Mantle source compositions of the Paraná Magmatic Province (southern Brazil): evidence from trace element and Sr–Nd–Pb isotope geochemistry

An integrated study based on incompatible trace elements and Sr–Nd–Pb isotopes is presented in order to assess the mantle sources involved in the genesis of the Paraná Magmatic Province (PMP) tholeiites. Particular emphasis is given to 33 new Pb isotope and concentration data obtained in representative samples of low-TiO₂ (LTiB) and high-TiO₂ (HTiB) flood basalts that occur in the province. Results show important differences with respect to type and location of these rocks. The LTiB and HTiB from northern PMP exhibit very similar initial Pb isotope ratios (average LTiB vs HTiB: ²⁰⁶Pb/²⁰⁴Pb_i=17.78±0.03 vs 17.65±0.02; ²⁰⁷Pb/²⁰⁴Pb_i=15.53±0.01 vs 15.52±0.01; ²⁰⁸Pb/²⁰⁴Pb_i=38.12±0.03 vs 38.05±0.04). The LTiB from southern PMP, with initial ⁸⁷Sr/⁸⁶Sr_i≤0.7060, show small variation in initial Pb isotope compositions (average ²⁰⁶Pb/²⁰⁴Pb_i=18.20±0.07; ²⁰⁷Pb/²⁰⁴Pb_i=15.61±0.01; ²⁰⁸Pb/²⁰⁴Pb_i=38.32±0.10), which are highly enriched in radiogenic Pb in comparison to the northern PMP analogues. The HTiB from southern PMP have initial Pb isotope ratios (average ²⁰⁶Pb/²⁰⁴Pb_i=17.45±0.09; ²⁰⁷Pb/²⁰⁴Pb_i=15.50±0.01; ²⁰⁸Pb/²⁰⁴Pb_i=37.89±0.03) slightly less radiogenic compared with the HTiB from northern PMP. The data cover a large range of isotope compositions, which are accompanied by systematic changes in incompatible trace element ratios and Sr–Nd isotopes, indicating contributions from different mantle sources. The remarkable chemical and isotope differences between PMP basalts, N-MORB and Tristan da Cunha least evolved volcanics indicate that these asthenospheric sources did not play a significant role in the basalt genesis, suggesting generation from the melting of heterogeneous lithospheric mantle sources. The close similarity between the radiogenic isotopes of the Cretaceous carbonatites that surround the PMP and those of the HTiB rock-types and the LTiB from the northern Paraná suggests the involvement of the same mantle components in their genesis: a dominant EMI end member and a radiogenic isotope enriched component of EMII-type, as some phlogopite-peridotite mantle xenoliths (Japecanga) from the Alto Paranaíba Igneous Province. The latter component seems also to have an important role in the origin of the LTiB from the southern Paraná, where the other end member is highly depleted in radiogenic lead similar to DMM.     

The geoid in southeastern Brazil and adjacent regions: new constraints on density distribution and thermal state of the lithosphere

