GP51B-0750
A BAYESIAN ARCHAEOMAGNETIC FIELD MODEL IN EUROPE FOR THE PAST 2000 YEARS
A detailed representation of the Earth's magnetic field and its secular variation is a crucial requirement for understanding the Earth's core dynamics. The geomagnetic field is usually represented by means of spherical harmonics using a worldwide distribution of sparsely distributed data of unequal quality. It is thus generally believed that only the first spherical harmonic degrees are statistically significant. During the past decades, a significant effort has been made to enrich the available database, in particular by measuring the direction and intensity of the magnetic field imprinted in archaeological artefacts. The data density is comparatively higher in Europe for the past 2000 yr than in any other part of the world. This situation offers the opportunity to regionally resolve the magnetic field at spatial scales higher than permitted by spherical harmonics. In principle, with a regional approach, we would be able to better explore the reliability of the 'rapid' spatial and time variation observed during the last 2000 yr in Europe. Recently, Pavon-Carrasco et al. (2008) proposed a regional archaeomagnetic field model in Europe based on a Spherical Cap Harmonic representation (SCHA). However, we argue that SCHA suffers from serious drawbacks that require strong prior information to be stabilized. Some theoretical problems are avoided if the Revised Spherical Cap Harmonic Analysis in 2D is used instead (R-SCHA2D; Thébault, 2008). In this paper, we represent the archaeomagnetic directional and intensity data with R-SCHA2D basis functions in a spherical cap centred over Europe with a 50 yr sliding window covering the past 2000 yrs. In addition to this regional modelling approach, we favour a Bayesian inversion in order to avoid the recourse of strong a priori information. This allows us to tackle non-linear, and sometimes underdetermined, inverse problems, which often represent a tedious issue. The marginal probabilities of each regional parameter are obtained with a Monte Carlo Markov Chain for all epochs. This directly translates into a probability distribution for the magnetic field components. This Bayesian model is then compared with archaeomagnetic records and the available Bayesian European Palaeosecular Variation Curves. Using the most up-to-data archaeomagnetic European database, we are thus able to assess the reliability of the 'rapid' spatial and temporal archaeomagnetic field variations occurring over Europe.
GP51B-0751
The Archaeomagnetic Field and the Historical Field: Never the Twain Shall Meet?
We compare predictions of the geomagnetic field in Europe from 1590 to 1800 AD from a field model from archaeomagnetic data in Europe, developed for archaeomagnetic dating and from the historical field model gufm1 (Jackson et al., 2000). A consistent discrepancy pre-1800 AD is observed in inclination (I), with gufm1 producing higher values of I than those predicted from archaeomagnetic data. Between 1590 and 1800 AD there are 354 archaeomagnetic data and 133 historical data; in general, the archaeomagnetic data are in Eastern Europe and the historical data in Western Europe (and the Atlantic). Thus if the two data sets were combined, they would provide much better spatial resolution; however, the apparent discrepancy between the datasets makes simple simultaneous modelling problematic. Is this offset to higher I values an indication of a systematically lower value of I measured in archaeomagnetic materials? Alternatively is this discrepancy caused because of inadequate error estimates? When long time series of historical data at single locations, e.g. London and Paris are plotted, it can be clearly seen that the scatter in I is substantially higher pre-1800 AD than post-1800 AD, but the quality is still likely to be higher than from archaeomagnetic measurements. We are currently testing improvements to the handling of errors in our models by considering correlation of temporal errors and also using the gradient of the current model iteration to convert temporal errors into measurement errors. We present comparisons of models produced using these different error calculations and discuss their effect on the discrepancy between archaeomagnetic models and gufm1.
