GP21A-0768
High paleosecular variation during the Cretaceous Normal Superchron
The Cretaceous Normal Superchron (roughly 84-120 Ma) marks one of the anomalously long periods in the Earth's history when the magnetic field did not reverse its polarity. Presumably changes in the convective behavior of the geodynamo trigged such a superchron. Paleosecular variation, which quantifies the convecting vigor driving the geodynamo, can be used to track these changes. To help constrain paleosecular variation during the Cretaceous Normal Superchron, we sampled an extensive section of marine sediments in northern Peru, which spans in time nearly the entire superchron. Paleosecular variation in our section, expressed by the S-value, is significantly higher than that of the last 5 Myr, and far higher than the S-value calculated for the Cretaceous Normal Superchron by McFadden et al. (1991). Moreover, because Peru was at the geomagnetic equator in the Cretaceous, the S-value derived from our sampling site represents the baseline S-value for all latitudes. Our study thus suggests a much higher paleosecular variation during the Cretaceous Normal Superchron than previously thought. References: McFadden, P.L., Merrill, R.T., McElhinny, M.W., Lee, S.: Reversals of the Earth's Magnetic Field and Temporal Variations of the Dynamo Families. Journal of Geophysical Research, 96/B3, 3923-3933, 1991.
GP21A-0769
Chron 24r: Cryptochrons or Subchrons?
A comparison is made between the "tiny wiggles" seen in marine magnetic profiles over Chron 24r and their proposed counterparts in a corresponding stratigraphic section in the foothills of the Canadian Rockies of Alberta. The age of the latter is constrained by palynology, radiometric dating, and the presence of the carbon isotope excursion that is the accepted marker for the Paleocene-Eocene boundary. Whereas the marine record suggests the presence of 10 (or 11) cryptochrons with durations on the order of 10 kyr, the land-based record implies much longer normal-polarity magnetozones on the order of 100 kyr. The polarity and apparent lengths suggested by the Canadian record can be satisfied by an orthodox 2-dimensional block model of the ocean floor consisting of appropriately wide (subchron-sized) blocks with strongly diminished average magnetic intensity. This may reflect a generally weaker geomagnetic field during these intervals, although the presence of blocks of mixed normal and reversed oceanic crust cannot be excluded.
GP21A-0770
Geomagnetic reversal frequency during the Earth's Middle Age
We obtained new magnetostratigraphic results from two Precambrian carbonate sections from Eastern Siberia and Southern Urals dated between ~1100 Ma and ~800 Ma. Sample magnetizations from the uppermost Mesoproterozoic Talakh-Khaya in the Siberian Uchur-Maya region appear to be mostly carried by a mixture of magnetite and hematite. A sequence of 33 magnetic polarity intervals is obtained within the section. All reversals occur in the first ~24 meters, while the upper ~160 meters are characterized by a single magnetic interval of normal polarity assuming a northern hemisphere position of Siberia around 1000 Ma. The Uralian Minyar section dated from the lower Neoproterozoic also possesses an ancient magnetization carried by magnetite and hematite. The high-temperature magnetization component obtained from the Minyar section allows one to establish a sequence of 43 magnetic polarity intervals. Positive reversal tests obtained from our two data sets and from other previously analyzed Proterozoic sections indicate that there is no convincing evidence yet for a long-standing asymmetric geomagnetic field during the Proterozoic, that would make the Precambrian field markedly less dipolar than during the Phanerozoic. The new data further show the occurrence of sharp transitions and alternations between long periods without any reversal (one superchron is observed from the Talakh-Khaya section) and periods with high reversal frequencies, possibly larger than 5 to 10 reversals par Myr. This characteristic may well be an important property of the Precambrian field, although rather sudden transitions between reversing and non-reversing states of the geodynamo may not only be restricted to this period.
GP21A-0771
Very Rapid Transitional Geomagnetic Change Recorded During Thermal Remagnetization of a Warm Lava Flow
Detailed thermal demagnetization of samples from three vertical profiles through a 3.9 m thick lava flow provide evidence of very rapid field change during a geomagnetic polarity transition at 15.6 Ma. The lava flow, referred to here as "Flow 20", occurs near the top of a stack of ~50 lavas in the Sheep Creek Range (north central Nevada) that records the transitional-to-normal part of an R-to-N geomagnetic polarity transition. The top meter of Flow 20 is magnetized in a NNW-down direction identical to that of Flow 21, the 7.9 m thick lava that overlies it. Lower in Flow 20, a higher blocking temperature (Tb) remanence component identical in direction to that of the underlying flow (Flow 19) becomes increasingly prominent. This higher Tb component points SSE-down, about 50 degrees away from the NNW-down component, and is clearly transitional in direction. The Tb at which SSE-down component is first discernible decreases steadily with depth in the flow, reaching a minimum of approximately Tb=350C at the flow base. Flow 20 apparently erupted while the field was pointing in the SSE-down direction and then was partially thermally remagnetized in the NNW-down direction. Experiments with a flexible, 1-dimensional conductive cooling model show that the base of Flow 20 is unlikely to have reached a temperature greater than approximately 200C if it was fully cooled when reheating by Flow 21 began. In order to explain how the base of Flow 20 reached 350C during the reheating, we speculate that that the interior of Flow 20 was still warm (near 300C) when capped by Flow 21. The implication of this interpretation is that the 50 degrees of field change recorded by the two remanence components must have occurred during the incomplete initial cooling and early reheating (by Flow 21) of Flow 20. The conductive cooling model shows that the timescale for the field change in this scenario is of the order of 0.5 year, implying that the transitional geomagnetic field change occurred at a rate of at least 10 degrees/month -- many orders of magnitude greater than typical of secular variation.