Least-squares collocation technique was used to process regional gravity data of the SE South American lithospheric plate in order to map intermediate (10–2000 km) wavelength geoid anomalies. The area between 35–10° S and 60–25° W includes the Paraná CFB Province, the Southern São Francisco Craton and its marginal fold/thrust belts, the Brazilian continental margin and oceanic basins. The main features in the geoid anomaly map are: (a) Paraná CFB Province is characterized by a 1000 km long and 500 km wide, NE-trending, 9 m-amplitude negative anomaly which correlates with the distribution of sediments and basalts within the Paraná basin. (b) A circular (600–800 km in diameter) positive, 8 m-amplitude geoid anomaly is located in the southern S. Francisco craton and extends into the northeastern border of the Paraná CFB Province. This anomaly partially correlates with Alto Paranaíba Igneous Province (APIP), where alkalic volcanism and tholeiitic dikes of ages younger than 80 Ma are found and where a low-velocity zone in the mantle has been mapped using seismic tomography. This positive geoid anomaly extends towards the continental margin at latitude 21° S and joins a linear sequence of short wavelength positive geoid anomalies associated with Vitoria–Trindade seamounts. (c) A NE-trending, 1000 km long and 800 km wide, 4 m-amplitude, positive geoid anomaly, which is located along the southeastern coast of Brazil, from latitude 24 to 35° S. The northern part of this anomaly correlates with the Ponta Grossa Arch and Florianopolis dyke swarm provinces. The age of this intrusive volcanism is 130–120 Ma. (d) A circular positive anomaly with 9 m of amplitude, located over the Rio Grande and Uruguay shields and offshore Pelotas basin. Few alkaline intrusives with ages between 65 and 80 Ma are found in the region and apatite fission track ages in basement rocks indicates cooling at around 30 Ma. A semi-quantitative analysis of the observed geoid anomalies using isostatic considerations suggests that the mechanism which generated Paraná CFB Province did not change, in a significant manner, the lithospheric thermal structure, since the same geoid pattern observed within this province continues northward over the Neoproterozoic fold/thrust belts systems separating the São Francisco and Amazon cratons. Therefore, this observation favours Anderson’s idea of rapid basaltic outpouring through a pull-apart mechanism along a major suture zone. A thermal component may still be present in the Southern São Francisco Craton and in the Rio Grande Shield and contiguous continental margins, sites of Tertiary thermal and magmatic reactivations.     

Paleomagnetism of the trap intrusive bodies in arctic Siberia: Geological and methodical implications

New paleomagnetic data are reported for the dikes, sills, and lava flows in the Arctic part of the Siberian Platform, which has not been covered by previous systematical paleomagnetic investigations. The analysis of the newly obtained and previously published data provides important time constraints for the duration and character of evolution of the Permian-Triassic magmatic events in the studied regions. Our results once again illustrate the conclusion that, in order to obtain an exact estimate for the location of the paleomagnetic pole in the northern paleolatitudes, at least 30?C40 rapidly cooled magmatic bodies (dikes, flows, or minor sills) should be sampled if secular variation is commensurate with the intensity of the presentday variations.     

Paleomagnetism, geochronology, and magnetic mineralogy of Devonian dikes from the Kola alkaline province (NE Fennoscandian Shield)

The new paleomagnetic data on forty dikes and two intrusive plutons of Devonian age located in different parts of the Kola Peninsula, which have not been previously covered by systematic paleomagnetic studies, are reported. We describe the results of the rock magnetic, petrographic, and microprobe investigations of the Devonian dikes and present their isotopic ages (⁴⁰Ar/³⁹Ar, stepwise heating). Within the studied area, almost all the Devonian dikes, metamorphic Archaean-Proterozoic complexes of the Fennoscandian Shield, and Proterozoic dikes have undergone low-temperature hydrothermal-metasomatic alteration, which resulted in the formation of new magnetic minerals with a secondary (chemical) component of magnetization. The comparison of the paleomagnetic poles indicates the Early Jurassic age of the secondary component. We suggest that regional remagnetization event was caused by endogenic activity genetically related to the formation of the Barents Sea trap province 200–170 Ma ago. On the basis of the obtained data, the preliminary Devonian paleomagnetic pole of the East European Platform is determined.     

Paleomagnetism of the western Cape Fold belt, South Africa, and its bearing on the Paleozoic apparent polar wander path for Gondwana

In order to test two different proposals for the poorly defined African Paleozoic apparent polar wander path (APWP), a paleomagnetic study was carried out on Ordovician through Carboniferous clastic sediments from the Cape Fold belt, west of the 22nd meridian. One proposal involves a relatively simple APWP connecting the Ordovician Gondwana poles in North Africa with the Late Paleozoic poles to the east of South Africa in a more or less straight line crossing the present equator in the Devonian. The other proposal adds a loop to this path, connecting Ordovician poles in North Africa with poles to the southwest of South Africa and then returning to central Africa. This loop would occur mainly in Silurian time. New results reported herein yield paleopoles in northern and central Africa for Ordovician to lowermost Silurian and Lower to Middle Devonian formations. The best determined paleopole of our study is for the Early Ordovician Graafwater Formation and falls at 28°N, 14°E (k = 25, α₉₅ = 8.8°, N = 28 samples). The other paleopoles are not based on sufficient numbers of samples, but can help to constrain the apparent polar wander path for Gondwana. Our results give only paleopoles well to the north of South Africa and we observe no directions within the proposed loop. Hence, if the loop is real, it must have been of relatively short duration (60–70 Ma) and be essentially of Silurian/Early Devonian age, implying very high drift velocities for Gondwana (with respect to the pole) during that interval.     