GP51B-0752
Integrated Magnetic and Raman Spectroscopy Study of Pre-Columbian Potteries From the Mesoamerican Formative Village of Cuanalan, Teotihuacan Valley, Mexico
I report a detailed archeomagnetic and micro-Raman spectroscopy investigation on pre-Columbian pottery fragments from Cuanalan (a Formative village in the Valley of Teotihuacan, Central Mexico). Available radiocarbon ages range from 2320 (80) to 2060 (90) BP. Continuous low-field susceptibility vs. temperature curves performed in air indicate Ti-poor titanomagnetites as magnetization carriers. Few samples, however, show two ferrimagnetic phases with Curie temperatures compatible with both Ti-poor and Ti-rich titanomagnetites. Hysteresis parameter ratios fall essentially in the pseudo-single-domain region, which may indicate a mixture of multi-domain and significant amount of single domain grains. Archeointensity values have been determined from seven pottery fragments (47 samples) out of 10 (70 samples) analyzed. Thermoremanent magnetization (TRM) anisotropy and the cooling rate effect upon TRM intensity acquisition have been investigated in all the samples. The mean archeointensity values obtained in this study range from 24.2 to 40.0 microT and corresponding Virtual Axial Dipole Moments (VADMs) from 4.8 to 8.0 (1022 Am2). This corresponds to mean VDM value of 5.9 x 1022 Am2 which is lower than present day field strength and predicted values by global models (Korte et al. 2005) and data compilation (Genevey et al. 2008).
GP51B-0753
A Regional Archaeomagnetic Model for Europe for the Last 3000 Years: the SCHA.DIF.3K Model. Applications to Archaeomagnetic Dating.
The available European database of archaeomagnetic field values and instrumental data has been used to produce a regional model for the geomagnetic field in Europe for the last 3000 years. This new model, SCHA.DIF.3K, constitutes an improvement with respect to the initial regional archaeomagnetic model SCHA.DI.00-F, which used relocated values and was only valid for the last 2000 years. The new model has been obtained by least squares inversion of palaeomagnetic data using spherical cap harmonics (SCHA) for the spatial representation of the field and sliding windows in time. An algorithm has been developed to jointly model the three archaeomagnetic elements declination, inclination, and intensity. The resulting model provides the direction and intensity of the Earth's magnetic field over the European continent, Northern Africa and western Asia for the last 3000 years. The fit to the European archaeomagnetic database is more accurate than that provided by global archaeomagnetic models. In addition, this model represents a new step forward in archaeomagnetic dating studies (since the relocation error is avoided) and can also be used to study the rapid changes of the geomagnetic field (archaeomagnetic jerks) that have been recently proposed.
GP51B-0754
Testing the Accuracy of the European Archaeomagnetic Database
Prior to the era of instrumental measurements, the variations of both direction and intensity of the Earth's magnetic field can be inferred from palaeomagnetic studies. Archaeological heated structures, such as pottery or domestic kilns or ceramics, provide the most suitable records to study the geomagnetic secular variation over the past few millennia. These archaeological artefacts indeed often possess a stable thermoremanent magnetisation and for some of them they can be accurately dated using archaeological or historical constraints. The number of available archaeomagnetic data is currently rather small at the worldwide scale. The European continent has comparatively the highest concentration of archaeomagnetic results for the past 2000 years. In this study, we estimate the reliability of the present European archaeomagnetic database using different approaches. We first make an analysis of the a priori data information, for instance those concerning the measurement and dating uncertainties provided in the database. In a second step, we perform an iterative regional modelling in order to regionally fit the archaeomagnetic data. This procedure yields a posteriori data uncertainties that can easily be compared with the a priori information. The iterative reweighted least-squares method is further of interest for evaluating the weight of a given archaeomagnetic dataset in the modelling process. We hence use the recent Revised Spherical Cap Harmonic Analysis in 2 dimensions (R-SCHA2D, Thebault, 2008), with a spherical cap covering all available European data. Our preliminary results are two-folds. Firstly, the uncertainties reported in the database seem to be generally underestimated. This leads to a poorest spatial resolution than what was initially expected from such a dense data distribution. Secondly, the distribution of the a posteriori data errors does not seem to be randomly distributed in Europe. We will show several spatio-temporal maps illustrating the differences between a priori and a posteriori errors.
