T23C-01 13:45h
Preservation/exhumation of ultrahigh-pressure (UHP) complexes
Recognized UHP terranes around the World share several unique characteristics: (1) All reflect descent of superjacent segments of continental crust to depths of ~90 km or more in compressional orogens. (2) All are thoroughly overprinted by lower pressure mineral assemblages, and rare relict UHP phases are preserved only in kinetically inhibiting circumstances such as inclusions in refractory host minerals. (3) Preserved UHP complexes consist exclusively of thin, allochthonous sheets. (4) Dense mafic + peridotitic rock types make up less than ~10 % of each UHP terrane, the major lithologies being lower density quartzofeldspathic units serpentinites. (5) Prograde + retrograde P-T paths are completed in ~5-10Myr, and rates of ascent approximate descent velocities. In addition, domical uplifts from mid-crustal levels characterize many UHP complexes. Continental crust can be carried down to great depths only in subduction zones, reflecting underflow of a chiefly oceanic plate prior to continental collision/suturing. Because exhumation involves decompression through the P-T stability fields of much lower pressure metamorphic facies, UHP relics are partially preserved mainly in tough, watertight minerals (e. g., zircon and garnet) in which their isolation prevents back-reaction. Thin-aspect-ratio, ductile-deformed nappes, generated by subduction-zone shear forces, conduct heat away from UHP complexes as they rise along the subduction channel, effectively cooling the sheets. Ascent is driven by buoyancy propulsion because the aggregate density of a dominantly quartzofeldspathic slice is considerably less dense than the mantle it had displaced. Relatively rapid ascent is required to defeat the establishment of a more normal geothermal regime in the subduction-zone environment. Tectonic aneurysms locally carry mid-crustal UHP decollements surfaceward. Where any one of these situations is not satisfied, UHP complexes will be totally transformed to lower pressure mineral assemblages, obliterating all evidence of past burial.
T23C-02 INVITED 14:00h
Intermediate Depth Earthquake Faulting in the Deep Continental Crust: Insights From Exhumed High-Pressure Rocks
One of the most intriguing and surprising seismological observations in recent years is the detection of earthquakes in the deep continental crust (e.g. Jackson, 2002) suggesting a strong lower crust to depths to 70 km (or more) in some continental collision zones. How the mechanical strength of the continental crust and mantle lithosphere varies as a function of depth has widespread implications for geodynamics. Information gleaned from exhumed gabbroic and anorthositic high-pressure complexes complexes in the Bergen area and the Lofoten Islands of Norway demonstrate that deeply subducted dry rocks in the lowermost continental crust may withstand metamorphic re-equilibration for geologically significant periods of time. The persistence of metastability had a significant effect on the mechanical properties of these rocks, as evidence by the abundant pseudotachylyte faults documenting seismic failure under eclogite facies conditions at depths exceeding 60 km in the thickened continental crust. Recent field research reveals that the localized earthquake fault systems are regionally extensive throughout the exhumed high-pressure complexes (100's of km$^{2}$). Outcrops of metastable mafic granulite $>$ 9 km$^{2}$ are transected by mm-cm thick pseudotachylyte veins and associated cataclasites and fractures; the pseudotachylyte accounts for $<$$<$ 1% of the total volume. For a fault area of ~ 4 km$^{2}$ (the observable extent of the largest pseudotachylyte sheets) and coseismic slip of 1 m (based conservatively on field observations), this leads to an estimate of a minimum earthquake magnitude estimate of $\sim$6.3. The fracturing associated with the faults could have provided access for the infiltration of fluids that instigated the localized conversion of the metastable granulites to eclogite. The observations indicate that very low volatile contents may control the ability of some lower crustal rock to remain metamorphically metastable and in a brittle seismogenic state at depths of $>$60 km in continental collision zones, despite prevailing conditions of T= 600-$700\deg$C and P= 1.5-2.0 GPa. This is consistent with recent experimental studies that suggest nominal amounts of in-situ water may be key in the initiation of seismic failure in essentially dry rocks at intermediate depths (Zhang et al., 2004). Together the data are compatible with present-day earthquake focal depth distributions in some continental collision zones, such as beneath the Himalayas, and may provide an explanation for the generation of deep-crustal earthquakes in these settings.
T23C-03 14:20h
High-pressure partial melting of eclogite and garnet amphibolite rocks during decompression and heating, the Tromso Nappe, Norway.