Ar–Ar geochronology of Santo Antão, Cape Verde Islands

Santo Antão, the northernmost island of the Cape Verde Archipelago, consists entirely of silica-undersaturated volcanic products and minor intrusions. ⁴⁰Ar–³⁹Ar incremental heating experiments have been carried out on 24 samples that cover the entire exposed chronological sequence. The oldest lavas (7.57±0.56 Ma), representing an older volcanic basement, are exposed about 620 m above mean sea level. After an interval of quiescence of up to 4.3 Ma the volcanic activity resumed and continued at low eruption rates. The older basement is unconformably overlain by a ca. 810-m-thick lava sequence that spans an age range from 2.93±0.03 to 1.18±0.01 Ma. This sequence is cut by many dykes and sills. Simultaneous volcanic activity occurred in the northeastern, central and eastern part of the island. A phonolitic pumice deposit that forms a noteworthy feature over most of the island has an estimated age of 0.20 Ma. This predates volcanic activity that formed the highest point of the island (Tope de Coroa) which has an age of 0.17±0.02 Ma. The most recent eruption on the island formed nephelinitic lavas in the Porto Novo region at 0.09±0.03 Ma. The oldest volcanism exposed on Santo Antão, which took place about 7.6 Ma ago, was simultaneous with waning activity on Maio at the eastern end of the Cape Verde Archipelago.     

Reassembling the Paleogene–Eocene North Atlantic igneous province: New paleomagnetic constraints from the Isle of Mull, Scotland

The paleomagnetic data sets from the British Tertiary Igneous Province (BTIP) have recently been criticized as being unreliable and discordant with data from elsewhere in the North Atlantic Igneous Province (NAIP) [Riisager et al. Earth Planet. Sci. Lett. 201 (2002) 261–276; Riisager et al. Earth Planet. Sci. Lett. 214 (2003) 409–425]. We offer new paleomagnetic data for the extensive lava flow sequence on the Isle of Mull, Scotland, and can confirm the paleomagnetic pole positions emanating from important earlier studies. Our new north paleomagnetic pole position for Eurasia at 59 ± 0.2 Ma has latitude 73.3°N, longitude 166.2°E (dp/dm = 5.2/7.0).A re-evaluation and an inter-comparison of the paleomagnetic database emanating from the NAIP were carried out to test for sub-province consistency. We find a general agreement between the Eurasian part of NAIP (BTIP and Faeroes) and East Greenland data. However a compilation of West Greenland data displays a large and unexplained dispersion. We speculate on if this is related to different sense of block rotation of the Tertiary West Greenland constituents. Combining all data from the NAIP constituents, give a pole position at 75.0°N, 169.9°E (N = 25, K = 84.3, A₉₅ = 3.2) in Eurasian reference frame.     

Paleomagnetic evidence for the relative positions of Timor and Australia in the Permian

An Upper Permian paleomagnetic pole has been determined for the Cribas Formation in eastern Timor. The co-ordinates for the mean pole are 159.8°E and 56.6°S,α₉₅ = 9.0. The reliability of the pole is ascertained through thermal demagnetization, a fold test, comparison between red beds and a lava flow, and the presence of normal and reversed polarities. The Timor pole is in excellent agreement with the Australian Upper Permian and Triassic poles. From this it is inferred that autochthonous Timor formed part of the Australian continental margin at least since the Upper Permian.     