GP51B-0755
Uncertainty Estimates for Millennial Scale Geomgagnetic Field Models
Continuous geomagnetic field models spanning several millennia have recently been developed using various selections of archeo- and paleomagnetic data and their inferred ages. In each case the geographic and temporal distribution of available data is far from uniform and both the magnetic data and ages have large uncertainties. We estimate error bars for both the models and their predictions using two statistical resampling techniques and a combination thereof. First, we used what we call the spatial and temporal (ST) bootstrap yielding different spatial and temporal distributions taken randomly from the original dataset. Second, we kept the original (temporal and spatial) distribution of data, but varied each datum randomly within the expected distributions of uncertainty in both the magnetic observation and assigned ages. We call this the magnetic/age (MA) Bootstrap. We produced a large number of models based on resampled data using each of the ST and MA bootstrap methods and then obtain standard deviations for both global model coefficients and predictions of field components. The ST and MA methods yield model uncertainties of the same order of magnitude. A sequential combination of MA and ST resampling takes into account the influence of uncertainties in both magnetic elements and ages as well as the unsatisfactory data distribution. We present global and regional results from this analysis and compare the uncertainties obtained from model predictions to the assigned data errors. The uncertainties obtained for magnetic field elements vary depending on whether they are obtained by error propagation from uncertainties in the model coefficients or by computing the standard error in the individual element predictions for all resampled models. The propagated uncertainties do not currently allow for covariance among the coefficients. Hence, they can be too large in some geographic regions and time intervals with good data coverage. Individual element uncertainty predictions incorporate any such covariance automatically, and can in principle better accommodate regional variations in model accuracy.
GP51B-0756
Investigating the use of Pyroclastics for Palaeointensity Determinations
Palaeointensity experiments are time consuming and prone to failure, making it difficult to obtain good data. This limits the global palaeointensity database, which has inadequate spatial and temporal resolution. Developing reliable yet efficient experimental techniques is important for improving data quality, as is expanding the range of materials to investigate. By characterizing and understanding the behaviour of different materials, such as lithic clasts found within pyroclastic flows, we can assess their potential usefulness as palaeomagnetic recorders. Pre-existing lithic fragments are reheated as they are incorporated into pyroclastic density currents (PDCs) and their subsequent deposits. This partially resets their magnetization to record the ambient field at the time of eruption. The explosive nature of volcanic eruptions associated with PDCs frequently results in a wide range of lithic fragments in such deposits and the high temperatures involved (up to ~1200°C) can allow acquisition of an entirely new magnetization. We have studied the pyroclastic deposits of four historic volcanoes: Láscar in the Chilean Andes, Colima in Mexico, Mount St. Helens, USA, and Vesuvius, Italy. At Láscar, the emplacement temperature of the deposits exceeds the Curie temperature of the magnetic minerals, which maximizes the potential temperature range for obtaining palaeointensity determinations. Triple heating palaeointensity experiments, with strict selection criteria, yield a mean palaeointensity of 24.3±2.1 μT (N=22), which is in good agreement with the expected value of 24 μT. This indicates that the method has promise. The use of multiple lithologies in a single palaeointensity determination also provides confidence that the result is not biased by alteration within one of the lithologies. Pyroclastics, however, still suffer from the problems associated with palaeointensity experiments on lava flows. Samples collected from Mount St. Helens show a high degree of thermal alteration during the experiment, resulting in the failure of all the samples. At Vesuvius the influences of MD grains, magnetic interactions and CRM overgrowth lead to failure of the palaeointensity experiments. Working with pyroclastic materials has one additional pitfall, as is illustrated by data from Colima. Discrimination between hot pyroclastic deposits and cold mudflow deposits based on field observations alone can be extremely difficult. Inconsistent palaeomagnetic directions recorded by samples from Colima suggest that they are mudflow deposits, the cold nature of which precludes them from meaningful palaeointensity analysis.
