TY - THES A1 - Michalk, Daniel M. T1 - An appraisal of a new method for the full-vector reconstruction of the Earth’s magnetic field - applied to volcanic rocks from Mexico T1 - Bewertung einer neuen Methode zur Gesamtvektor-Rekonstruktion des Erdmagnetfeldes - angewendet auf Vulkanite aus Mexico N2 - Das Magnetfeld der Erde wird durch Konvektionsströmungen im elektrisch leitfähigen, flüssigen eisenreichen äußeren Erdkern erzeugt. Eine drastische Ausprägung der dynamischen Prozesse im äußeren Erdkern sind sowohl Polaritätswechsel über geologische Zeiträume, als auch geomagnetische Feldexkursionen (kurze Umpolungen). Letztere sind in geologischen Archiven häufig unzureichend dokumentiert. Für ein verbessertes Verständnis über die Entwicklung des Erdmagnetfeldes in geologischer Vergangenheit benötigen wir Informationen über die Geometrie des gesamten Vektorfeldes, wofür neben der Bestimmung der Feldrichtungen auch die Bestimmung der absoluten Paläointensität und des Alters notwendig ist. Insbesondere Vulkanite bieten die Möglichkeit, Daten über die Richtung und vor allem auch die Intensität des Erdmagnetfeldes zur Zeit ihrer Platznahme zu gewinnen. Bisweilen ist eine genaue Charakterisierung der Entwicklung des Erdmagnetfeldes in Zeit und Raum schwer möglich, was sich in erster Linie auf den generellen Mangel an Paläointensitätsdaten zurückführen lässt. Ein Grund hierfür ist, dass die meisten Methoden zur absoluten Paläointensitätsbestimmung, auf Modifikationen der Thellier Methode basieren, welche nur auf magnetische Minerale im Einbereichs-Domänenzustand anwendbar ist und zudem hohe Ausschussraten liefert. Eine alternative Methode zur Bestimmung der absoluten Paläointensität ist die kürzlich entwickelte „multispecimen parallel differential pTRM“ (MS) Methode, welche im Vergleich zur Thellier Methode den Vorteil hat, dass sie theoretisch unabhängig ist vom Domänenzustand der magnetischen Minerale und somit auf alle Vulkanite anwendbar ist. Ein Schwerpunkt dieser Arbeit lag darauf, neue Informationen über das Auftreten und gegebenfalls die globale Gültigkeit von geomagnetischen Feldexkursionen zu gewinnen. Hierfür wurden etwa 75 Lavaflüsse des Transmexikanischen Vulkangürtels für paläomagnetische Studien beprobt. Eine Korrelation der mittleren Paläorichtungen von 56 mexikanischen Laven mit einer um Feldexkursionen ergänzten geomagnetischen Polaritätszeitskala, lieferte Hinweise auf 4 Exkursionen. Ein bedeutendes Ergebnis dieser Arbeit sind annährend komplett inversen Richtungen zweier Laven der Brunhes Chron. Dies gibt einen Hinweis darauf, dass diese Exkursionen kurze Zeitintervalle inverser Polarität mit globaler Gültigkeit repräsentieren könnten. Ein weiterer Schwerpunkt der vorliegenden Arbeit war, die neue MS Methode auf ihre Anwendbarkeit und Genauigkeit hin zu testen. Hierfür wurden Paläointensitätsexperimente an 11 historischen Laven aus Mexiko und Island durchgeführt. Ein Vergleich der Paläointensitäten mit Daten von magnetischen Observatorien ergab, dass die MS Methode einen generellen Trend zur Überschätzung der Paläointensität aufweisst, welcher anhand von komplementierenden gesteinsmagnetischen Daten mit magnetischen Mineralen im Mehrbereichsteilchen-Zustand in Verbindung gebracht werden konnte. Diese Beobachtung liefert demnach einen ersten Beweis dafür, dass die MS Methode möglicherweise nicht wie ursprünglich angenommen unabhängig vom Domänenzustand der Trägerminerale ist. Im weiteren wurde eine Komplementierung der Richtungsdaten mexikanischer Laven durch absolute Paläointensitätsbestimmungen angestrebt. Hierfür wurde die MS Methode herangezogen und zum ersten Mal in großem Umfang auf Vulkanite mit Altern von bis zu 3,5 Millionen Jahre angewendet. Ein Vergleich mit Rekonstruktionen des Dipol-Momentes, welche auf den Daten der gegenwärtigen globalen Paläointensitätsdatanbasis basieren, ergaben, dass diese MS Daten mit hoher statistischer Wahrscheinlichkeit im Mittel etwa 30% höher sind. Die generell zu hohen Paläonintensitäten nach der MS Methode bekräftigen daher die Ergebnisse von