The Medina Wrench: a key to the kinematics of the central and eastern Mediterranean over the past 5 Ma

The Medina Wrenth in the central Mediterranean is a transform fault connecting the plate collision in northwest Africa and northern Sicily with that occurring at the Aegean plate boundary, south of Greece. The more than 800 km long crescent-shaped wrench zone is currently seismically quiet but exhibits major deformation since 5 Ma within a belt 30–100 km wide. It forms the southern boundary of two microplates moving eastward with respect to Africa and Europe. A simple plate rotation model constrained by recent paleomagnetic data indicates that a continental Iblean microplate and a hybrid continental/oceanic Ionian microplate, separated along the Malta Escarpment, have rotated anticlockwise by 11° and 12°, respectively, around poles in southern Italy. These rotations involved some 100 km of dextral eastward movement relative to Africa of the Ionian Basin north of the Medina Wrench since 5 Ma. Combining the published 26° clockwise rotation of the Peloponnesus and northwest half of the Aegean with the 12° anticlockwise rotation of the Ionian microplate results in (a) a 99% agreement between the length of the seismic Benioff Zone beneath Greece and the total convergence of the microplates, and (b) an average rate of convergence across the Aegean plate boundary southwest of the Peloponnesus of 6.6 ± 1cm a⁻¹ since the Miocene. Relative motion between microplates in a collision zone thus may be as much as 6 times faster than convergence between the major plates which spawned them, and they can be considered rigid to the first order over the time span involved.     

Trans-Atlantic correlation of eruptive sequences and individual silicic volcanic units within the Paraná-Etendeka igneous province

The Mesozoic volcanic rocks of the Serra Geral Formation in the Paraná Basin, South America, and of the Etendeka Group in northwestern Namibia were erupted shortly before the opening of the South Atlantic. The major widespread silicic volcanic units in the Etendeka Group are interpreted as rheoignimbrites (Milner et al., 1992) and are interbedded with tholeiitic basalts and basaltic andesites.The southern portion of the Etendeka Group is subdivided into a basal Awahab Formation which is overlain disconformably by the Tafelberg Formation. Both formations contain silicic and mafic units. Bulk composition, initial ⁸⁷Sr/⁸⁶Sr ratios, phenocryst assemblages and mineral compositions are used to correlate silicic units of the Awahab Formation with the basal units of the Palmas silicic volcanic rocks in the southern Paraná Basin. Silicic units of the Tafelberg Formation can similarly be correlated with silicic units in the upper portion of the Palmas succession, which are also disconformable on the units below them. Not all silicic units in these successions are present in both the Etendeka and Paraná areas, but where correlation of individual units is possible, then this is found to be consistent with the overall stratigraphic sequence.Silicic units in the Awahab Formation were erupted from the Messum Igneous Complex in Namibia and their correlation into Brazil indicates that individual eruptive units must have travelled over 340 km from their source. Serial changes in the composition of silicic units in the Awahab Formation and their correlatives indicates that they were erupted from a single magma system from which a total of ˜ 8600 km³ of material was erupted.     

Paleomagnetism of early cretaceous arapey formation (Northern Uruguay)

We report detailed rock-magnetic and paleomagnetic directional data from 35 lava flows (302 standard paleomagnetic cores) sampled in the Central-Northern region of Uruguay in order to contribute to the study of the paleosecular variation of the Earth’s magnetic field during early Cretaceous and to obtain precise Cretaceous paleomagnetic pole positions for stable South America. The average unit direction is rather precisely determined from 29 out of 35 sites. All A₉₅ confidence angles are less than 8°, which points to small within-site dispersion and high directional stability. Normal polarity magnetizations are revealed for 19 sites and 10 are reversely magnetized. Two other sites yield well defined intermediate polarities. The mean direction, supported by a positive reversal test is in reasonably good agreement with the expected paleodirection for Early Cretaceous stable South America and in disagreement with a 10° clockwise rotation found in the previous studies. On the other hand, paleomagnetic poles are significantly different from the pole position suggested by hotspot reconstructions, which may be due to true polar wander or the hotspot motion. Our data suggest a different style of secular variation during (and just before) the Cretaceous Normal Superchron and the last 5 Ma, supporting a link between paleosecular variation and reversal frequency.     