GP51B-0757
A new Technique for Probing Thermal Alteration in Paleointensity Studies: Double Thermal Demagnetization of 3-components of Anhysteretic Remanent Magnetization (ARM)
It is well known that changes in thermal remanent magnetization (TRM) due to thermal alteration of magnetic grains are the main reasons to result in failure in traditional type paleointensity studies. Blocking temperature (Tb) and/or unblocking temperature (Tb) of magnetic carriers are the comprehensive parameters in evaluating such changes. We have recently developed a double thermal demagnetization of 3-components ARM technique for checking thermal alteration in paleointensity determination experiments. By comparison thermal demagnetization curves of ARM before and after a given maximum temperature Tn, it is possible to check whether thermal alteration has occurred. It is also possible to carry out a thermal alteration check at every successive thermal demagnetization step Ti by acquiring a new ARM at room temperature. Three components thermal demagnetization of ARM technique is thus very useful for distinguishing magnetic behavior of microphenocryst grains (which are crystallized in early stage before eruption of lava) and intersertal titanomagnetite grains (which are crystallized in secondary stage after eruption of lava). We will present results from our case study on the Columbia River Basalt to illustrate these points.
GP51B-0758
Why are the Results of Microwave and Thermal Paleointensity Experiments Systematically Different?
The use of microwave radiation in place of conventional heating in absolute geomagnetic palaeointensity studies is a useful innovation since it minimises the potential for alteration to bias the results and since the configuration of the apparatus allows experiments to be performed quickly on a sample-by-sample basis. It has been argued both theoretically and empirically that the microwave (de)magnetisation process is equivalent to its thermal counterpart and several studies have produced indistinguishable results using the two methods. However, here it is shown that the majority of studies in which microwave and thermally-derived results can be directly compared exhibit systematic differences in some or all of the results obtained. Specifically, the thermal results tend to be higher than the microwave results derived from the same rock units or even the same core samples. Furthermore, the thermal experiments more frequently produce concave-up Arai plots. Explanations for these observations which rely on some intrinsic difference between the microwave and thermal approaches are not convincing in the majority of cases. Instead, it is argued that the offsets are not in large part related to the issue of whether microwave or thermal energy is used to (de)magnetise the samples but are instead related to the particular measurement protocol used in the experiments and the fact that most of the samples do not contain only ideal single-domain magnetic grains and therefore exhibit some multidomain thermoremanence (MD TRM) behaviour. A phenomenological model of MD TRM can be used on a case-by-case basis to demonstrate that differences in the results of microwave and thermal experiments are generally consistent with the various protocols employed in the two types of experiments producing results which are affected by MD TRM behaviour to different degrees. Broadly, microwave experiments favour the perpendicular protocol of palaeointensity measurement whereas the thermal studies by contrast use double heating protocols. Here it is shown that samples exhibiting MD TRM behaviour will tend to produce slight underestimates of the palaeointensity in the former type of experiment but can produce significant overestimates in the latter type. The findings of this study suggest that in most cases where discrepancies between microwave and thermal results have been observed, the microwave result was the more reliable of the two. They also suggest that biasing of Thellier-type palaeointensity results to high values by MD TRM behaviour may be more widespread and more severe than previously believed. Future palaeointensity studies should therefore seek to design their experiments to minimise this source of bias as well as those associated with thermally-induced alteration and secondary magnetisations. Fortunately, this can be done with relative ease and without losing the benefits of Thellier-type approaches that have led to them gaining their current levels of popularity.
GP51B-0759
A Possible Record of the Laschamp Excursion From High Sedimentation Rate Sediments in the Southern Hemisphere
A record of a geomagnetic excursion has been obtained from Core MD07-3128 taken at (52 degrees S) off the Pacific coast of Southern Chile, during the IMAGES XV-MD159-PACHIDERME cruise of the R/V Marion Dufresne (IPEV). The primary component of magnetization has been isolated after AF demagnetization beyond 20 mT, after removal of a strong (80 per cent of the NRM) secondary component. Between 20 mT and the end of the demagnetization a single component was precisely defined on 7-9 steps of demagnetization. The directional excursion is recorded over about 1 meter of sediments (between 19,65 and 21,5) and corresponds to a longer marked low in the relative paleointensity record. Radiocarbon datings (for which reservoir influences cannot be excluded), extend to 36.3 kyr BP at 18 meters and indicate large changes in sedimentation rates and a thick glacial interval. Linear extrapolation of the last 2 dates downcore, gives an age of 40.7 at 20 m. The event , therefore , is most probably the Laschamp Excursion. Details of the directional and relative paleointensity changes will be discussed.