historischen Laven dieser Arbeit, sowie anderer experimenteller Studien an synthetischen Proben, bei denen Überschätzungen von MS Paläointensitäten von bis zu 30% festgestellt wurden. Der Process, aus dem diese Überschätzung der Paläointensität resultiert ist eine Asymetrie des Entmagnetisierungs- und Remagnetisierungsprozesses heisst, dass ein effektives Entmagnetisieren während der Remagnetisierung im angelegten Laborfeld erfolgt. Diese Asymetrie scheint besonders bei pseudo-Einbereichsteilchen ausgeprägt zu sein. Es wird allerdings davon ausgegangen, dass diese Überschätzung nicht größer ist, als was man bei einem Thellier Experiment an Proben mit ähnlicher magnetischer Korngröße erwarten würde. N2 - The Earth’s magnetic field (EMF) is generated by convections in the electrically conducting liquid iron-rich outer core, modified by the Earth’s rotation. A drastic manifestation of the dynamics of this fluid body is the occurrence of geomagnetic field reversals in the Earth’s history but also geomagnetic excursions, which are more frequent features of otherwise stable polarity chrons, but often poorly constrained in the geological record. To better understand the origin of the field, we need to know how the field has varied on different geological timescales. This includes not only information about changes in the ancient field’s direction but also about the absolute intensity (palaeointensity) and the age. This palaeointensity record is needed for compiling a full-vector description of the field. A palaeomagnetic and palaeointensity study on lava flows allows gaining insights about the evolution of the EMF through time and space. However, constraining the EMF evolution over different geological timescales remains a difficult objective due to the paucity of available palaeointensity data. One new alternative approach in palaeointensity studies is the recently proposed multispecimen parallel differential pTRM (MS) method, which has potentially several advantages over the commonly used Thellier method, because it is in theory independent of magnetic domain state, less prone to biasing effects, such as thermal alteration and significantly faster to perform in the laboratory. A study of highly active volcanic regions, such as the Trans-Mexican Volcanic Belt, seems promising when attempting a full-vector reconstruction or when looking for field excursions. One aim of this thesis was to gain new information about the occurrence and global validity of geomagnetic excursions from the Brunhes- or Matuyama Chron. For this purpose some 75 lava flows from within the Trans-Mexican Volcanic Belt were sampled for palaeomagnetic analyses. The scatter of virtual geomagnetic poles from lavas younger than 1.7 Ma was used for estimating palaeosecular variation and was found to be consistent with latitude dependent Model G and other high quality palaeomagnetic data from Mexico. The palaeomagnetic mean-vectors of 56 lavas were correlated to the Geomagnetic Polarity Timescale supplemented with information on geomagnetic excursions. On the grounds of their associated radioisotopic ages, four lavas were tentatively correlated with known excursions from marine records. Two lava flows dating of Brunhes Chron were associated with the Big Lost and Delts/Stage 17 excursions, respectively. From further two flows dating of Matuyama Chron, one flow was associated with either the Santa Rosa- or Kamikatsura excursions, while the other could have been emplaced during the Gilsa excursion. The most significant outcome was the finding that both Brunhes excursional flows display nearly fully reversed directions that deviate almost 180°C from the expected normal polarity direction. This observation could indicate that in particular the Big Lost and Delta/Stage17 excursions may represent other short periods during which the field completed a full reversal for a short time, such as was previously found for other older cryptochrons or tiny wiggles. Another focus of this thesis was set on estimating the feasibility of the new MS method for routine