underground objects;- to diffuse, most largely possible, these data and tools.
We propose to describe how to store and give access to the scientific community of the GéoFrance 3D program the geological data, geophysical data... For that it is necessary to provide an environment for 2D and 3D modeling and data management fulfilling the requirements of professionals of geology and geophysics. This environment must be accessible using the Internet network and provide the mechanisms for:- input news data;- consult catalogues of available data with geographical requests and sets of themes;- download these data on workstations and use then in the 3D modeling applications.
A pilot application was developed from the data of the Armor project in spring 1997. The geological and principal geophysical data were prepared in order to be able to proceed to requests sets of themes and geographical. It was proven that in operational phase, the repatriation of the data was possible. We will describe the results obtained in this phase of development, which allowed the passage of a prototype to a public application. We will also point out the link, which exists between our project and the European project, GEIXS (Geological Electronic Information Exchange System) whose objective is to give free access to the public catalogues of digital data available in the 15 European geological services. After having described the whole of the problems to be solved, we will see the architecture choices, by using the last data-processing techniques. These choices were done in order to return an easy access to the data (by integrating management rules) by consultation sets of themes and/or geographical, then we will detail examples of use of the proposed solution. Lastly, we will present how all these results are integrated in a concept of 3D modeling of the geological objects.
The production of a geological model 3D, is the realization of a model among an infinite whole of models possible. Our objective is to describe a method making it possible to provide a description of the space of the 3D geological models by using the inverse problem. To provide a geological 3D probabilistic model, we will use the inverse problem, applied to a whole of various data: geology, gravity, magnetism, petrophysics properties.
We will define this step like a total litho-inversion. We will present initially the information contained in a geological model then we will specify this one by showing that the topological contents are of primary importance. We will see that this topology rises from a geological assumption, that this perhaps precise topology but that the geometry of the model remains often undetermined. While respecting topology and in consequence the a priori geological assumption, we will propose a method allowing to walk around the possible whole of the geometries and taking in account for the geophysical and geological data.
Description in 3D of the geological objects is essential in the activities of the Earth sciences applied to the mining exploration, the regional land-use and the 3D cartography. To this purpose, we propose to develop a programmable method to obtain a 3D probabilistic description of the geological objects, which take into account the whole of the data available like: geological map, borehole data, specific observations of structural geology, geophysical measurements, petrophysic properties measurements.
The difficulty lies in the developpement of a method making possible to walk around the whole of the possible models to obtain a representative distribution law of the models. The method suggested here allows us to obtain such a law with reasonable calculating times with a workstation. Indeed, for a geological model composed of twenty seven formations represented by a grid of 60 000 cells and in the presence of a gravity and magnetic field of 1000 points of measurements, we obtain a description of the whole of the possible models in 9 hours of calculation.
After having posed the problem, we will quickly describe the construction method of the 3D models then, we will clarify the method adopted to realize the inversion, namely an algorithm of Monte-Carlo (Metropolis) adapted to the problem. Lastly, we will apply the method to the geological model of the Cadomian belt in Armor.
In spite of their success for modeling simple surfaces, classical automatic mapping systems are unable to model complex surfaces and, more generally, complex geological regions of the subsurface affected by severe tectonic events with overturned faults, salt domes and reverse faults. Similarly, experiences of using classical Computer Aided Design methods developed for the car industry indicated that they can produce nice abstract models but they are unable to fit all the complex and imprecise data encountered in real applications in the realm of geosciences.
In order to address the modeling problems specific to the geometry and the properties of complex geological structures, a completely new mathematical approach based on discrete modeling was proposed in the frame of the gOcad research project. In this discrete modeling approach, the geometry of any geological object is defined by a set of nodes (points) in 3D space while its topology is modeled by links bridging these nodes.
If the object to be modeled is a geological surface (horizon or fault), then the links are arranged in such a way that the mesh so defined generates triangular or polygonal facets. It is easy to imagine that this discrete modeling approach presented for surfaces can be extended in a straightforward way to the modeling of curves and volumes. Moreover, the nodes of the models can not only be used for holding the (x,y,z) coordinates defining the geometry at this point but they can also store the physical properties at this location if any.
From a practical point of view, the above discrete proposed approach is without any interest if we do not have a powerful mathematical tool able to interpolate both the geometry (x,y,z) and the properties of the nodes defining the geological objects in the 3D space. For this reason, a new method called Discrete Smooth Interpolation and abbreviated DSI is proposed for modeling natural objects while taking into account a wide range of complex and more or less precise data.
In the framework of this paper, the basic principles of this discrete modeling approach and some applications will be presented.
In the frame of GeoFrance 3D program, geologists are requested to produce 3D models i.e a solid description of relevant geological formations.
Any 3D geological model relies on georefererenced maps and cross-sections which reflect the state of the interpretation at a given moment. These can change during the 3D modeling process, or new ones can be added. Therefore an efficient modeling method must: (1) represent data in 3D (maps and sections); (2) allow edition/modification of maps and sections in their proper space (2D edition; (3) insure the consistency of intersecting maps and sections; (4) provide facilities for 3d Solid Modeling.
We develop a software that fullfils those requirements and provides basic modeling facilities which can evoluate according to continuously improving modeling methods.
Section edition is based upon the use of parametric surfaces so that the 2D-3D relations can be efficiently handled. In sections the user draws the geological formations boundaries that may be uncomplete. in order to insure the 3D consitency of all the sections (i.e. only one geological formation can exist at a given point of space) a 3D topology is built (Boundary Representation Model).
Structural data (the dip - azimuth of main anisotropy of the formations) can be equally entered in sections. Those structural data are used to interpolate 3D contacts known at some points. This helps to draw or extend the data boundaries in all the sections of the model.