GP51B-0760
Paleosecular Variation of Plio-Pleistocene Lavas from the Loiyangalani Region of Kenya
The data reported here is part of a study of Pliocene-Pleistocene lavas in Kenya to document the paleosecular variation and time-averaged geomagnetic field direction near to the Equator. We sampled 32 sites (10 oriented cores each) in lavas to the south and the northeast of Loiyangalani that are mapped and dated as Plio-Pleistocene in age (less than ~5 Ma) and associated with Mt. Kulal and the Longipi eruption centers. The samples from this collection were returned to the US, sliced into samples and progressively demagnetized using alternating field demagnetization. The Loiyangalani sites yielded excellent results and are seemingly unaffected by lightning, which seems to be infrequent at this latitude, in this arid environment; all but one site gave acceptable data with an alpha95 of 10° or less. There are 17 reverse sites (Dec = 183.4°, Inc = 0.9°, alpha95 = 6.7°) and 15 normal sites (Dec = 358.4°, Inc = -1.2°, alpha95 = 4.7°). The reversal test is positive suggesting that the normal and reverse polarity populations both represent a reasonable time average. The site means were combined yielding an overall mean direction of Dec = 1.1°, Inc = -1.1°, alpha95 = 4.1°. The inclination is shallower than expected for a geocentric axial dipole field (delta I = -6°); accordingly, the site VGPs give a mean pole position at Lon = 205.1° E, Lat = 86.8° N, Alpha95 = 3°, which is significantly far-sided with respect to the geographic axis. The angular standard deviation of the VGPs is 9.3°, which is a relatively low angular dispersion compared to most PSVL models such as Model G.
GP51B-0761
Geomagnetic field variations during the last 400 kyr in the western equatorial Pacific: Paleointensity-inclination correlation revisited
A paleomagnetic study was conducted on four piston cores newly obtained from the West Caroline Basin in the western equatorial Pacific in order to investigate variations in paleointensity and inclination during the last 400 kyr. An inclination-intensity correlation was previously reported in this region using giant piston cores, but the quality of the paleomagnetic data of the younger end, the last ca. 300 kyr, was needed to be checked because the upper part of the giant piston cores could suffer from perturbation by oversampling. Age control is based on the oxygen-isotope ratios for one core and inter-core correlation using relative paleointensity for other cores. The mean inclinations of the four cores show negative inclination anomalies ranging from -5.2 to -11.2 degree. The western equatorial Pacific is documented as a region of a large negative inclination anomalies, and the observed values are comparable to those expected from the time-averaged field (TAF) models [Johnson and Constable, 1997; Hatakeyama and Kono, 2002]. Stacked curves of paleointensity and inclination were constructed from the four cores. It was confirmed that geomagnetic variations on the order of 10 to 100 kyrs occur in inclination as well as paleointensity. A cross-correlation analysis showed that significant in-phase correlation occurs between intensity and inclination for periods longer than about 25 kyr, and power spectra of both paleointensity and inclination variations have peaks at ~100 kyr periods. The regional paleointensity stack with higher resolution than the Sint-800 stack [Guyodo and Valet, 1999] should be useful for paleointensity-assisted chronostratigraphy.
