ALBERT

Beschreibung: The more than 1000 km long trans-continental Dead Sea Transform (DST) forms the boundary between the African and Arabian plates in the Middle East. Magnetotelluric (MT) data were recorded at more than 200 sites, focusing on the DST in the Arava valley in Jordan. 2D inversion results of the MT data indicate very clearly that the DST is associated with a strong lateral conductivity contrast. The most prominent feature on the MT image is a conductive half-layer beginning at a depth of approximately 1.5 km, which may be caused by brines in porous sediments. The DST can be identified as a sharp vertical conductivity boundary on the east side of the feature and directly beneath the surface trace. On a coincident high-resolution seismic tomography image of the upper crust, a strong increase of the P wave velocities to values exceeding 5 km/s is observed west of the DST, where the MT model indicates lower conductivities. The seismic velocities are consistent with metamorphic basement rocks; however the observed resistivities (50-250 $Omega m$) are unusually low for unaltered metamorphic rocks. Fractured metamorphic rocks with interconnected fluid bearing veins could explain both the seismic and MT observations. However, the conductivity model suggests furthermore that the DST acts as an impermeable barrier to cross-fault fluid flow. In stark contrast, a prominent flower structure is observed along the central segment of the SAF and interpreted as evidence of pervasive along-fault fluid flow. Here, the high conductivity is attributed to the circulation of saline fluids within the damage zone of the fault system. The width of the conductive zone (0.5 km) is in the same order of magnitude as the width of a seismic low-velocity zone inferred from fault-zone-guided wave observations, while its depth extent (3 km) coincides with the occurrence of a cluster of small earthquakes. It is possible, that the damage zone of the DST is so narrow that it cannot be resolved even with the dense site spacing of the MT experiment. This observation is supported by preliminary results from geological mapping and a seismic study using fault-guided waves that suggest a very narrow low-velocity wave-guide of 3 to 10 m width. The reason for this difference between the DST (very narrow fault zone) and the SAF (wide gouge zone) is not yet clear, but seems to coincide with generally slower slip rates and the relatively low recent seismicity associated with this segment of the DST.

Beschreibung: The Bohemian Massif (BM) is the largest coherent surface exposure of basement rocks in central Europe. It is a geodynamically active part of the Hercynian orogenic belt representing a collage of magmatic arcs and micro-continents caused by the collision of Laurasia (Laurentia-Baltica) and Africa (Gondwana). The general northwest direction of accretion is typical of the northern part of the Hercynian belt. Irregularly-shaped colliding blocks resulted in a very complicated structure of convergence, lithospheric subduction, and crustal shortening, followed by extensional processes and rifting. The western part of the Bohemian Massif is the well-known health and resort landscape of Bohemia, Saxonia, and Bavaria, with Karlovy Vary (Karlsbad) as the flagship of the famous spa towns of the region (Figure 1). Allegedly, the Emperor Charles IV founded the spa in the years 1347–1349 at the site, which was already well known for its hot springs. For centuries, 12 springs in Karlovy Vary ranging in temperatures between 42°C and 72°C have been exploited, especially for the treatment of digestive system disorders and metabolic diseases.