Based on entered data in sections and maps, a 3D solid model of all geological formations can be automatically built using Voronoi diagrams. Starting from a generated set of points issued from data this method computes a partition of space according to the nearest neighbour. Built solids are topologically closed and by construction they share common boundaries. Since the solid construction is automatic, it is expected that the proposed solution allows multiple cycles of interpretation-construction until the model is statisfactory. This is illustrated by examples from Geofrance-3D program.
This methodlogy have been proven to be more efficient than using off-the-shelf CAD/CAM tools in several modelization processes. In addition, the proposed solution includes functionalities to make benefit from existing work in 2D and 3D as well as connection with information system infrastructure.
This presentation illustrates the interest of combining geophysical and solid volumic modeling to converge to satisfying geometrical models.
Initial data encountered within the framework of GeoFrance3D program mainly consist in (i) geological maps and (ii) cross-sections calibrated onto geological observations and available geophysical constraints (gravity, magnetism, seismic ...). In order to reconstruct the 3D geological volumes, we use a method based on Voronoi diagrams which enables all volumes to be automatically constructed starting from a set of points derived from the geological map and cross-sections. The method provides a partition of space into regions according to geological formations. This partition is very dependant to the initial distribution of the set of points. However, it is used to automatically provide a best order solution that is in most cases topologically correct i.e. homotopic to the actual geological model. The advantage of using Voronoi method compared to surfacic one is (i) the volumic character which allows geophysical computations to be performed, (ii) the automatic character allowing to test several hypothesis. From this 3D volumic model, simulation of geophysical effect can be computed based on the assumed physical properties of the medium. This computed contribution is then compared to the observed one. Discrepancies are analysed in order to discuss the geological uncertainties and eventually to converge to the best satisfying model.
However, a remaining problem concerns how and where to interact on the initial sections to reduce discrepancies. Some solutions are explored using Monte-Carlo inverse method to randomly walk around the model space solution.
This method has been applied to Armor and Massif Central projects (GeoFrance3D program) using gravity and seismic simulations. It appears to be a powerful tool for 3D model validations.
As part of a larger study pursuing the use of earthquake data instructural model building, we examined the 3D geometry of the source fault of the 1994 Northridge Earthquake, as outlined by after shocks,using Gocad. Gocad is a powerful tool for 3D model building from data as well as viewing and manipulating spatial and property information. It allowed us to get a clear 3D view of the aftershocks distribution and to interpolate it to generate fault surfaces.
We used about 8,000 after shocks (SCEC database). Before loading them into Gocad, the after shocks hypocenters were clustered using a computer algorithm by Jones and Stewart (1997). We also imported a set of focal mechanisms by Seeber (SCEC database), to be displayed together with the after shocks. The surfaces were generated from the clustered after shock set.
The Northridge Earthquake source fault shows a well marked slip-parallel step near its SE margin, and this same feature can be recognized from the nodal planes. The fault strike changes by 10° from the SE section to the NW section. The two sections are joined by a segment striking 95°. The geometry at shallow depths is more complex and there are indications that either the main fault splits into several low-angle splays, or it is truncated by low-angle faults at a depth of about 3 km. Other faults can be outlined in the area, though for some of them there are not enough data available to reliably determine the geometry. However, the main fault is bounded to the NW by a surface dipping 45° to the North. West of the main fault there is a small fault, possibly strike-slip, dipping 70° to the NE. Finally we projected the slip model by Wald et al. (1996) onto the fault surface. The rupture appears to initiate at the step, and then it propagates to the NW section.
Jones RH & Stewart RC, J G R, 102, 8245-8254, (1997).
Wald DJ, Heaton HT & Hudnut KW, Bull. Seis. Soc. Am, 86, S49-S70, (1996).
This project is aimed on the researching of the methodology of interactive structural 3D modelling. Here we present our attempt of using the geoinformation systems and computer aided design systems for structural 3D modelling of such geological object as a real deposit. Several approaches for deciding such problems are given.
Our study is realized on the example of Kholba gold ore deposit, which is located in the South-East part of the Eastern Sayan, Buryatia Republic, Russia. Geological structure, tectonics, stratigraphy, metamorphism and geodynamics of this area of the Eastern Sayan are in detail described in two monographs: Belichenko et al. (1988) and Dobretsov et al. (1989). Here we would like only note that polygene and polychron nature of the genesis of the deposit, several stages of its formation history have stipulated a very difficult structure of the object, so it was quite interesting to try the modelling exactly on such difficult object. As source working materials we have used cartographic materials on the object of the study: topographical map of the region of the deposit, geological map, and geological plans of the underground horizons on the different levels of the deposit as well.
The main idea of our modelling is that using lithological and structural heterogeneousnesses of the deposit and resting on the data of the cartography of underground horizons we can interactively reconstruct a real structure and morphology of the deposit in the most full range (on the area limited by available reliable materials). For this, using the facilities of Arc/INFO we realize underground horizons of the deposit as independent layers (coverages), on which the interpolation in bCAD is conducted to achieve 3D model of the deposit. DEM (digital elevation model) of the area of the deposit was created in Arc/INFO and PCI systems. The initial principle model of the ore-bearing horizon of the deposit was created in Surfer. Also it is possible to decide some auxiliary problems of the practical significance such as creating electronic geocoded maps of the surface and underground horizons of the deposit. And then our results will be integrated into one model containing as more as possible different geoscience data.
Belichenko VG et al., Geology and metamorphism of Eastern Sayan, 1988
Dobretsov NL et al., Geology and ore genesis of Eastern Sayan, 1989
Investigations of fractal geological objects (most of them are 3D bodies) as a rule are based on their 1D - 2D sections, or some indirect data (as in seismology). In fact this is true for all geological objects that are more than 1 m in size. Such investigations, however, do not provide a strong basis for calculation of fractal dimensions of these 3D bodies. Other source of misinterpretation is the indirect data (as in seismology).