GP51B-0762
The Brunhes/Matuyama polarity transition recorded as Be-10 flux changes in deep-sea sediments
Fluxes of meteoric cosmogenic radionuclide, Be-10, is thought to be varied due to changes of incoming comic-ray flux modulated by geomagnetic field intensity variation. Enhanced production rate of the nuclides during a geomagnetic polarity transition period is expected as a result of the low dipole field strength. We therefore measured Be-10 concentrations in deep-sea sediments including the Brunhes/Matuyama geomagnetic polarity transition to reconstruct the detailed structures of the geomagnetic field behavior. A piston core, MD982187 was taken from the West Caroline Basin, the western equatorial Pacific Ocean, during the IMAGES IV campaign. The water depth of the site of MD982187 core is about 4600 m, which is close to the carbonate compensation depth (CCD) in this area at present (Berger et al., 1976). Measurement of Be-10 was conducted using the accelerator mass spectrometry (AMS) of the University of Tokyo, Japan. The result shows significant increase of Be-10 concentration during the polarity transition, indicating that the geomagnetic field intensity was low during this interval. In detail, well-defined double highs of Be-10 concentration are recognized. These highs are thought to correspond to the B/M polarity boundary and the "precursor" event, 15 kyr before the M/B boundary (e.g., Hartl and Tauxe, 1996; Singer et al., 2005), respectively. This feature is very similar to the relative paleointensity record of MR982187 core by Yamazaki and Oda (2005) and other published relative paleointensity records of the Brunhes/Matuyama geomagnetic polarity transition, indicating that Be-10 concentration of the deep-sea sedimentary sequence well records the variation of the geomagnetic field intensity. However, ca. 18 cm of clear depth offset between the Be-10 concentration and the relative paleointensity record was observed from the same sedimentary sequence of MR982187 core. This indicates that the relative paleointensity record of MR982187 core is offset by ca. 18 cm below the actual level of the polarity transition, which is thought to be the paleomagnetic lock-in depth effect.
GP51B-0763
The Upper Olduvai boundary recorded in Chinese loess at Lingtai: Evidence for transitional geomagnetic field behaviour or delayed remanence acquisition?
A detailed remanent magnetisation record of the upper Olduvai geomagnetic polarity boundary from loess/palaeosol sediments at Lingtai (central Chinese Loess Plateau) is presented. The polarity transition occurs within loess layer L25 (mean sedimentation rate 11.5 cm/kyr), where rock magnetic parameters such as natural remanent magnetisation (NRM) intensity, F-factor and magnetic low-field susceptibility show considerable fluctuations. The characteristic remanent magnetisation record, obtained after thermal cleaning, yields 23 polarity changes within L25, 19 of them occur within a depth interval corresponding to about 10,000 years. Compared to marine data, the upper Olduvai boundary appears to be shifted below its expected stratigraphic level of occurrence (i.e. marine oxygen isotope stage 63 or resp. palaeosol S24 cf. Heslop et al. (2000), which should correspond to an interglacial climate period. Although lags between benthic δ18O records and magnetic susceptibility records may possibly occur, they cannot explain the downward shift of the geomagnetic polarity boundary. Antiphase behaviour of climatic events can be excluded in order to explain the observations. Theoretical considerations, lacking observational evidence from other upper Olduvai records and from geodynamo dipole models give no evidence for such complicated transitional field behaviour as suggested from the characteristic remanent magnetisation record. Instead, it is proposed that complex acquisition processes of detrital and pedogenic remanences, which are almost always co-existent on the central Chinese Loess Plateau, causes a long delay in the acquisition of the total remanence, implicating a strong dependence of the acquired remanence on lithogenic variations. It is concluded that loess/palaeosol sediments from the central Chinese Loess Plateau do not track confidentially geomagnetic field behaviour during polarity transitions, palaeosecular variation and short geomagnetic excursions.
GP51B-0764
Paleomagnetic Study of the Bundelkhand Dykes Swarms (Madhya Pradesh, Central India)
The early Proterozoic Bundelkhand massif granite of Central India (ca. 2550 Ma) is extensively intruded by suites of NW--SE and NE--SW trending mafic and ultramafic dykes (ca. 2000 Ma). These dolerite dykes display chilled margins and do not shown any interaction with the host rocks. We have recently collected a series of samples from 21 sites distributed in 16 dykes. Both thermal and alternating field demagnetizations were conducted in order to check for the consistency of the results, but also to provide some preliminary clues on magnetic properties and their suitability for studies of absolute paleointensity. In most cases a characteristic component was isolated by thermal demagnetization beyond 450°C. Three groups of directions were identified. The first one lies in the horizontal plane with a declination pointing south. The second one shows a vertical downward inclination. The third one has been recorded by the samples from the ENE--WSW Great Mahoba Dyke with a northward declination (D=~20°) and a negative inclination (I=~30°). Though there is no clear correlation between the dykes strikes and the paleomagnetic directions, except for the Great Mahoba Dyke, we cannot rule out completely the possibility of relationships in the absence of accurate geochronometric determinations. Paleointensity and geochronologic experiments are under way. Implications of the present study would concern the True Polar Wander Hypothesis, Apparent Polar Wander of the Bundelkhand Craton, and paleointensity of the Earth magnetic field around 2000 Ma.