It has recently been argued that melting of high-P, high-T rocks in thickened arcs is an important process in magma genesis (e.g. Petford and Atherton, 1993; Rapp et al., 2003), producing rocks such as adakites. Such melting, however, has rarely been studied in-situ, and has instead relied on inferences from experimental and numerical studies, often based on quite different oceanic crustal lithologies. 452 Ma eclogites and garnet amphibolites in the Tromso Nappe, Norway provide one of the first opportunities to examine such melting in-situ. Evidence for a variety of melt forming reactions is preserved, involving melting of some or all of the eclogitic or retrograded eclogite components to form melt and peritectic garnet or amphibole. Thermobarometry shows that melting of eclogite involving peritectic garnet occurred at the highest pressures (P: 1.8 - 2.3 GPa), but lower than peak eclogite conditions of 3.3 GPa (Krogh-Ravna, pers.comm. 2004). Partial melting involving peritectic amphibole and melting of garnet amphibolite involving peritectic garnet both yield lower pressures (P: 1.1 GPa). These results suggest that melting of eclogite rock in continental arcs may be intimately linked to exhumation of those arcs. Temperatures remained high (T: 800 C), most likely in response to the emplacement of the neighbouring Skattora Migmatite Complex (Selbekk and Skjerlie, 2002). This supports the modelling of Petford & Gallagher (2001) that suggests that melting in lower continental arcs may be in response to intrusion of magmatic bodies nearby, and suggests that melting of the eclogites and garnet amphibolites in the Tromso Nappe was due to both decompression and heating. Four reactions are inferred qualitatively: (1) Omphacite +/- Zoisite +/- Kyanite +/- Phengite +/- Quartz goes to Garnet + Melt; (2) Omphacite + Garnet +/- Zoisite +/- Kyanite +/- Phengite +/- Quartz goes to Amphibole + Melt; (3) Omphacite +/- Zoisite +/- Kyanite +/- Phengite +/- Quartz goes to Amphibole + Melt; (4) Amphibole +/- Omphacite +/- Quartz +/- Biotite goes to Garnet + Melt. Current work is focusing on fully quantifying these reactions and the composition of melts produced. References: Petford, N. and Atherton, M. 1996. Na-rich partial melts from newly underplated basaltic crust: the Cordillera Blanca Batholith, Peru. J. Pet. 37, 1491-1521. Petford, N. and Gallagher, K. 2001. Partial melting of mafic (amphibolitic) lower crust by periodic influx of basaltic magma. EPSL, 193, 483-499. Rapp, R. P., Shimizu, N. and Norman, M. D. 2003. Growth of early continental crust by partial melting of eclogite. Nature, 425, 605-609. Selbekk, R. and Skjerlie, K-P. 2002. Petrogenesis of the Anorthosiste Dyke Swarm of Tromso, North Norway: Experimental evidence for Hydrous Anatexis of an Alkaline Mafic Complex. J. Pet. 43, 943-962.
T23C-04 14:35h
Exhuming Large UHP Terranes: Insights From an Amphibolite to Ultrahigh-Pressure Transition in Western Norway.
Various mechanisms have been proposed to account for the exhumation of continental ultrahigh-pressure (UHP) terranes; these include (i) lithosphere-scale normal faulting; (ii) extrusion of a `pip' or `sliver' along a basal thrust; and (iii) buoyant return of continental lithosphere after slab detachment. However, critical aspects of these models, most notably the existence of a sole thrust beneath the UHP crust, are yet to find compelling support in studies of large UHP terranes ($>$5000 km$^{2}$). We present the results of detailed structural analysis, eclogite thermobarometry and single grain U/Pb zircon geochronology across the southern boundary of the Nordfjord UHP province, Western Gneiss Region, Norway. Specifically, we find that peak metamorphic grade increases steadily over a 40 km distance from high-P amphibolite facies to the coesite-eclogite facies. This is accompanied by a continuous evolution from hornblende and epidote-bearing, prograde-zoned, incompletely recrystallized eclogites to blastic, high variance, unzoned UHP assemblages. Most importantly, this transition is not disrupted by any significant structural or metamorphic breaks. We conclude from these observations that the UHP transition results from in-situ prograde metamorphism, to increasingly greater depths northwestward. These data do not support thrust emplacement of the UHP rocks over lower pressure crust in the Nordfjord area. Furthermore, late-orogenic extrusion of the HP/UHP terrane is not proven by seismic evidence of a buried decollement beneath the WGR, or by simultaneous thrusting in foreland regions. Therefore, we argue that the HP/UHP rocks in the Nordfjord area remained essentially attached to the Baltica craton, and were exhumed by a combination of extensional unroofing and erosion, possibly coupled with flexural rebound of the continental lithosphere after slab failure at greater depth.