Leaf breakdown in two tropical streams: Differences between single and mixed species packs

We assessed leaf breakdown of five native riparian species from Brazilian Cerrado (Myrcia guyanensis, Ocotea sp., Miconia chartacea, Protium brasiliense, and Protium heptaphyllum), incubated in single and mixed species packs in two headwater streams with different physico-chemical properties in the Espinhaço Mountain range (Southeastern Brazil). Leaves were placed in plastic litter bags (15 cm×20 cm, 10 mm mesh size) and the experiments were carried out during the dry seasons of 2003 and 2004. Leaf nitrogen and phosphorus contents were similar in all species, but polyphenolic contents were different (P<0.001). M. guyanensis showed higher polyphenolics content (8.48% g⁻¹ dry mass) and leaf toughness. Individually, higher breakdown rates were found in M. guyanensis at Indaiá stream (k=0.0063±0.0005 d⁻¹) and in Ocotea sp. at Garcia stream (k=0.0088±0.0006 d⁻¹). However, P. brasiliense and P. heptaphyllum showed lower breakdown rates at Indaiá and Garcia streams (Indaiá: k=0.0020±0.0002 and 0.0019±0.0001 d⁻¹; Garcia: k=0.0042±0.0001 and 0.0040±0.0002 d⁻¹). Single and mixed breakdown processes of each species were not statistically different on both streams. However, all species showed higher breakdown rates at Garcia stream (P<0.01). These results suggest that leaf breakdown is not altered when litter benthic patches are composed by a mixture of species in the same proportions that they occur on riparian leaf falls.     

Loading effect of a self-consistent equilibrium ocean pole tide on the gravimetric parameters of the gravity pole tides at superconducting gravimeter stations

The gravimetric parameters of the gravity pole tide are the amplitude factor δ, which is the ratio of gravity variations induced by polar motion for a real Earth to variations computed for a rigid one, and the phase difference κ between the observed and the rigid gravity pole tide. They can be estimated from the records of superconducting gravimeters (SGs). However, they are affected by the loading effect of the ocean pole tide. Recent results from TOPEX/Poseidon (TP) altimeter confirm that the ocean pole tide has a self-consistent equilibrium response. Accordingly, we calculate the gravity loading effects as well as their influence on the gravimetric parameters of gravity pole tide at all the 26 SG stations in the world on the assumption of a self-consistent equilibrium ocean pole tide model. The gravity loading effect is evaluated between 1 January 1997 and 31 December 2006. Numerical results show that the amplitude of the gravity loading effect reaches 10⁻⁹ m s⁻², which is larger than the accuracy (10⁻¹⁰ m s⁻²) of a SG. The gravimetric factor δ is 1% larger at all SG stations. Then, the contribution of a self-consistent ocean pole tide to the pole tide gravimetric parameters cannot be ignored as it exceeds the current accuracy of the estimation of the pole tide gravity factors. For the nine stations studied in Ducarme et al. [Ducarme, B., Venedikov, A.P., Arnoso, J., et al., 2006. Global analysis of the GGP superconducting gravimeters network for the estimation of the pole tide gravimetric amplitude factor. J. Geodyn. 41, 334–344.], the mean of the modeled tidal factors δ_m = 1.1813 agrees very well with the result of a global analysis δ_CH = 1.1816 ± 0.0047 in that paper. On the other hand, the modeled phase difference κ_m varies from −0.273° to 0.351°. Comparing to the two main periods of the gravity pole tide, annual period and Chandler period, κ_m is too small to be considered. Therefore, The computed time difference κ_L induced by a self-consistent ocean pole tide produces a negligible effect on κ_m. It confirms the results of Ducarme et al., 2006, where no convincing time difference was found in the SG records.     