This study is based on the data from Kola Banded Iron Formation (BIF). The Kola BIF is situated in the northern part of the Baltic shield, in the Kola peninsula. An extremely thorough geological and geophysical survey of the Kirovogorskoye BIF deposit resulted in such a great amount of data (30000 points) that there is a possibility to determine 3D fractal dimension of the BIF bodies. Calculations made by an original programme gave the dimension value of 2.14.
What can it mean? It is unlikely that fractal ore structures could have formed by a sedimentary process; I believe, they indicate a metamorphic genesis of the BIF. Such fractal structures correspond to the model of diffusion limited aggregation: initial fluctuations of iron content appeared to be cluster aggregation centers, which controlled subsequent ore deposition. Processes of this kind are described by systems of parabolic differential equations. Solutions to these equations can be illustrated as a trajectory in the phase space. Using the canonical procedure of Takens, it is possible to correlate the phase trajectory with a long series of values of one relevant variable. The plot of this variable can be studied as a fractal by classical Takens-Grassberger-Procaccia algorithm. The analysis of 100 BIF's magnetic distributions (drill-hole logging data) showed that spatial variations of magnetite are an exhibition of the deterministic time history of the ore systems. However, as it was shown (Jedyak et al., 1994), a reliable identification of the process would require a series of data N=42D. With this restriction, in this study the initial massif was obtained by combining the drill-hole logging data on magnetite distribution at several sections of the ore body. It is established that inclinations of the D=f(d) plots defining a fractal dimension of attractors D=2.05-2.3. Digital experiments in modelling of re-distribution of components, according to the model system of 8 variables, show that there is a possibility that structures analogous to BIF bodies could be formed. This testifies that fractal structures of the Kola BIF's are not a random coincidence, but an evidence of functioning of a metamorphic-self-organization system.
Jedyak A., Bach M., & Timmer J., Physical Review E, 50, 1770-1780, (1994).
South Australia is currently undergoing intra-plate deformation in the Flinders and Mount Lofty Ranges as evidenced by active seismicity. Several models have been proposed in order to reconcile traditional rigid plate tectonic theory with the occurrence of intra-plate deformation (e.g. Campbell, 1978, Liu and Zoback, 1997). These models include lithospheric weakening due to changes in the compositional and/or thermal state of the lithosphere (e.g. Liu and Zoback, 1997) and stress amplification due to density inhomogenities (e.g. Campbell, 1978). Knowledge of the mechanical properties and stress state of the lithosphere is critical for developing a robust model of intra-plate deformation. In order to understand the cause of and control on the intra-plate deformation occurring in South Australia, these two properties are being analysed. Prior to this project the South Australian stress field was poorly constrained, with existing information summarized in Denham and Windsor (1991). Hillis et al. (1998) presented more in situ stress data for Australia as a whole, including South Australia. The majority of stress indicators for South Australia are located within sedimentary basins where active petroleum exploration is being undertaken, namely the Cooper-Eromanga, Otway and Duntroon Basins. The Cooper-Eromanga Basin shows a broad E-W orientation of SHmax, while both the Otway and Duntroon Basins (Messent and Yacopetti, 1997) display a NW-SE orientation of SHmax. As well as confirming these orientations, our results have placed constraints on the magnitudes of the three principal stresses, showing Shmin<SV>=SHmax. Hence the South Australian crust appears to be in a strike slip fault regime, on the boundary of strike slip and normal faulting.
The strength of the South Australian lithosphere has been measured in terms of it effective elastic thickness (Te), using a technique described by Forsyth (1985). South Australia displays Te ranging from 36 to 96 km, corresponding to moderately weak to extremely strong lithosphere. A general trend of decreasing Te from west to east is observed, however the northern Flinders Ranges displayed the largest Te value and greatest error. The southern Flinders and entire Mount Lofty Ranges were found to have a relatively low Te. The observed variations in Te are consistent with those expected for the varying geological settings, which range from a stable cratonic region to an actively deforming region.
These two datasets provide a powerful tool for modeling the intra-plate deformation occurring in the Flinders and Mount Lofty Ranges. This may in turn provide further insight into other regions of intra-plate deformation, both past and present.
Campbell DL, Geophyics Research Letters, 5, 477-479, (1978).
Denham D, Windsor CR, Exploration Geophysics, 22, 101-106, (1991).
Hillis RR, Meyer JJ, Reynolds SD, Exploration Geophysics, In Press
Liu L, Zoback MD, Tectonics, 16, 585-595, (1997).
Messent BEJ, Yacopetti CM, APEA Journal, 301-314, (1997).
Forsyth DW, Journal of Geophysical Research, 90, 12,623-12,632, (1985).
The HDR (Hot Dry Rock) project, developed on the Soultz geothermal site (Alsace, France), is aimed to extract energy from a thermal doublet. The microseismicity data induced during the stimulation tests on 1993, 1994, 1995, 1996 have been reinterpreted. They indicate that the permeability of the reservoir increases from 3x10-17 to 10-15 m2 during 1993 to 1996. These values are in excellent agreement with the upper limits of the permeability estimates from thehydraulic experiments at Soultz. From the 9400 events recorded during the 1993 test, the permeability tensor K is then be estimated. The largest component of K indicates that fractures are induced mainly in the vertical direction, while its horizontal principal axes are oriented according to N130 and N40, respectively. Previous work by Schild and co-authors report similar directions of the two maximal-density systems of healed microscopic cracks in quartz. They suggest that the N130 and N40 direction may correspond to the paleostress directions probably resulting from superposition of the internal cooling- and external tectonic stresses. It seems to be that the same directions are reactivated during the stimulation tests, indicating that paleo healed cracks might have plaid an important role during hydraulic fracturation. The travel time and evolution of the temperature of fluids between the injection and the pumping wells is then deduced from a 3D modeling of the fractured reservoir using an equivalent porous medium approach. One of the most important parameter which governs the fluid circulation within the reservoir is the effective porosity. Indirect methods have been used to estimate this parameter using the tracer monitoring tests. Depending on the hypothesis assumed for fracture opening, porosity values ranging from 0.01 to 0.3% have been estimated. The travel time between the injection and pumping wells is then estimated at 78 h for a flow rate at 18 l/s. This order of magnitude is in good agreement with the tracer monitoring tests which indicate a first income of the tracer after 72 h. The model gives over-pressure between the injection and pumping wells at 3 Mpa, the macro mean permeability of the reservoir being 10-15 m2. 3D numerical modeling techniques have been used to characterize the thermal state of the upper crust around the Soultz geothermal site before and after exploitation.