Beschreibung: 〈P〉Detailed analysis of melt and fluid inclusions combined with an electron-microprobe survey of boron-bearing minerals reveals the evolution of boron in a highly evolved peraluminous granite-pegmatite complex and the associated high- and medium temperature ore-forming hydrothermal fluids (Ehrenfriedersdorf, Erzgebirge, Germany). Melt inclusions in granite represent embryonic pegmatite-forming melts containing about 10 wt% H〈SUB〉2〈/SUB〉O and 1.8 wt% B〈SUB〉2〈/SUB〉O〈SUB〉3〈/SUB〉. These melts are also enriched in F, P, and other incompatible elements such as Be, Sn, Rb, and Cs. Ongoing differentiation and volatile enrichment drove the system into a solvus, where two pegmatite-forming melts coexisted. The critical point is at about 712°C, 100 MPa, 20 wt% H〈SUB〉2〈/SUB〉O and 4.1 wt% B〈SUB〉2〈/SUB〉O〈SUB〉3〈/SUB〉. Cooling and concomitant fractional crystallization from 700°C to 500°C induced development of two conjugate melts, a H〈SUB〉2〈/SUB〉O-poor (A-melt) and a H〈SUB〉2〈/SUB〉O-rich melt (B-melt) along the opening solvus. Boron is a major element in both melts and preferentially partitioned into the H〈SUB〉2〈/SUB〉O-rich melt. Temperature-dependent distribution coefficients 〈FONT FACE="Times"〉〈IMG SRC="wdc4:/usr/www/httpd/data/pb4/pg1/formeln/Image16.gif" WIDTH="69" HEIGHT="17" NATURALSIZEFLAG="3" ALIGN="BOTTOM"〉〈/FONT〉〈FONT COLOR="#ff00ff" FACE="Times"〉 〈/FONT〉are 1.3 at 650°C, 1.5 at 600°C, and 1.8 at 500°C. In both melts, boron concentrations decreased during cooling because of exsolution of a boron-rich hypersaline brine throughout the pegmatitic stage. Boromuscovite containing up to 8.5 wt% was another sink for boron at this stage. The end of the melt-dominated pegmatitic stage was attained at a solidus temperature at around 490°C. Fluid inclusions of the hydrothermal stage reveal trapping temperatures of 480°C to 370°C, along with varying densities and highly variable B〈SUB〉2〈/SUB〉O〈SUB〉3 〈/SUB〉contents ranging〈FONT COLOR="#808000"〉 〈/FONT〉from 0.20 to 2.94 wt%. A boiling system evolved, indicating a complex interplay between closed- and open-system behaviour. Pressure switched from lithostatic to hydrostatic and back generating of hydrothermal convection cells where meteoric waters were introduced and mixed with magmatic fluids〈FONT COLOR="#808000"〉.〈/FONT〉 Boron-rich solutions originated from magmatic fluids, whereas boron-depleted fluids were mainly of meteoric origin. This highlights the potential of boron for discriminating fluids of different origin. Tin is continuously enriched during the evolution because tin and boron are cross-linked by formation of boron-, fluorine- and tin-fluorine-bearing complexes and finally deposited within quartz-cassiterite veins during the transition from closed- to open-system behaviour.〈/P〉 〈P〉Boron does not only trace the complex evolution of the Ehrenfriedersdorf complex but exerts, together with H〈SUB〉2〈/SUB〉O, F and P, an important control on the physical and chemical properties of pegmatite-forming melts, and particularly on the formation of a two-melt solvus at low pressure. We discuss this with respect to experimental results on H〈SUB〉2〈/SUB〉O solubility and the critical behaviour of the haplogranite-water system that contained variable concentrations of volatiles.〈/P〉

Beschreibung: The geometry and dynamics of the Mesozoic basins of the Weald-Boulonnais area have been controlled by the distribution of preexisting Variscan structures. The emergent Variscan frontal thrust faults are predominantly E-W oriented in southern England while in northern France they have a largely NW-SE orientation.Extension related to Tethyan and Atlantic opening has reactivated these faults and generated new faults that, together, have conditioned the resultant Mesozoic basin geometries. Jurassic to Cretaceous N-S extension gave the Weald-Boulonnais basin an asymmetric geometry with the greatest subsidence located along its NW margin. Late Cretaceous-Palaeogene N-S oriented Alpine (s.l.) compression inverted the basin and produced an E-W symmetrical anticline associated with many subsidiary anticlines or monoclines and reverse faults. In the Boulonnais extensional and contractional faults that controlled sedimentation and inversion of the Mesozoic basin are examined in the light of new field and reprocessed gravity data to establish possible controls exerted by preexisting Variscan structures.