palaeointensity determination. This was accomplished by applying the MS method to samples from 11 historical lava flows from Mexico and Iceland from which the actual field intensity was either known from contemporary observatory data, or deduced from magnetic field models. Comparing observed with expected intensity values allowed to test the accuracy of the MS method. It a was found that the majority of palaeointensity estimates after the MS method yielded results that were very close or indistinguishable within the range of uncertainty from the expected values. However, a general trend towards an overestimate in the palaeointensity was also observed, which, on the grounds of corroborating rock magnetic analyses, was associated with multidomain material. This observation was taken as first evidence that the MS method is not entirely independent of magnetic domain state, as was originally claimed. However, a second experiment in which a modification of the most widely used Thellier method was applied to sister samples from 5 Icelandic flows revealed that, in comparison to the MS method, the latter produced more accurate and statistically better defined palaeointensities. Thus, from these first results, the MS method appeared as a viable alternative for future palaeointensity studies. Subsequently it was attempted to corroborate the directional record from Mexican lavas with palaeointensity data. It was possible to acquire palaeointensity estimates for 32 out of 51 investigated lava flows. These new results revealed that the new MS palaeointensities for Mexico are, with a high degree of statistical significance, around 30% higher than expected. The generally high palaeointensities seem to corroborate the results obtained from historical lava flows in this study and other previous studies on synthetic samples where domain state effects were found to cause overestimates in the palaeointensity of up to 30 per cent in the MS method. The primary process that leads to this overestimate is assigned to an asymmetry in the demagnetisation and remagnetisation process. Yet, this overestimate is expected to be no larger than what might be expected from Thellier experiments performed on samples with a given degree of multidomain behaviour. KW - Earth's magnetic field KW - palaeointensity KW - palaeomagnetism KW - rockmagnetism KW - Mexico KW - Erdmagnetfeld KW - Paläointensität KW - Paläomagnetik KW - Gesteinsmagnetik KW - Mexiko Y1 - 2009 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus-31868 ER - TY - THES A1 - Liu, Jiabo T1 - Dynamics of the geomagnetic field during the last glacial T1 - Dynamik des geomagnetischen Feldes während der letzten Eiszeit N2 - Geomagnetic paleosecular variations (PSVs) are an expression of geodynamo processes inside the Earth’s liquid outer core. These paleomagnetic time series provide insights into the properties of the Earth’s magnetic field, from normal behavior with a dominating dipolar geometry, over field crises, such as pronounced intensity lows and geomagnetic excursions with a distorted field geometry, to the complete reversal of the dominating dipole contribution. Particularly, long-term high-resolution and high-quality PSV time series are needed for properly reconstructing the higher frequency components in the spectrum of geomagnetic field variations and for a better understanding of the effects of smoothing during the recording of such paleomagnetic records by sedimentary archives. In this doctorate study, full vector paleomagnetic records were derived from 16 sediment cores recovered from the southeastern Black Sea. Age models are based on radiocarbon dating and correlations of warming/cooling cycles monitored by high-resolution X-ray fluorescence (XRF) elementary ratios as well as ice-rafted debris (IRD) in Black Sea sediments to the sequence of ‘Dansgaard-Oeschger’ (DO) events defined from Greenland ice core oxygen isotope stratigraphy. In order to identify the carriers of magnetization in Black Sea sediments, core MSM33-55-1 recovered from the southeast Black Sea