A 3D stratigraphic database has been compiled for an intracratonic flexural basin, the Paris Basin, in order to better constrain the relationships between tectonic controls - and their different wave-lengths - and the stratigraphic record. Around 1100 wells have been correlated using the principles of high-resolution sequence stratigraphy. These data were interpolated using a kriging method in order to build a 3D-solid model. By this way layers can be separately visualized so that layers relationships and the basin dynamic evolution can be illustrated.
Long term subsidence, from base Trias to lower Pleistocene, is the thermal consequence of a Permian extension. From Middle Quaternary, the Paris Basin is uplifting.
The Paris basin records the main west-european tectonic events - consequences of the Tethys and Atlantic evolutions - which perturbate this long term subsidence. This database makes possible a quantitative appraisal of the cinematic, wave-length and amplitude of these movements.
These main events occur during Norian (Upper Trias) and Trias/Jurassic boundary (early Cimmerian unconformities), Aalenian (base middle Jurassic, mid-Cimmerian unconformity), early Cretaceous (late Cimmerian disconformities), late Aptian (lower Cretaceous, "Austrian phase"), late Turonian (upper Cretaceous), late Cretaceous and multiple Triassic one. They record an evolution from a subsident extensional (Trias-Jurassic) step to a non subsident compressional (Tertiary) period. On 3D geometrical data, Triasic to Tertiary unconformities are medium wave-length (multiple of 100 km), low amplitude (few tens of metres) - Triassic, Jurassic - to short wave-length (multiple of 10 km), high amplitude (few hundreds of metres) - Upper Cretaceous to Miocene. 3D data leads to a quantification of the tectonic displacements and the associated erosion.
A 5-month passive seismological experiment was conducted in 1996 in the southwestern Alps within the framework of the Geofrance 3D project. The quality of this dataset (10 km interstation spacing, 350 local earthquakes selected) provides a unique opportunity to compute high-resolution 3D images of the crustal structure from traveltime inversion. To improve the resolution at depth, we included in the dataset a hundred of deeper earthquakes (between 15 and 110 km) recorded by the permanent network of DISTER Genova. We use the technique of Thurber (1983) which inverts P and S-P traveltimessimultaneously for velocities and event locations. The most striking feature of the tomographic images is the high velocity anomaly associated with the Ivrea body. Starting at 8 to 10 km depth under the Dora Maira Massif, the anomaly (Vp > 7 km/s) is very narrow (< 10 km) and NS elongated. To the south, it stops abruptly under the surface trace of the Penninic front along the northwestern boundary of the Argentera Massif. At greater depths (20 km) it appears to shift to the east under the Po plain. Because of the lack of deep earthquakes under the french side of the study region, we bring no constraint on the possible extension of the deep Ivrea body under the brianconnais zone. At shallower depths between 3 and 8 km, we find another high velocity anomaly which could be associated with the MonViso ophiolites. Between 0 and 5 depth, the Digne nappes appear as a low velocity anomaly. We will show how the use of 3D visualization techniques presenting the velocity model in combination with the relocated hypocenters can improve our understanding of the crustal structure of the southwestern Alps and the relationship between the structure and the present dynamics of the chain.
The external sector of the northern Apennines across Marche and Romagna regions, on and off shore, were studied with the integration of a detailed field mapping and the analysis of seismic reflection profiles and composite well logs derived from the hydrocarbon exploration. Three geological cross sections were balanced and restored in order to constrain the geometry of the major structures and calculate the amount of deformation.
Then an advanced computing package as 3DMove™ by Midland Valley Exploration Ltd has been used in order to construct a 3D model from the already balanced cross sections. This software, among the other things, allows the importing 2D cross-sections, rescaling and positioning on a digitised base map. Once having built the net of sections, it is possible to correlate the lines by different interpolations.
The construction of the model has been developed via the simplest way, i.e. connecting horizons and faults lines of the three cross-sections by surfaces giving a first block-diagram, rough but really useful in realising that some problems among the parallel balanced sections were still unsolved. For example, the along-strike relationships of the thrust related structures (folds and faults) are not explained by the already balanced 2D cross sections and structural maps do it only for separate levels or horizons. Some system is made with a quite simple pop-up structure. This is constituted by a thrust fault, detaching the sedimentary cover at the top of the basement, and an associated back thrust with a number of splays. Some other sistem is related also to reactivated basement faults in a mixed thin/thick skin tectonics.
In the interpretative part of the model, especially when diverging opinions are strongly debated such as the involvement of the basement in the thrusting, the third dimension results useful in demonstrating the hypothesis.