Beschreibung: Basaltic andesites of the Ongeluk Formation, Transvaal Supergroup in Griqualand West constitute part of a large flood volcanic province that extruded at ca. 2.22 Ga onto the partly submerged Kaapvaal Craton. Pods and short veins of megaquartz, with albite, epidote and traces of calcite and Cu-Ni-Co sulphides are closely associated with beds and pods of jasper and chert that are products of low-temperature seafloor alteration. Characterization of the megaquartz pods and veins yields evidence for an origin in a ''passive'' seafloor alteration regime. Fluid inclusion studies suggest the involvement of two aqueous fluid end members, one NaCl-dominated with salinity similar to modern seawater, the other Ca-dominated and with distinctly elevated salinity. The chemical composition of both fluids is akin to that of seawater, modified in its composition by interaction with the volcanic host rock. The low salinity fluid appears only little affected by fluid-rock interaction processes; the composition of the more saline Ca-rich fluid is more distinctly modified. The chemical composition of the two fluids has important implications for our understanding of the composition of ocean water during the Paleoproterozoic Era. Cl/Br ratios, widely regarded as being conservative in hydrothermal solutions, are significantly below those of present-day seawater, but remarkably similar to that predicted for Archean seawater. This observation suggests that Paleoproterozoic seawater was still buffered by vent fluids, and lacked sufficient organic matter to fractionate Cl from Br.

Beschreibung: We report about Self-Potential (SP) anomalies registered 150 m from the active lava dome of Merapi Volcano, which are associated with Ultra-Long-Period (ULP) seismic signals (periods 〉100 s). During a 5-month period of simultaneous SP and seismological measurements, 50 ULP events were detected seismologically. If SP time series corresponding to ULP events are aligned to the onset times of ULP-events and stacked, the resulting traces show anomalous SP with an amplitude of 5–20 mV/km and 40–60 min duration. In contrast, the anomalous signals were not present in ground temperature data nor in SP data recorded ≈1 km from the summit. SP anomalies associated with ULP-seismicity might be caused by electro-kinetic effect of fluid flow in subsurface. This result is consistent with the ULP-generation model based on seismological observations and adds new information towards the understanding of ULP-seismicity.

Beschreibung: Drilling into modem systems such as the KTB borehole in Germany indicates that the upper crust is pervasively fractured and saturated with surface-derived fluids at hydrostatic pressures. The recognition of surface-derived fluids in palaeohydrothermal systems is limited to the effectiveness of geochemical tracers. At low total fluid fluxes, fluids tend to interact with wall rocks and become isotopically, chemically and thermally equilibrated. At high fluid fluxes, such as found in upper crustal mineralisation systems, fluid inclusion halogen systematics can be used to identify the origin of the fluids, even when the fluids have equilibrated isotopically and thermally with deeper parts of the hydrothermal system. Unlike the modem systems, fluid inclusions in vein fills in both mineralised and unmineralised systems show strong evidence for episodic fluid flow and fluctuating fluid pressures, even in upper crustal rocks.

Beschreibung: As part of Project International Deep Profiling of Tibet and the Himalaya III, a 400-km-long, densely spaced array of 57 broadband and short-period seismic stations was deployed in central Tibet from August 1998 through May 1999. Although originally designed to image the lithosphere with teleseismic events, the array also recorded numerous local and regional seismic events. More than 900 local and regional events were detected on at least 10 stations during the 1-year deployment, and we were able to locate 267 local earthquakes. A substantial number of the events were found to cluster in or near large grabens and along known strike-slip faults, while other events show no obvious correlation with known structures. In addition to spatial clustering, at least one of the large clusters also exhibits temporal clustering that may be associated with magmatic or geothermal activity in the upper crust. The average Vp and Vs are estimated to be 5.85 and 3.35 km/sec for the upper crust and 7.0 and 3.9 km/sec for the lower crust, respectively. The 50 focal mechanisms computed from this set of events are consistent with north–south shortening and east–west extension; there are no clear indications of significant local perturbations in the regional stress field induced by the collision between India and Eurasia. The majority of the focal mechanisms indicate normal and strike-slip faulting. At least six of the newly computed focal mechanisms, however, indicate thrust faulting, which is a phenomenon not well documented previously. Ninety-nine percent of the local earthquakes have focal depths shallower than 25 km, and the locations of the few deeper events are poorly constrained. The shallow earthquake focal depths are consistent with high temperatures and proposed ductile or aseismic behavior in the middle to lower crust of central Tibet.