T23C-05 14:50h
Exhumation of the Western Gneiss Region Ultrahigh-Pressure Terrane
Coordinated structural, petrological, and geochronological analysis of the Norwegian HP-UHP terrane, the world's largest, lends new insight into these remarkable rocks. New UHP eclogites, new HP eclogites, and new 40Ar/39Ar ages define three discrete ultrahigh-pressure (UHP) domains that are separated by distinctly lower pressure, but still eclogite-facies, rocks. 40Ar/39Ar mica and K-feldspar ages show that this outcrop pattern is the result of gentle regional-scale folding younger than 380 Ma, and that the UHP domains comprise antiformal culminations beneath a HP veneer. The shapes of the UHP antiforms overturn the long-held idea that P and T increase monotonically toward the coast. The sizes of the UHP antiforms range from 2500 km2 to 100 km2; if the UHP culminations are part of a continuous sheet at depth, the UHP terrane underlies a minimum of 11,000 km2 and is overlain by a HP veneer of 60,000 km2. No recognized structure separates the UHP and HP domains, and the transition in the Nordfjord area is defined by a gradual increase in P and T from the amphibolite to the coesite-eclogite facies. Metapelites in the UHP core of the orogen indicate that the (U)HP metamorphism was followed by near-isothermal decompression to 1.2-1.7 GPa/750 C and then to 0.5 GPa/750 C; metapelites in the HP area show a supra-Barrovian overprint at 1.0-1.2 GPa/750 C and then 0.5 GPa/750 C. New U/Pb zircon, U-Th/Pb monazite, and 40Ar/39Ar ages, combined with extant geochronology, require decompression from eclogite-facies conditions at 410-405 Ma to mid-crustal depths in a few million years. The short timescale and consistently high temperatures imply adiabatic exhumation of a UHP body with a minimum dimension of 30-40 km. 40Ar/39Ar muscovite ages of 397-380 Ma show that this extreme heat advection was followed by rapid cooling (30 C/m.y.) at mid-crustal levels caused by large-scale extension.
T23C-06 15:10h
A Kinetic Model for Scattered Radiometric Ages in HP/UHP Terrains Using the Western Gneiss Region of the Scandinavian Caledonides as an Example
Attempts to date the "peak" of metamorphism in HP/UHP terrains are complicated by radiometric dates that scatter over significant time periods: on the order of tens of millions of years in some terrains. For example, eclogites dated by Sm-Nd mineral isochron and U-Pb zircon techniques in the Western Gneiss Region (WGR) of the Norwegian Caledonides give ages that range from 402 to 422 Ma (the oldest age is a re-determination of an eclogite previously dated at 447 Ma). Some recent studies have proposed that the older ages should be discarded and the youngest ages should taken to date the HP/UHP metamorphism that accompanied the collision of Baltica and Laurentia during the Scandian Orogeny. This interpretation implies that HP/UHP metamorphism was a discrete, short-lived event that occurred during or shortly after the achievement of peak temperatures. However, if HP/UHP conditions are achieved through the subduction of continental crust into the mantle, eclogite-facies assemblages have the potential to form throughout the interval that the crust is within the eclogite stability field, during both subduction and eduction. Recent evidence suggests that eclogitization requires the introduction of fluids or the application of strain or the action of some other catalytic process and that where these processes do not occur the rocks can persist metastably as non-eclogite facies assemblages. This means that eclogitization can occur locally, wherever and whenever fluids are introduced or strain is localized, rather than occurring coherently within a short interval throughout the entire terrene. If so, all determined ages from the WGR could be correct with each one dating the formation of eclogite-facies assemblages in a particular place at a particular time. If so, the WGR resided at depths within the eclogite facies stability field for ca. 20 m.y. A testable consequence of this model is that eclogites from the part of the slab that was subducted to the deepest levels should have both the oldest and youngest eclogites with the ages dating the entry into, and the departure from, respectively, the eclogite stability field.
T23C-07 15:25h
Apatite geochronology: Applications to exhumation histories in orogenic settings
The Lu-Hf system in apatite represents a robust dating tool for assessing the extent, timing and duration of high-grade metamorphic processes. Until recently, garnet has been the only common phase in high-grade terranes suitable for Lu-Hf geochronology. However, apatite has elevated $^{176}$Lu/$^{177}$Hf ratios compared to garnet (up to 90 versus less than 30 [e.g., 1]). The $^{176}$Lu/$^{177}$Hf ratios are heterogeneous on a mm-scale allowing, in some cases, for dating of single apatite crystals [2]. Moreover, mineral inclusions in apatite can be largely eliminated by dissolution of the host apatite in highly dilute acid [2]; whereas one of the major challenges in garnet geochronology is assessing the influence from possible inherited inclusions [e.g., 3]. Apatite from three Proterozoic terranes yields Lu-Hf ages that are consistently older than their respective Pb step leaching (PbSL) ages. Isotopic closure for the Lu-Hf system, therefore, occurs prior to U-Pb closure in this mineral. In the Adirondack Lowlands, where H$_{2}$O activity was low, Lu-Hf ages of 1270 Ma and 1230 Ma are obtained for 3 cm and 0.3 cm diameter apatite crystals, respectively. These ages significantly predate the peak metamorphism conditions during the Elzevirian orogeny in the Adirondack Lowlands that reached upper amphibolite facies at 1170-1130 Ma. In contrast, apatite from the fluid-rich Otter Lake area and Black Hills record surprisingly young Lu-Hf and PbSL ages, presumably reflecting late exchange facilitated by fluids. The Lu-Hf ages for metamorphic apatites are shown to be controlled by temperature and grain size, and, in areas where fluid activity was high during peak metamorphism and exhumation, by fluid composition. [1] Anczkiewicz et al., 2004, EPSL, 255, 147-161; [2] Barfod et al., in press, GCA; [3] Scherer et al., 2000, GCA, 64, 3413-3432.