was subjected to detailed rock magnetic and electron microscopy investigations. The younger part of core MSM33-55-1 was continuously deposited since 41 ka. Before 17.5 ka, the magnetic minerals were dominated by a mixture of greigite (Fe3S4) and titanomagnetite (Fe3-xTixO4) in samples with SIRM/κLF >10 kAm-1, or exclusively by titanomagnetite in samples with SIRM/κLF ≤10 kAm-1. It was found that greigite is generally present as crustal aggregates in locally reducing micro-environments. From 17.5 ka to 8.3 ka, the dominant magnetic mineral in this transition phase was changing from greigite (17.5 – ~10.0 ka) to probably silicate-hosted titanomagnetite (~10.0 – 8.3 ka). After 8.3 ka, the anoxic Black Sea was a favorable environment for the formation of non-magnetic pyrite (FeS2) framboids. Aiming to avoid compromising of paleomagnetic data by erroneous directions carried by greigite, paleomagnetic data from samples with SIRM/κLF >10 kAm-1, shown to contain greigite by various methods, were removed from obtained records. Consequently, full vector paleomagnetic records, comprising directional data and relative paleointensity (rPI), were derived only from samples with SIRM/κLF ≤10 kAm-1 from 16 Black Sea sediment cores. The obtained data sets were used to create a stack covering the time window between 68.9 and 14.5 ka with temporal resolution between 40 and 100 years, depending on sedimentation rates. At 64.5 ka, according to obtained results from Black Sea sediments, the second deepest minimum in relative paleointensity during the past 69 ka occurred. The field minimum during MIS 4 is associated with large declination swings beginning about 3 ka before the minimum. While a swing to 50°E is associated with steep inclinations (50-60°) according to the coring site at 42°N, the subsequent declination swing to 30°W is associated with shallow inclinations of down to 40°. Nevertheless, these large deviations from the direction of a geocentric axial dipole field (I=61°, D=0°) still can not yet be termed as 'excursional', since latitudes of corresponding VGPs only reach down to 51.5°N (120°E) and 61.5°N (75°W), respectively. However, these VGP positions at opposite sides of the globe are linked with VGP drift rates of up to 0.2° per year in between. These extreme secular variations might be the mid-latitude expression of the Norwegian–Greenland Sea excursion found at several sites much further North in Arctic marine sediments between 69°N and 81°N. At about 34.5 ka, the Mono Lake excursion is evidenced in the stacked Black Sea PSV record by both a rPI minimum and directional shifts. Associated VGPs from stacked Black Sea data migrated from Alaska, via central Asia and the Tibetan Plateau, to Greenland, performing a clockwise loop. This agrees with data recorded in the Wilson Creek Formation, USA., and Arctic sediment core PS2644-5 from the Iceland Sea, suggesting a dominant dipole field. On the other hand, the Auckland lava flows, New Zealand, the Summer Lake, USA., and Arctic sediment core from ODP Site-919 yield distinct VGPs located in the central Pacific Ocean due to a presumably non-dipole (multi-pole) field configuration. A directional anomaly at 18.5 ka, associated with pronounced swings in inclination and declination, as well as a low in rPI, is probably contemporaneous with the Hilina Pali excursion, originally reported from Hawaiian lava flows. However, virtual geomagnetic poles (VGPs) calculated from Black Sea sediments are not located at latitudes lower than 60° N, which denotes normal, though pronounced secular variations. During the postulated Hilina Pali excursion, the VGPs calculated from Black Sea data migrated clockwise only along the coasts of the Arctic Ocean from NE Canada (20.0 ka), via Alaska (18.6 ka) and NE Siberia (18.0 ka) to Svalbard (17.0 ka), then looping clockwise through the Eastern Arctic Ocean. In addition to the Mono Lake and the Norwegian–Greenland Sea excursions, the Laschamp excursion was evidenced in the Black Sea PSV record with the lowest paleointensities at about 41.6 ka and