Beschreibung: We have constructed a composite image of the fault systems of the M 6.7 San Fernando (1971) and Northridge (1994), California, earthquakes, using industry reflection and oil test well data in the upper few kilometers of the crust, relocated aftershocks in the seismogenic crust, and LARSE II (Los Angeles Region Seismic Experiment, Phase II) reflection data in the middle and lower crust. In this image, the San Fernando fault system appears to consist of a decollement that extends 50 km northward at a dip of 25� from near the surface at the Northridge Hills fault, in the northern San Fernando Valley, to the San Andreas fault in the middle to lower crust. It follows a prominent aseismic reflective zone below and northward of the main-shock hypocenter. Interpreted upward splays off this decollement include the Mission Hills and San Gabriel faults and the two main rupture planes of the San Fernando earthquake, which appear to divide the hanging wall into shingle- or wedge-like blocks. In contrast, the fault system for the Northridge earthquake appears simple, at least east of the LARSE II transect, consisting of a fault that extends 20 km southward at a dip of 33� from 7 km depth beneath the Santa Susana Mountains, where it abuts the interpreted San Fernando decollement, to 20 km depth beneath the Santa Monica Mountains. It follows a weak aseismic reflective zone below and southward of the main-shock hypocenter. The middle crustal reflective zone along the interpreted San Fernando decollement appears similar to a reflective zone imaged beneath the San Gabriel Mountains along the LARSE I transect, to the east, in that it appears to connect major reverse or thrust faults in the Los Angeles region to the San Andreas fault. However, it differs in having a moderate versus a gentle dip and in containing no mid-crustal bright reflections.

Beschreibung: The San Andreas Fault System (SAFS) in central and northern California is a family of sub-parallel strike-slip faults that accommodate the transform motion between the Pacific and the North America plates. We present here a fully coupled numerical thermo-mechanical model of the SAFS as it evolved over the past 20 Myr following the northward migration of the Mendocino triple junction. The strike-slip displacements and fault perpendicular compression (over the last 5 Myr) are implemented as side boundary conditions in the upper 20 km of the 100 km deep model box, other parts of the side boundaries and bottom boundary being free for material flow. The model employs realistic visco-elasto-plastic rheology and allows for spontaneous generation of faults. The modeling replicates the evolution of multiple strike-slip faults, the overall eastern migration of the San Andreas fault itself from the western edge of the Coast Ranges near Cape Mendocino to the eastern edge in central California, the variation of heat flow in the Coast ranges along and across the SAFS as well as high angle (60-80$deg$) between the direction of the maximum horizontal stress and the local fault orientation in the upper crust. The modeling shows that the key conditions for such behavior are the cooling (and strengthening) of the lithosphere following passage of the triple junction which causes ductile deformation at depth to migrate eastward with time and the drop of frictional strength at high strain in the upper crust which results in a small number of major transform faults to develop, rather than a broad distribution of such faults. Our modeling suggests that the major faults in the SAFS in Central and Northern California must have been weak during most of the 20 Myr period over which it evolved.