The internal western Alps currently undergo a moderate seismicity along two main arcs: the "Piedmont and Briançonnais seismic arcs". This upper-crust seismicity appears mainly extensive in the internal western Alps as a whole. The current stress field in this region is inferred from inversion of sets of focal solutions and shows extensive axes radial with respect to the alpine arc. The 3D geometry of the seismic arcs strongly suggest that their development is driven by inherited crustal structures. The geometry of the Briançonnais seismic arc that follows the arcuate geometry of the belt, and the radial extensive stress field indicate that they could correspond, at the belt scale, to the re-activation of the crustal penninic front, a major Oligocene crustal thrust between internal and external units. At local scale, the seismicity of the Briançonnais arc, to the SE of the Pelvoux external crystalline massif, follows fairly well a late-alpine fault network studied from a structural point of view. The corresponding hypocenter depths range mainly from 0 to 10 km. Thus, several faults have been shown to be active. The late-alpine fault network is linked to the crustal penninic front, and could branch with it at depth. We propose that the crustal penninic front is currently re-activated as a crustal extensive detachment. The Piedmont seismic arc presents the same extensive tectonics but with deeper hypocenters (5 - 20 km) and straighter shape. The southern part of the Piedmont seismic arc is a very active vertical straight alignment of earthquakes. This seismic area as a whole corresponds to the western side of the Ivrea body, as shown by the superposition of seismicity and Bouguer anomaly in this region. The Ivrea body is a mantle slice which indents the upper-crust and is an evidence of important shortening in the western Alps. Thus, the 3D modelling of the western alpine crust allows us to establish that the present-day activity of this part of the belt is driven by inherited crustal shortening structures re-activated in an extensive current tectonic context.
This airborne gravity survey was realized in the frame of the «GEOFRANCE 3D Alpes» project in the Western Alps, from the Rhone Valley to the Pô-Plain in Italy. Four objectives were followed: accurate airborne measurements of the alpine gravity field, combination of both GPS data and those provided by a classical inertial navigational system (INS), comparison between airborne gavity data and upward prolongated ground data, 3D modelling of the alpine crust using several Bouguer anomaly maps.
In February 1998, 18 NS lines were flown, crossed by 16 EW lines, spaced respectively by 10 and 20 km. The survey was performed with a Twin-Hotter aircraft owned by the Swiss Federal Directorate of Cadastral Surveying at barometric altitude of 5100 m above sea level. The aircraft was equipped with 5 GPS receivers (working frequencies from 1 to 8 Hz) for relative positionning to 7 ground fixed GPS stations in Europa and for monitoring both aircraft velocities and accelerations. Gravity field was measured with a L&R airborne gravimeter, type SA, mounted on a laser gyro stabilized platform (working frequencies: 1 Hz for the spring tension, 10 Hz for the beam position).
Various strategies for computing the kinematic properties of the aircraft (softwares, choice of the reference GPS ground station) are discussed. A preliminary long-wavalength "free-air" anomaly map is presented and compared with some previous published.
Three dimensional geometric modeling has been performed on the external crystalline massifs of the western Alps (Massifs de Belledonne and Grand-Chatelard). This zone corresponds to the internal part of the European passive margin developped during the liasic rifting and thrusted during the Miocene compression (tectonic inversion).
The contact between basement and cover is representative of the massif geometry and has been modelized. 3D geometrical model has been build up using 30 cross-sections (NW-SE, NE-SW) and geological outcrops projected on the Digital Elevation Model (DEM) of the area, taking into account local dip measurements.
Three interfaces are shown in the model which correspond to identified geological events: (i) the unconformity of triasic series over the metamorphic Paleozoic basement, supposed to be horizontal in the beginning of the Alpine cycle, (ii) the main stage of the passive margin development (Lias, Carixian) marked by ENE-WSW extensional faults, (iii) the main compressive stage of the Alpine collision (Miocene) marked by steep NNE-SSW trending faults.A kinematic study has been performed for each of these structures. The Miocene deformation, highly heterogeneous, results into networks of anastomosed brittle-ductile reverse shear zones. It corresponds to a WNW-ESE compression associated to an important vertical stretching on the two studied massifs. At some places, eastern border of the Massif de Belledonne is back thrusting to SE showing a dextral main component.
This tectonic style is characteristic of the deformation of the eastern European margin during its involvment into the Alpine collision and very similar to the deformation of other external crystalline massifs as the Aar. Thus, 3D geometric modeling reveals the overall anastomosed shape of the main faults, ondulating around crustal scale boudins vertically stretched, made of crystalline basement. Former structures, as the liasic extensional faults of the Massif du Chatelard, showing spectacular syntectonic sedimentation and hydrothermal alteration at the top of a granitic massif, are preserved within low strain zones located at the core of these boudins.
A 3D reconstruction of the north Limousin crustal block (90 km x 90 km x 6 km) has been performed using the Gocad modeler. The aim of this project is to better understand, at the regional scale, the relationships between geological structures and ore deposits (Rare Metals, Au, Sn, W and U) associated with Variscan granites and during the 300±30 Ma time period. The data used in the modelling are : revised 1/50 000 geological and structural maps of the area, 10 interpreted vertical cross sections, 3D gravimetric models of granites, whole rock and stream sediment geochemical data, age determinations mainly on granites and the BRGM data bank on mineral occurences.
All the granites appears as relatively thin slabs with multiple deeper roots. The Gueret biotite-cordierite granite is very thin (0.5 to 2 km). The Saint Sylvestre biotite-muscovite complex is a 1 to 3 km thick laccolith with numerous deeper zones generally corresponding to late intrusions. The complex extends below enclosing metamorphic rocks. The isodepth map highlights N20 and N120 regional structures controlling the granite emplacement. The Western Marche complex is composed of Gueret granites injected by Limousin type granites. Gravimetric inversion shows that Gueret granite outcrops are associated with thinner zones (1.5 km) whereas Limousin granite outcrops correspond to thicker zones (about 3 km), up to an average of 3.5 km to the NE (Saint Sulpice). Nevertheless, deeper zones observed below some Gueret granite outcrops, are interpreted, taking into account the low density contrast between Gueret granite and surrounding rocks, as possible occurences of hidden leucogranitic bodies. Two series of deeper zones oriented EW to N120 are observed : one south of the Marche fault and another one north of the Arrenes-Ouzilly fault to the south. The Blond massif appears as a 1.5 km thick slab in average with a thickness increasing from the north to the south and SE. The northern contact is vertical whereas the southern one is low deeping, extending below the Cieux granite. The granite seems also to extend under cover beyond the Oradour fault to the NW.