Paleomagnetism of the triassic nicola volcanics and geotectonics of the quesnellia subterrane of terrane I, British Columbia

Paleomagnetic data from 46 sites (674 specimens) of the Westcoast Crystalline Gneiss Complex on the west coast of Vancouver Island using AF and thermal demagnetization methods yields a high blocking temperature WCB component (> 560°C) with a pole at 335°W, 66°N (δ_p = 4°, δ_m = 6°) and a lower coercivity WCA component ( 25 mT, < 500°C) with a pole at 52°W, 79°N (δ_p = 7°, δ_m = 8°). Further thermal demagnetization data from 24 sites in the Jurassic Island Intrusions also defines two high blocking temperature components. The IIA component pole is at 59°W, 79°N (δ_p = 7°, δ_m = 8°) and IIB pole at 130°W, 73°N (δ_p = 12°, δ_m = 13°). Combined with previous data from the Karmutsen Basalts and mid-Tertiary units on Vancouver Island and from the adjacent Coast Plutonic Complex, the geotectonic motions are examined for the Vancouver Island segment of the Wrangellian Subterrane of composite Terrane II of the Cordillera. The simplest hypothesis invokes relatively uniform translation for Terrane II from Upper Triassic to Eocene time producing 39° ± 6° of northward motion relative to the North American craton, combined with 40° of clockwise rotation during the Lower Tertiary.     

On farsidedness of palaeomagnetic poles: Magnetic refraction,sediment compaction and dipole off-set

Summary Three possible ways to explain the Caenozoically observed farsidedness of paleomagnetic poles (apart from lithospheric plate movements) are discussed: magnetic refraction, sediment compaction and dipole off-set. The dipole off-set, being a possible geomagnetic field property, will be of opposite sign on opposite hemispheres, and hence will not tend to smoothe out by sectorial averaging. Sediment compaction shallows the inclination on both hemispheres, and hence will tend to smoothe out by sectorial averaging, provided that sediment properties, site latitude coverage and number of investigations are equal (fairly unlikely).Magnetic refraction causes systematic directional distortions of the remanent magnetization in rocks of moderate to high magnetic intensity (or apparent susceptibility k_app=k(1+Q)) such as in many volcanics, some metamorphics, as well as in baked clays and slags, etc. A detailed discussion of this effect is given: If the k_app of the material is above 0.001 emu/cm ³ (×4 SI), this effect is likely to cause a significant palaeomagnetic refraction error of the NRM (typically a TRM or a CRM) of the rock. An apparent susceptibility of this order of magnitude is quite common in volcanic rocks; e.g. for oceanic floor basalts the average of k_app is about 0.02 emu/cm ³ corresponding to systematic errors (flattening) of some 3° to 6° in the inclination of a horizontal flow, depending on the latitude.To improve paleomagnetic results in general, a simple refraction correction is therefore suggested to be applied in the case of common two-dimensional (i.e. flat, elongated) geological bodies such as dykes, sills, lava flows and baked clays. Numerical solutions are given for the horizontal case, while a graphical solution is given for the general two-dimensional case.Being of systematic types, the refraction error together with the sediment compaction effect may be responsible for a major part of the observed farsidedness of the Caenozoic palaeomagnetic pole positions, the apparent farsidedness not yet beeing masked by the scatter of pole positions produced by older individual lithospheric plate movements.Presented at 2nd conference on New Trends in Geomagnetism, Castle of Bechyn, Czechoslovakia, September 24–29, 1990.     

The polar spirals of Mars may be due to glacier surges deflected by Coriolis forces

All previous accounts of the spiral patterns at the Martian poles emphasize that the north polar spiral is centered about the geographic pole, whereas that of the south polar region is off-set by about 4°. This paper demonstrates that the patterns near both poles are centered on topographic highs rather than the spin poles themselves. This is circumstantial evidence in favour of the relatively unexplored mechanism of radial outflow of viscous rock by gravity spreading.The hypothesis developed here is that the spiral patterns are essentially due to crevasse patterns formed perpendicular to flow lines which are perturbed by Coriolis forces. In order to account for a crevasse pattern that has a form concave to the east the angular deflection of an hypothetical ice flow emanating from the topographic high centered about the geographical north pole, must be about 40° or 0.7 radians in a westward direction at 85°N latitude.The polar cap rock has previously been assumed to consist mainly of either frozen carbondioxide or water ice. Corresponding viscosities (at 190 K) allow for the occurrence of radial outflow or gravity driven tectonics at a maximum rate of 1 cm a⁻¹, but the flow pattern remains unaffected by Coriolis forces.The spiral patterns of the Martian poles can be explained if the flowing mass has an occasional effective kinematic viscosity as low as about 7 × 10⁶ m² s⁻¹, because gravity tectonics will then be deflected by Coriolis forces resulting in appropriately curved flowlines. A tensile fracture pattern, resembling an anticlockwise spiral pattern perpendicular to the clockwise deflected flowlines may subsequently form by local brittle failure.The occasional kinematic viscosity 7 × 10⁶ m² s⁻¹ would cause flow rates of 0.2 m s⁻¹ along the slopes of the topographic highs. This velocity and the corresponding viscosity is tentatively thought to be possible when thermal and pressure runaway occurs in the polar layered deposits. This would mean glacier surges on the Martian poles are two orders of magnitude faster than those hitherto observed on Earth.     