Beschreibung: We use data from recently installed broad-band seismographs on the islands of Crete, Gavdos, Santorini, Naxos and Samos in the Hellenic subduction zone to construct receiver function images of the crust and upper mantle from south of Crete into the Aegean Sea. The stations are equipped with STS-2 seismometers and they are operated by GFZ Potsdam, University of Chania and ETH Zürich. Teleseismic earthquakes recorded by these stations at epicentral distances between 35° and 95° have been used to calculate receiver functions. The receiver function method is a routinely used tool to detect crustal and upper-mantle discontinuities beneath a seismic station by isolating the P-S converted waves from the coda of the P wave. Converted P-S energy from the oceanic Moho of the subducted African Plate is clearly observed beneath Gavdos and Crete at a depth ranging from 44 to 69 km. This boundary continues to the north to nearly 100 km depth beneath Santorini island. Because of a lack of data the correlation of this phase is uncertain north of Santorini beneath the Aegean Sea. Moho depths were calculated from primary converted waves and multiply reflected waves between the Moho and the Earth’s surface. Beneath southern and eastern Crete the Moho lies between 31 and 34 km depth. Beneath western and northern Crete the Moho is located at 32 and 39 km depth, respectively, and behaves as a reversed crust-mantle velocity contrast, possibly caused by hydration and serpentinization of the forearc mantle peridotite. The Moho beneath Gavdos island located south of Crete in the Lybyan Sea is at 26 km depth, indicating that the crust south of the Crete microcontinent is also thinning towards the Mediterranean ridge. This makes it unlikely that part of the crust in Crete consists of accreted sediments transported there during the present-day subduction process which began approximately 15 Ma because the backstop, i.e. the boundary between the current accretionary wedge of the Mediterranean ridge and the Crete microcontinent, is located approximately 100 km south of Gavdos. A seismic boundary at 32 km depth beneath Santorini island probably marks the crustal base of the Crete microcontinent. A shallower seismic interface beneath Santorini at 20-25 km depth may mark the depth of the detachment between the Crete microcontinent and the overlying Aegean subplate. The Moho in the central and northern Aegean, at Naxos and Samos, is observed at 25 and 28 km depth, respectively. Assuming a stretching factor of 1.2-1.3, crustal thickness in the Aegean was 30-35 km at the inception of the extensional regime in the Middle Miocene.

Beschreibung: From poroelasticity theory we know that fluid diffusion will induce matrix deformation and vice versa. In practice, well known phenomena for such coupled processes are, for example, occurrence of seismo-tectonically induced groundwater fluctuations, land subsidence as a result of fluid extraction from subsurface reservoirs, production-induced surface strain near the vicinity of wells, reservoir- or injection-induced seismicity. Modeling of deformation and pore-pressure data that have been observed near the surface can help to image the dynamics and to assess the hydraulic properties of subsurface aquifers. We here present a semi-analytical Haskell propagator method to fully handle linear poroelastic problems in a multilayered half-space. Our method is a powerful tool for various reasons: (1) It is faster than traditional numerical schemes when respective discretization of the object region is chosen and solutions are sought for single locations only; (2) a problem is easily formulated, as only a set of five poroelastic parameters per layer plus the layers' thicknesses need to be specified; (3) the method is highly flexible, as forcing functions of point injection, single force (e.g., surface loading), double couple dislocation (earthquakes), etc. may be readily incorporated; (4) the so-called loss-of-precision problem of the original propagator algorithm has been fully overcome using the orthonormalization technique. The effectiveness of the new tool has been demonstrated by modeling pump-induced near-surface tilt data obtained at a test site near Sopron in western Hungary. The results show that the hydraulic diffusivity of the shallow subsurface aquifer can be assessed with an accuracy better than half an order of magnitude, if other elastic parameters and the geometry (depth and thickness) of the water-bearing formations are sufficiently known from, for example, bore-log records. Moreover, the present method can be applied to model induced seismicity based on the Coulomb failure criterion.

Beschreibung: Obliquely convergent subduction margins commonly exhibit strain partitioning leading to trench-parallel material transport in the forearc. If deformation is localised, large margin-parallel strike-slip faults may form (e.g. in Peru, North-Chile or Sumatra). These faults are typically located far arcward from the deformation front (100-300km) in the internal part of the forearc. To date sandbox experiments, however, show strain partitioning in the deforming accretionary prism only. We therefore designed a series of scaled 3D sandbox experiments to monitor the deformation of the rearward region of the forming wedge. We further test how deformation is dependent on modes of shear force transmission at the base of the wedge. The latter is modelled by shifting the S-line relative to the deformation front. The S-line is the location on the plate interface landward of which basal shear forces are not further transmitted into the overlying wedge. Such a singularity is inescapable in analogue or numerical modelling and may have its natural analogy in the line of intersection (transition zone) of the subducting slab with some brittle-ductile boundary in the continental wedge. Analysis of the 3D surface displacement using an image correlation technique (PIV: particle imaging velocimetry) is used to evaluate material transport and to differentiate kinematic domains on the forming sand wedge. First results for a setting with shear force being transmitted along the entire backstop base show that most of the oblique convergence is accommodated in the fore- and back-thrust system. With proceeding convergence, a fault system evolves on the arcward side of the accretionary wedge, formed by the surface traces of back-thrusts. The main back-thrust is reactivated with a noticeable strike-slip component at the end of each fore-thrust forming cycle. Finally, the back-thrust becomes inactive and a new, more favorably oriented fault is formed in the backstop. This shows that the seemingly decoupled rearward region of the forearc system accommodates a trench parallel displacement component.------------------------