The superimposition of inverted gravimetric data and ore deposit occurences shows that U deposits are located in the vicinity deeper root zones along a N20 trend in the Saint Sylvestre complex and along an EW trend in the Western Marche massif corresponding to late specialised magmatic bodies. Some hidden cupolas may represent interesting exploration targets. Inversely, W and Au mineralisation are mainly observed in the Lower Gneiss Unit around the granites.
The Châtaigneraie area is located on the southern border of the French Massif Central, between the western end of the Margeride granite to the East, and the Sillon Houiller to the West. This area is characterized by W-As mineralizations and an important granitic magmatism. From petrographic observations and 40Ar/39Ar dating (Monié et al., this vol.), two generations of granites are known. The older ones are porphyritic granites of minimum age of 315 Ma. They are, from West to East, the Marcolès, Veinazès, Soulaques plutons and the Entraygues massifs which correspond to the western end of the Margeride laccolith. The younger granites are leucogranitic dykes and sills of 305 Ma age widespread in the Châtaigneraie area. Plutons and mineralizations host rock consist of paragneiss and micaschist divided into two structural units, namely the Lower Gneiss Unit resting on the Para-Autochtonous Unit. Porphyritic plutons cross-cut the nappe contact herited from the Hercynian crustal thickening (between 350 and 335 Ma). Their intrusions is responsible for contact metamorphism. All the W-As mineralizations of the study area are located in the spotted schists of the contact metamorphism aureole showing a spatial link between granite and mineralizations. On the other hand, 40Ar/39Ar dating and isotopic geochemistry (Lerouge et al., this vol.) show that the W-As hydrothermal fluids are, in fact, related to the 305 Ma leucogranitic magmatism. 3D modelling of the plutons and metamorphic units constrain by structural measurements and gravimetric modelling was performed in order to better understand the close spatial relationship between porphyritic granites and mineralizations. Contact metamorphism is not only limited to the outcropping plutons margins because small areas of spotted schists, hosting the W Enguiales deposits, are observed between the Entraygues and Veinazès granites. These spots of contact metamorphism were interpreted as evidences for a hidden granitic body at depth. The lack of negative gravity anomaly centered on the spotted schists and gravimetric modelling are inconsistent with this interpretation. Therefore, no granitic body can be located under these contact metamorphism spots, then Veinazès and Entraygues massifs are not be linked at depth. We interpret the small areas of spotted schists as the result of contact metamorphism due to the floor of a larger partly eroded pluton (i.e the Margeride laccolith) from which the Veinazès and Entraygues appendix are relics which could represent the root zones. According to this interpretation, W-As bearing fluids associated with the 305 Ma leucogranitic magmatism were trapped under the floor of a the larger pluton (i.e the Margeride laccolith).
The Blond peraluminous leucogranitic pluton (315±5 Ma) intrudes the Upper Gneiss Unit of Limousin to the North and the Cieux-Vaulry biotite-cordierite granite (352±12 Ma) to the South during late Variscan time. The pluton is bounded to the West by the N130°E Oradour fault. It shows an early leucogranitic pluton made up of 4 main non concentric and discontinuous units elongated E-W, followed by an ongonitic unit with Li-muscovite and topaz leucogranitic bodies scattered to the North-West of the pluton.
Bouguer anomaly satisfactorily individualize the pluton. Near the Oradour fault, local change in direction of iso-anomalies suggests a dextral shear sense of this fault. After computation of the shape of the pluton's floor by 3-D gravity data inversion, the massif appears as a sill deepening to the SE and separated in two domains. The southern domain, 2 to 4 km in width, is the deepest. It is NE-SW trending, oblique to the E-W maximum lengthening of the pluton. To the south, the Blond leucogranite extends about about 2 km under the Cieux-Vaulry biotite ±cordierite granite. Maximum depths occurs in two funnels near the Oradour fault. They coud represent the main root-zones of the massif. The northern domain shows a laccolithic shape (<2 km) with several funnels (3 km deep) lined up according a N95-100°E trend near the northern limit of the pluton. Some of these zones which are correlated with outcroping differentiated ongonitic leucogranites coud attest to the presence of feecder zones for such late injections.
Each petrographic subtypes has been deformed before complete crystallisation. Magmatic fabrics belong to two main orientation families. The first family shows subhorizontal foliations to the NW and the center of the pluton. Elsewhere, foliations are E-W in strike and dip moderately to the SE. These planar structures form a convex dome E-W trending. Local changes in direction near the Oradour fault indicates the dextral shear sense of this fault. Lineations have several directions and a shallow plunge outwards of the pluton. The second family forms conspicuous subvertical NE-SW corridors, hectometers in width. The foliation trajectories underline the sigmoidal transitions occuring from the first family to the second one in accordance with a senestral shearing.
Considering geophysical and structural data, the geometry of the Blond pluton appears as a slab, rooted to the SE, and spreading out to the NNW. Syncinematic ascent of magma is constrained by the dextral shear of the Oradour fault which induce a set of combined ENE-WSW left-lateral late-magmatic shear zones controlling the distribution of late plutonic injections.
The 3D model of the coal basin of Alès (Massif Central) is built from data produced by the collaboration between BRGM and HBCM. These used data are: bibliography, DEM, 27 geological cross-sections from the mining data, field geology validating the existence of the superficial structures, geomety of the known or exploited coal layers, faults built from the exploitations data, bottom topography of the basin basement from deep wells and galeries.