GP51B-0765
Detection and Reliability of Model Signatures in Paleomagnetic Time Series
We model the geomagnetic field as a transient phenomenon, and therefore relative paleomagnetic intensity as a set of growths and decays, punctuated by occasional steady states. Our model originates from theory and experimental observations of rotating parametric instabilities that are possible in Earth's fluid outer core due to tidal and precessional strains. Since evaluation of this model using paleomagnetic time series depends on errors in both paleointensity and chronology, our analysis includes both sources of error. Accordingly, we have constructed an algorithm that recovers growth and decay rates from paleomagnetic time series. Our algorithm has been applied to both simulated data and relative paleointensity from ODP983 and NAPIS- 75. Analysis of simulated data returns estimates that agree with expected rates to within the calculated error. For real data from single records, recovered rates are consistent with earlier results using less objective methods, but can have errors as large as the rates themselves. Thus reliable rates and uncertainties of growth and decay rates need to be found from multiple time series and correlated to identifiable events.
GP51B-0766
Paleomagnetic Sampling of Holocene Silicic Eruptive Products at Medicine Lake Volcano, Southern Cascade Range
Thick blocky lava flows, particularly those that are silicic and thus erupt at lower temperatures than mafic andesitic and basaltic flows, present special challenges for paleomagnetic sampling when no erosion or artificial cuts expose their solid interiors. These viscous flows in many cases continue to move after the exterior carapace has cooled below its Curie temperature, leaving a mantle of loose blocks with variably tilted directions of magnetization. We drilled a variety of uneroded, late Holocene, silicic eruptive products at Medicine Lake volcano: crease structures on flow surfaces, near-vent agglutinated spatter and bombs, mafic magmatic inclusions (MI), prismatically jointed peripheral blocks, and levee environments including channel walls and shear planes exhibiting plastic morphology. At the 950-yr-old Glass Mountain fissure rhyolite and dacite, 14 sites were sampled in the main flow and adjacent domes, including 8 agglutinate deposits, 2 flattened crease structures, 2 sets of MI in the wall of a deep crack, and 2 levee walls. The latter 2 sites yielded tilted magnetic directions and one site had unacceptable scatter. The remaining 11 sites gave excellent results with an a95 of 2.1°. At the ~1000-yr-old Little Glass Mountain fissure rhyolite, 10 sites were drilled. Half were crease structures, 4 were agglutinate, and one was in prismatically jointed blocks that had presumably tumbled off the flow when they were hot. The latter attempt failed, as did one dome block site, but 8 sites yielded an a95 of 2.4°. Similar low values of a95 were obtained for other late Holocene rhyolites to andesites. Viscous lava flows of andesite composition can present similar difficulties to those exhibited by lava flows of dacite or rhyolite, where identifying plastic behavior of the lava flow in a dominantly brittle environment can be problematic. By comparison, 2 sites in the ~3-ka basalt of Black Crater and Ross Chimneys easily produced an a95 of 1.2°. Lava flows of mafic composition or welded silicic ash-flow tuffs present fewer sampling challenges and more opportunities to sample relatively quickly chilled magmatic products. Sampling a variety of volcanic flow structures may be necessary to check the accuracy of the results from blocky rhyolite and dacite lava flows.
GP51B-0767
Structurally induced errors in paleomagnetic analysis of fold and thrust belts: Types, causes and detection techniques.