Beschreibung: GPS (Global Positioning System) radio occultation aboard the German CHAMP satellite (CHAllenging Minisatellite Payload) was activated in February 2001. It brought significant progress for establishing this innovative remote sensing technique as a standard method for precise monitoring of the Earth’s atmosphere on a global scale. We characterize the operational occultation data analysis at GFZ, review the results obtained so far and focus to future prospects. CHAMP provides continuously up to 200 vertical profiles of refractivity, temperature and water vapour daily. Since the experiment is currently expected to last until 2007, the first long-term set of GPS occultation data is expected. The measurements are analysed by an automatically working processing system, results are provided to the scientific community via the CHAMP data center at GFZ (ISDC, Information System and Data Center). Beginning in January 2003 a continuous transmission of analysis results for the purpose of data assimilation for numerical weather prediction is tested. An average delay of ~5 hours between each measurement and arrival of the data at the assimilation center (MPI Hamburg) is reached continuously. Data from meteorological analyses (ECMWF) and radio sondes are used to permanently validate the vertical atmospheric profiles from CHAMP to ensure high data product quality. Between 10 and 35 km the derived refractivity is nearly bias-free in relation to ECMWF e.g. for a set of ~60,000 vertical profiles, measured between March 2002 and 2003. The standard deviation is less than ~1%. This is confirmed by comparisons with ~ 10,000 coincidencing radio sonde measurements. In the middle and lower troposphere a negative refractivity bias (in relation to the analyses and radio sonde data) is observed in the Geometrical Optics (GO) retrievals. A maximum value of 5% is reached in the Tropics in the vicinity of the Earth’s surface. This bias is observed also by other analysis centers and was found to be caused by multipath propagation, receiver tracking errors or signal loss due to critical refraction. It also results in dry-biased water vapour profiles. In order to reduce the bias improved techniques for the data analysis in the middle and lower troposphere are applied (e.g. Canonical Transform sliding spectral, CTss or Full Spectrum Inversion sliding window, FSIsw). These techniques reduce the refractivity bias by a factor of more than 2 and are foreseen for the implementation to the operational processing system. Recent results of the analyses of global tropopause parameters and stratospheric temperatures, derived from 2.5 years of CHAMP measurements, are presented and will be discussed. Special attention is given to the tropical tropopause.

Beschreibung: Cosmogenic nuclides, measured in quartz from recent river bedload, provide a novel tool to quantify catchment-wide erosion rates at geologically meaningful time scales. Here we present an analysis of the geomorphological evolution of the 350 km2 Wutach catchment in the uplands of the south-west German Black Forest. The robustness of the method is demonstrated by the fact that, although the area was affected by river capture at 18 kyr BP, the formed gorge is so narrow that spatially averaged erosion rates were not resolvably perturbed. However, because cosmogenic nuclides preserve an erosion memory of several thousand years, the only perturbation introduced was detected in the minor areas that have been subject to the last maximum glaciation. In unglaciated areas, an important relationship between lithology and erosion can by quantified: sandstone lithologies erode at 12–18 mm kyr-1, granite lithologies at 35–47 mm kyr-1 and limestone lithologies (as deduced from river load gauging) at 70–90 mm kyr-1.