New paleomagnetic pole and magnetostratigraphy of Faroe Islands flood volcanics, North Atlantic igneous province

A paleomagnetic sampling was carried out along four sections (altogether 86 lava flows, 548 samples) in the North Atlantic Igneous Province outcropping in Faroe Islands, Denmark. The four polarity zones in the 700-m-thick exposed part of the Faroes lower formation can be correlated with the geomagnetic polarity time scale as C26n-C25r-C25n-C24r. The seven lava flows erupted during C25n indicate a very low eruption rate in the upper part of the Faroes lower formation of ∼1/70 kyr. The Faroes middle and upper formations (composite thickness ∼2300 m) are all reversely magnetized corresponding to C24r. The eruption rate at the onset of middle formation volcanism was very high as evidenced by several thick lava sequences recording essentially spot readings of the paleomagnetic field. The shift in eruption rate between the Faroes lower and middle formations and evidence that onset of the Faroes middle formation volcanism took place in C24r are of particular importance, placing onset of middle formation volcanism in close temporal relation to North Atlantic continental break-up and the late Paleocene thermal maximum. After grouping flows recording the same field directions, we obtained 43 independent readings of the paleomagnetic field, yielding a paleomagnetic pole with coordinates 71.4°N, 154.7°E (A₉₅=6.0°, K=14, N=43); age 55-58 Ma. The pole is supported by a positive reversal test. Paleosecular variation, estimated as the angular standard deviation of the virtual geomagnetic pole distribution 21.7°+3.9°/−2.8°, is close to expected for the given age and paleolatitude. Our new Faroes paleomagnetic pole is statistically different from the majority of previously published poles from the British and Faroes igneous provinces, and we suggest that these older data should be used with care.     

Paleomagnetic study of Cambrian Potsdam Group sandstones,St. Lawrence Lowlands,Quebec

We report paleomagnetic results from oriented drill core samples collected at 10 sites (80 samples) from the Covey Hill and 19 sites (96 samples) from the overlying, fossiliferous Cha?teauguay Formations of the gently dipping Late Cambrian Potsdam Group sandstones exposed in the St. Lawrence Lowlands of Quebec. Stepwise thermal demagnetization analyses ave revealed the presence of two predominant groups of coherent magnetizations C-1 and C-2, after simple correction for bedding tilt. The C-1 group magnetization is a stable direction (D=332°, I=+18°) with unblocking temperatures (T_UB) between 550 and 650°C, present in the older Covey Hill Formation; this direction is probably a chemical remanence acquired during the Covey Hill diagenesis and carried predominantly by hematite. The C-2 group magnetization (D=322°, I=+9°) is present at 13 sites of the younger Cha?teauguay Formation; this is probably carried by magnetite and represents a penecontemporaneous, depositional DRM, characterized by T_UB spectra 400–550°C. We believe that C-2 is relatively younger than C-1 based on a combination of arguments such as the presence of opposite polarities, internal consistency, similarity and common occurrence of C-1 and C-2 respectively in the Covey Hill and Cha?teauguay members. The corresponding paleomagnetic poles C-1 (46°N, 149°E; dp, dm=3°, 5°) and C-2 (37°N, 156°E; dp, dm=2°, 5°) are not significantly different from most of the other Late Cambrian (Dresbachian-Franconian) poles derived from sediments exposed in the southern region (Texas) of the North American craton which are also believed to have been deposited during Croixian Sauk sea transgression similar to the Potsdam sandstones. Although adequate faunal control is lacking (in particular for the Covey Hill Formation), this comparison with the Cratonic poles suggests a Late Cambrian age to the Potsdam poles. The agreement between the results also gives the evidence for internal consistency of cratonic poles at least for Late Cambrian.The incoherent C-3 group remanence (D=250°, I=?15°) is commonly present at 7 sites in both the formations; this may not correspond to a reliable paleomagnetic signal. The other remanence C-4 (D=180°, I=+10°) is found only at 3 sites located in the uppermost stratigraphic levels of the Cha?teauguay Formation; the corresponding paleomagnetic pole (40°N, 107°E) does not differ significantly from the Ordovician and some Late Cambrian poles. The present data are insufficient to resolve a problem in apparent polar wander for Middle and Late Cambrian time posed by the existence of high-latitude poles for some strata of Middle Cambrian age and low-latitude poles for some strata of Late Cambrian age.     