The 3D geometric model was built with STRIM (a 3D mechanical modeller from MatraDatavison). This modeller provides tools to construct parametric surfaces from geometric data. In this case, the surfaces were created from two main kinds of data (1) digitized cross-sections including topographic profiles, traces of faults, coal layers, top and bottom of the basin and (2) the isohypses of the basement for the north part which was not covered by the cross-sections.
The main geological interest of this 3D reconstitution is to show the geometric and kinematic relations between the basin and the substratum.- to establish the geometric relations between the folded coal layers and the faults to determine the elementary cinematic of deformation- to hierachise the internal structures of the basin- to discuss the rooting of some faults in the substratum compared with the faults which affect only the sedimentary series-to propose a genetic model of the basin which takes into account the main regional faults which affect the substratum- to compare this cinematic with the conclusions get from the young deformations of the Stephanian formations which determine the various tectono-sedimentary styles of the basin.
This preliminary 3D geometric model is the first step to improve the knowledge of the Cévennes metallogeny (auriferous conglomerate of Bulidou, iron mines, Palmesalade mine) and it will be also the base to propose a genetic model of the basin.
Intraplate main structural units of the Iberian Plate (Spanish Central System and Tertiary Basins) results from a mid-Tertiary compressive regime, caused by the convergence of African and European Plates. Crustal structure of this zone has been studied by means of spectral analysis of gravity maps. A gravity survey has been carried out and a Gravity map has been produced. Bidimensional spectral analysis of the data allows us to obtain the mean depth of the sources contributing to the total observed field. From the plot of the natural logarithm versus the radial frequency, two major discontinuities are recorded. The deeper one, located to a mean depth of 28 ± 5 km, corresponds to the crust-mantle boundary, in good agreement with seismic data (Suriñach & Vegas, 1988), and the shallower has a mean depth of 7 ± 0.4 km, representing the upper crust-middle crust transition. In order to isolate the regional and residual sources, specific filters has been designed (Gupta & Ramani, 1980). Regional gravity map shows a relative low which expands in a NW-SE direction under the Spanish Central System and surrounding areas of Duero and Tajo Basins. Low and high anomalies distribution in residual gravity map reflects density changes in the upper crust. Two NE-SW trending strong gradients present in this map are related to the reverse faults which bound the northern and southern limits of the Spanish Central System with the Basins. The inversion method applied (Parker, 1972) permits us to obtain the three-dimensional Moho discontinuity, which is consistent with the two-dimensional models for the area (Tejero et al., 1996). The Moho geometry is characterised by a N 150º E trending wide through, transverse to the chain, reaching depths greater than 33 km. This through is located in the middle part of the chain, running to the basins. To the northeastern and southwestern part, the Moho rise to 30 km deep, which is the average depth value of the Iberian Plate crust. This gravity analysis together with seismic data are used to constrain spatial variations in density and crustal structure, which combined with thermal data, allow us to design a vertical strength distribution for the crust. Its thermal structure is derived from surface heat flow observations assuming heat production and thermal conductivity values for sediments, upper crust, lower crust and lithospheric mantle. The strength profiles show that ductile-brittle transition occurs between 10 and 15 km. Deepest earthquakes reach depths not far from 15 km, which is in good agreement with these results.
Gupta & Ramani, Geophysics, 45(9), 1412-1426, (1980).
Parker, J. R. astr. Soc, 31, 447-455, (1972).
Suriñach & Vegas, Phys. Earth Planet. Inter, 51, 226-234, (1988).
Tejero, Perucha, Rivas and Bergamin, Geogaceta, 20(4), 947-950, (1996).
Little is known about the 3D geometry of shear zones and the depth of their root zones. The Brasiliano/Panafrican orogeny in NE Brazil is characterized by a complex network of continental-scale shear zones developed to adjust the convergence of the West African/São Luís, Amazon and São Francisco/Congo cratons. This Borborema shear zone system, marked by EW trending megalineaments synchronous to NE and NS trending shear zones and metasedimentary belts, is orthogonal to actual brazilian coastline and agree to a geometrical correlation with intracontinental mechanical deformation in Western Africa on Gondwana restitution. We combine geologic-geophysic maps, satellite images and DEM in order to discuss their geometry in the deep crust and interference over the Moho. Remote sensing imagery (Landsat 5-TM and GEMS/X-Band) based on reflectance spectrometry procedures allowed the improvement of geological mapping by lithological discrimination (mainly syn-tectonic subalkaline to alkaline granitoids and HTLP metasediments), geometric and kinematic framework of the strain partitioning (foliations and lineations) characterized by transpressional and trantensional styles. Wavelenghts filtering of magnetic and gravimetric data in the frequency and space-time domains enhanced the mass and susceptibility contrasts among different continental crust-lithospheric mantle levels. Regional and residual geophysical anomalies reflect the Brasiliano fabric framework. This situation is demonstrated by the ductile shear zones control on the geometry of crustal blocks with contrasting geological characteristics (geochemical and geochronological); in the emplacement of syn- to late-tectonic alkaline to subalkaline granitoids of mantle origin; in the granulite facies metamorphism and migmatization. The fit among geophysical signatures and geological features is indicative for mantle rooting of the shear zones. These features indicate that the crustal framework is marked by Moho transposition and uplift of lithospheric mantle and/or mantle derived magmas following the transpressional and transtensional ductile shear zones.