a short-term (~500 years) full reversal centered at 41 ka. These excursions are further evidenced by an abnormal PSV index, though only the Laschamp and the Mono Lake excursions exhibit excursional VGP positions. The stacked Black Sea paleomagnetic record was also converted into one component parallel to the direction expected from a geocentric axial dipole (GAD) and two components perpendicular to it, representing only non-GAD components of the geomagnetic field. The Laschamp and the Norwegian–Greenland Sea excursions are characterized by extremely low GAD components, while the Mono Lake excursion is marked by large non-GAD contributions. Notably, negative values of the GAD component, indicating a fully reversed geomagnetic field, are observed only during the Laschamp excursion. In summary, this doctoral thesis reconstructed high-resolution and high-fidelity PSV records from SE Black Sea sediments. The obtained record comprises three geomagnetic excursions, the Norwegian–Greenland Sea excursion, the Laschamp excursion, and the Mono Lake excursion. They are characterized by abnormal secular variations of different amplitudes centered at about 64.5 ka, 41.0 ka and 34.5 ka, respectively. In addition, the obtained PSV record from the Black Sea do not provide evidence for the postulated 'Hilina Pali excursion' at about 18.5 ka. Anyway, the obtained Black Sea paleomagnetic record, covering field fluctuations from normal secular variations, over excursions, to a short but full reversal, points to a geomagnetic field characterized by a large dynamic range in intensity and a highly variable superposition of dipole and non-dipole contributions from the geodynamo during the past 68.9 to 14.5 ka. N2 - Die geomagnetischen Paläosäkularvarionen sind ein Ausdruck dynamischer Konvektionsprozesse im flüssigen äußeren Erdkern als Quelle des Erdmagnetfeldes, dem Geodynamo. Paläomagnetischen Zeitserien erlauben daher Rückschlüsse auf die veränderlichen Eigenschaften des Erdmagnetfeldes in Raum und Zeit, von stabilen Zuständen mit relativ hoher Feldstärke und dominierender Dipolgeometrie, bei der die magnetischen Pole nur in hohen Breiten migireren, über Feldkrisen wie Intensitätsminima und geomagnetischen Exkursionen mit komplexeren Geomentrien, bishin zur vollständigen und dauerhaften Umpolung des Dipolhauptfeldes. Eine geomagnetische Exkursion, als Extremfall der Paläosekularvariationen, ist charakterisiert durch kurzfristige (<104 Jahre) Abweichungen der paläomagnetischen Richtungen von der Richtung, die ein geozentrischer axialer Dipol produziert. Die aus paläognetisch bestimmten Richtungen abgeleiteten Positionen des virtuellen geomagnetischen (Nord-) Pols (VGP) liegen dabei per Definition südlich von 45°N. Für diese Doktoarbeit wurden sechzehn Sedimentkerne aus dem Schwarzen Meer herangezogen, um das Verhalten des Erdmagnetfeldes der letzten 70 ka zu untersuchen. Die dabei rekonstruierten Feldvariationen umfassen drei geomagnetische Exkursionen: die 'Norwegian-Greenland Sea excursion' (64.5 ka), die 'Laschamps excursion' (41.0 ka), sowie die 'Mono Lake excursion' (34.5 ka). Alle drei Ereignisse sind mit ausgeprägten Minima in der Feldintensität verbunden. Insbesondere während der 'Laschamps excursion', kam es zu einer kurzfristigen (0.5 ka) aber vollständigen Umpolung, während für die 'Mono Lake excursion' nur ein stark verzerrtes Dipolfeld anzunehmen ist. Die in der Literatur postulierte 'Hilina Pali excursion' (18.5 ka) konnte trotz in diesem Zeitraum zeitlich hochaufgelöster Datenreihen aus dem Schwarzen Meer nicht verifiziert werden. Für sie konnte, ähnlich wie für die 'Norwegian Greenland Sea excursion', lediglich stärker ausgeprägte, zum Teil zeitlich beschleunigte Änderungen in Inklination und Deklination der Magnetfeldrichtung nachgewiesen werden. KW - paleomagnetism KW - geomagnetic excursions KW - paleosecular variations KW - Black Sea KW - Paläomagnetik KW - geomagnetische Exkursionen KW - Paläosekularvariation KW - Schwarzes Meer Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:517-opus4-429461 ER -