Seasonal variations in pCO2 and its controlling factors in surface seawater of the northern East China Sea

Surface partial pressure of CO₂ (pCO₂), temperature, salinity, nutrients, and chlorophyll a were measured in the East China Sea (ECS; 31°30′–34°00′N to 124°00′–127°30′E) in August 2003 (summer), May 2004 (spring), October 2004 (early fall), and November 2005 (fall). The warm and saline Tsushima Warm Current was observed in the eastern part of the survey area during four cruises, and relatively low salinity waters due to outflow from the Changjiang (Yangtze River) were observed over the western part of the survey area. Surface pCO₂ ranged from 236 to 445 μatm in spring and summer, and from 326 to 517 μatm in fall. Large pCO₂ (values >400 μatm) occurred in the western part of the study area in spring and fall, and in the eastern part in summer. A positive linear correlation existed between surface pCO₂ and temperature in the eastern part of the study area, where the Tsushima Warm Current dominates; this correlation suggests that temperature is the major factor controlling surface pCO₂ distribution in that area. In the western part of the study area, however, the main controlling factor is different and seasonally complex. There is large transport in this region of Changjiang Diluted Water in summer, causing low salinity and low pCO₂ values. The relationship between surface pCO₂ and water stability suggests that the amount of mixing and/or upwelling of CO₂-rich water might be the important process controlling surface pCO₂ levels during spring and fall in this shallow region. Sea–air CO₂ flux, based on the application of a Wanninkhof [1992. Relationship between wind speed and gas exchange over the ocean. Journal of Geophysical Research 97, 7373–7382] formula for gas transfer velocity and a set of monthly averaged satellite wind data, were −5.04±1.59, −2.52±1.81, 1.71±2.87, and 0.39±0.18 mmol m⁻² d⁻¹ in spring, summer, early fall, and fall, respectively, in the northern ECS. The ocean in this study area is therefore a carbon sink in spring and summer, but a weak source or in equilibrium with the atmosphere in fall. If the winter flux value is assumed to have been the mean of autumnal and vernal values, then the northern ECS absorbs about 0.013 Pg C annually. That result suggests that the northern ECS is a net sink for atmospheric CO₂, a result consistent with previous studies.     

La β-decay constant estimated from geochronological studies

LaCe ages are reported for two sets of Finnish pegmatites, Lövböle and Mustikkamäki, and for an Amiˆtsoq gneiss, Greenland. When λ_β¹³⁸La value (2.29 × 10⁻¹² yr⁻¹) obtained by radioactivity measurement [1] is used for the chronological calculation, the LaCe ages (2129, 2325, 3271 Myr) evaluated for these rocks are 18–35% older than the SmNd ages for the same samples. To make the LaCe age fit to the SmNd age for the same sample, a new value of (2.77 ± 0.21) × 10⁻¹² yr⁻¹ is evaluated for λ_β¹³⁸La. In this calculation, the LaCe and SmNd ages reported for a Bushveld gabbro [2] have been also taken into account together with those for the Lövböle pegmatite and the Mustikkamäki pegmatite, while the Amiˆtsoq gneiss (GGU110999) has been omitted because of the complicated thermal history of this sample.