Magnetic modelling along profiles extracted from detailed aeromagnetic surveys are useful to elaborate 3D models of the crust when outcrops and relief are rare in a studied area. These magnetic models need to be constrained by sample measurements. Nevertheless, the question of the representativeness of samples data (small scales and different places) compared with aeromagnetic data (larger scale and global survey) is critical. We have applied a series of scaling statistical treatments (variogram-, fractal-, and wavelet-based methods) to field samples measurements and susceptibility profiles from North Armorican Massif. Thus a procedure for the integration of magnetic data at all scales is proposed.
Despite a wealth of geophysical data available for Iceland, the nature of the Icelandic crust is poorly understood. The long-standing debate, whether the Icelandic crust is thin and hot or thick and cold, focusses on a key problem: What and where is the "Moho" underneath Iceland? Some seismic refraction profiles show reflections interpreted as the "Moho", but on other profiles a similar reflector is absent.
We take a new approach to the problem with an integrated study of teleseismic receiver functions. The HOTSPOT network, a temporary seismic network consisting of 30 broad band seismometers distributed over all Iceland, provided excellent data for this purpose. During the period 7/96-8/98, about 60 big earthquakes (M>5.9) with epicentres 30-90 degrees away were recorded and could be used for the receiver function study.
Receiver functions are sensitive to velocity contrasts beneath the receiver site, such as the Moho or intracrustal reflectors. However, they do not contain information about absolute seismic velocities. Phase velocities of surface waves provide estimates of average S-wave velocities in the crust. In addition, P-wave velocities from seismic refraction studies are available for some areas of Iceland. The joint inversion of these data yields velocity-depth-profiles for the Earth beneath each seismometer site.
The first results of our receiver function analysis indicate considerable lateral variability of crustal structure. Whereas receiver functions from the north-east of Iceland have similar wave forms with weak P-S converted phases, the neovolcanic zone shows a variety of complicated receiver function wave forms with pronounced P-S converted phases. Scattered energy and a strong azimuthal dependence characterizes the receiver functions, for example at Askja.
An integrated interpretation of the results of the receiver function study with other geo scientific data will be performed in future in order to improve our understanding of the regional structure of the Icelandic crust.
We three-dimensionally retro-deformed (i.e. removed in reverse) the effects of the Upper Cretaceous (and later) deformations on a small, unique tectono-metamorphic unit on the western border of the Bohemian Massif (the Zone of Erbendorf-Vohenstrauss - ZEV), using a virtual computer model. Geological constraints for the Upper Cretaceous deformation were fulfilled when the whole ZEV was considered as a hanging wall which had moved 7 km due south above a linked system of faults including a flat detachment at 10 km depth and two orthogonal steep faults (see Tanner et al., 1998, for more details).
As part of the analysis, radiometric K-Ar biotite, hornblende and muscovite isotope ages were placed within the model at nodes corresponding to their three-dimensional sampling locations. Currently, the isotope data show a clear split into Devonian (>360 Ma) and Carboniferous ages (<340 Ma) without variation with depth or geographical location. After the retro-deformation of the Upper Cretaceous event, these data were statistically analysed. In the pre-Upper Cretaceous state, although Devonian and Carboniferous ages still co-exist at all depths, maximum Carboniferous ages decrease linearly with depth. For hornblende and biotite isotope ages these gradients are estimated as 410 m/Ma and 450 m/Ma, respectively. The precision of the hornblende data is limited by the large measurement error, but is independently confirmed by the more accurate biotite data. We interpret the correlation as a fossil, thermal cooling gradient caused by uplift after the Variscan Orogeny. Effectively the retro-deformation bestows the geographically 2D isotopic data a third dimension, thus allowing its three-dimensional analysis.
Tanner DC, Behrmann JH, Oncken O & Weber K, Sp. Pub. Geol. Soc. London, 135, 275-287, (1998).
Located at the southernmost extremity of the Hellenic island arc, the island of Crete is considered as an area of important tectonic deformation and high seismic activity resulting from the collision between the Eurasia and African plates and the subduction of the later under the former.
Using arrival times of P and S waves from 300 microearthquakes, recorded at 10 seismological stations during the period June-December 1995, the 3-D crustal velocity structure of the central part of Crete island was determined.
The method proposed by Thurber (1983) and refined by Eberhart-Phillips (1986) for the inversion of local earthquake arrival times in relatively small areas was applied. The initial velocity model was calculated after rigorous tests using the minimum 1-D model technique (Kissling et al, 1994).
Joint evaluation of the results with other available geophysical and seismotectonic data from the investigated region, reveals that crustal velocity structure is significantly influenced by the tectonic features and the geological regime of the region.
Eberhart-Phillips, D., BSSA, 76(4), 1025-1052, (1986).
Kissling E, Ellsworth WL, Eberhart-Phillips D & Kradolfer U, JGR, 99, 19635-19646, (1994).
Thurber, CH, JGR, 88, 8226-8236, (1983).
A detailed Bouguer gravity map of the western Alps has been produced in the frame of the French GEOFRANCE 3D program. This map is based on new gravity measurements performed in 1997 and 1998.
We assume that the gravity anomalies are mainly due to the upper mantle Ivrea body. Moho depth variation and heterogeneities in the crust of the inner part of the belt. The effect of the Moho depth variations can be estimated using recent three-dimensinoal interface modelling of the Alpine crust-mantle boundary (Waldhauser et al., 1998). A 3D seismic model of the southwestern Alps has been recently obtained by local earthquake tomography (Paul et al., 1998). This seismic model is converted in order to get an a priori density model whose the gravity effect is compared to the Bouguer gravity map. A least square procedure is therefore used to improve the consistency of the seismic and gravity data.
The final density model is discussed in term of deep structure of the inner part of the belt (Mount Viso, Dora Maira, ...).
Index of EUG 10 Volume
Further EUG 10 Information
Index of the Journal of Conference Abstracts
Cambridge Publications Home Page
Last Updated on Wednesday, March 17, 1999.
© 1997 Cambridge Publications