Paleomagnetic vectors are unique kinematics indicators allowing for the real understanding of the lateral transference of deformation processes and they are essential for a real 3D understanding of fold and thrust belts. The association with the bedding surface gives the only 3D reference system able to unambiguously relate the deformed and undeformed stages and their implications are, until now, relatively unexplored in structural geology. However paleomagnetic data are sometimes, misinterpreted or ignored due to the lack of reliability of some databases, where a geometric control of errors seems evident from the structural point of view. An analysis of the implicit assumptions in paleomagnetic studies of fold and thrust belts reveals three possible sources of error with an intrinsic structural (geometric) control: Assumption 1) The laboratory procedures are able to completely isolate of the original paleomagnetic vectors When this fails, the subsequent overlapped paleomagnetic directions (eg. primary record and the recent overprint) will display both declination and inclination errors, that will be controlled by the fold axis orientation, the degree of flank rotation (dip), the primary magnetic polarity as well as the degree of vector overlapping. Assumption 2) The rigid-body behavior during deformation and the absence of rock volume changes. When the rock volume undergoes active internal deformation during folding or shearing, the deformed paleomagnetic vectors will display again declination and inclination errors, but both polarities will behave similarly. In this case the errors will depend on the relation between the primary field orientation and the deformation tensor, which in fact, can be reduced to the orientation and magnitude of the shear in most cases. Assumption 3) The bedding correction is able to restore the bedding-vector couple to the ancient (paleo)geographical reference system. This restoration may fail in complex deformation zones affected by non-coaxial or inclined axes of deformation (conical, plunging, forced folds, etc…). Here, only the paleomagnetic declination will show deviations (spurious or apparent rotations) that will be a function (non-coaxiality case) of the obliquity and magnitude (external rotation) of the deformation axes. These errors may cause severe problems in the, so-called, stability tests: all three sources of error will change the result of the fold test resulting in a false synfolding magnetization that may display also an artificial oroclinal bending of the dataset. A fictitious negative reversal test may be derived from the overlapping of vectors (very severe) and the non-rigid-body behavior of the rock volume (internal deformation). The different errors can be detected by applying simple geometric techniques like the stereographic scattering, the inclination vs dip and the declination vs strike diagrams, the fold test, the small circle reconstruction during unfolding and the geometry of the dispersion of the net tectonic rotation axes. A list of reliability criteria to assess the quality of the paleomagnetic data in fold and thrust belts will be proposed to avoid these errors.
GP51B-0768
A controlled evaluation of magnetite formation in synthetic submarine basaltic glass: implications for paleointensity studies
Submarine basaltic glass (SBG) has gained increasing popularity in recent years as a material for use in magnetic paleointensity studies, as it behaves ideally in a Thellier-type paleointensity experiment. Questions have been raised regarding the timing and temperature of (titano-)magnetite formation, in part because the Ti content of the magnetite is considerably lower than that found in most crystalline basalt. We therefore generate a set of synthetic SBG under controlled cooling rate and oxygen fugacity conditions to evaluate both the timing of magnetite formation, as well as its suitability as a paleofield recorder. The starting material is a crushed basalt from the Pacific-Antarctic Ridge (FeO* = 11.8 wt%). Samples were melted and homogenized at 1400°C under QFM or QFM+1 conditions and were then allowed to cool at rates that varied between 5 and 140 °C min-1. Three additional compositions (two synthetic and one natural) with FeO* ~ 9.5 wt% were quenched to glass at ~200°C min- 1 under QFM conditions. Sample splits were reheated in air at 600°C or 750°C for 100 hrs to test the possible role of post-eruption oxidation on magnetic mineralogy. Room-temperature magnetic hysteresis and low-temperature susceptibility data are best explained by a mixture of single-domain and superparamagnetic sized grains; there are no obvious differences between synthetic and natural SBG samples. Natural remanence (NRM) and anhysteretic remanence (ARM) results from the 9.5 wt% FeO* samples show that quenched samples carry a remanence comparable to that of natural SBG. Samples reheated in air at 600°C have a magnetization intensity indistinguishable from that of the un-oxidized samples. Samples reheated to temperatures above the glass transition (~700°C) have magnetizations elevated by 1-2 orders of magnitude. Combined, these early results suggest that a magnetic phase is present at the time the glass is originally quenched from the melt. Furthermore, additional magnetic grains do not appear to nucleate or grow at temperatures below the glass transition, even under oxidizing conditions. We will additionally present results of Thellier-type paleointensity experiments to evaluate the ability of the synthetic samples to recover the known field. Blocking-temperature distributions will allow us to estimate Ti content and will supplement preliminary TEM and SEM results.