GP21A-0772
Using fracture zones magnetic anomalies to unravel the long-term behavior of the geomagnetic field
Marine magnetic profiles oriented along the spreading direction have been used to study paleointensity variations of the geomagnetic field. These data-sets have confirmed the link between crustal magnetization and past geomagnetic field variations. An additional set of globally distributed anomalies, oriented perpendicular to the spreading flow-line direction, are located above oceanic fracture zones. These anomalies have never been used to study the geomagnetic field nor has their origin been systematically studied. Here we present the first attempt to use these anomalies to study the long-term behavior of the geomagnetic field in the Cretaceous normal polarity superchron. We use archival data and new magnetic and bathymetry data acquired on geophysical cruise MGLN44MV aboard the R/V Melville on the Pioneer and the Murray fracture zones, North Pacific. These Fracture zones straddle the fast-spreading Pacific Cretaceous Quiet Zone (KQZ) and provide a relatively simple setting to test the feasibility of our approach. We use data from 150 crossings that cover a time-span of 27 million years (110 to 83 Ma). The anomalies demonstrate a remarkable uniform shape and size for thousands of kilometers implying that they were generated by spatially and temporally uniform processes. Two dimensional inversion solutions together with 3D forward models suggest that remanent magnetization governs the magnetic signal. These models also imply that enhanced magnetization, primarily situated within the uplifted regions adjacent to fracture zones, is responsible for the observed anomalies. The actual mechanism that creates these magnetization highs is not known, yet the inference of remanent magnetization allows us to use the anomalies to explore the behavior of the geomagnetic field. The nearly uniform amplitude of the anomalies suggests that the strength of the geomagnetic field remained relatively constant over most of the superchron. If a change in the geodynamo behavior is linked with changes in frequency of reversals then this change should have taken place after (and probably before) the superchron. Initial analysis of crossings located on crust younger then the KQZ suggests that in-fact, the Cenozoic field had similar strength as the superchron.
GP21A-0773
High Resolution Magnetostratigraphy from Plio-Pleistocene Sediments, Armenia
The Pleistocene geology of Armenia is dominated by the widespread occurrence of lacustrine diatomites. The onset of diatomite deposition in southern Armenia started at about 2 Myr. The importance of these sediments stems from the fact that they provide a high resolution archive of environmental conditions, especially during the emergence of early man in the region some 1.8 Myr ago (homo georgicus). Unfortunately, however, precise age models are still unavailable, except for the Shamb section where overlaying basalts yield an age of 819±18 kyr and two intercalated tuff layers yield ages of 1.1±0.01 and 1.2±0.01 Myr (Ollivier et al., in press). The study presented here, was originally designed to establish a robust magnetostratigraphy as a tool to further constrain the age of other diatomite deposits in southern Armenia. Six geographically distributed sections (Ashotavan, Darbas, Brnakot, Uyts, Shamb and Sisian) nearby the town of Sisian and ranging in thickness from 10 to 110 m have been sampled with minimum sampling spacing of 5 cm. The vast majority of the samples was thermally demagnetized and yield very stable and well defined characteristic directions of dual and intermediate polarity. These are interpreted to define the Olduvai and Jaramillo normal polarity subchrons. During a second field trip in 2008, sampling was expanded and special attention was given to those parts in the sections, where transitions where predicted. Except for the Darbas, where the magnetization is to weak to give robust directions, thermal demagnetization yields extremely well defined results. If estimates for the sedimentation rates in the order of 35 cmkyr-1 (Ollivier et al., in press) are correct, the duration of the polarity transitions observed here, is in the order of 5 kyr. Most surprisingly, however, the VGPs computed for the transitional directions do not define either a continuous path or show clusters but are randomly distributed in the Pacific and on the Americas. This observation is consistent with data published by Glen and Coe (1999) for the Searles Lake, California and underlines the high potential of these rocks for high resolution paleomagnetic studies.
GP21A-0774
PADM2M: A Time-varying Model of Paleomagnetic Axial Dipole Moment for 0- 2Ma
Earth's dynamic history can be investigated by looking at the mark left in the paleomagnetic record. Core processes drive the continuously changing geomagnetic field, which has been recorded on Earth's surface by newly forming igneous rocks and sediments. We construct a continuous time varying model of Paleomagnetic Axial Dipole Moment (PADM) variations over the past two million years using almost 1500 globally distributed absolute paleointensity data from the PINT08 and Geomagia50 databases and 90 distinct relative paleointensity records from sediment cores that are now archived in the MagIC PMAG database. Our modeling is carried out using maximum likelihood (ML) estimation to find a regularized cubic B-spline model with reasonable misfit to the data. As part of the modeling procedure we solve for scaling factors needed to transform the relative paleointensity records to virtual axial dipole moments (VADM). The resulting model is designated PADM2M and provides a new global estimate of the time varying PADM for 0-- 2~Ma. We compare our results to existing relative paleointensity stacks and find both similarities and differences. Similarities include the recovery of the expected paleointensity lows at field reversals and excursions, while differences are best illustrated by a statistical analysis of the properties of the model.
GP21A-0775
Geomagnetic Instability Time Scale 2008 (GITS-08) and dynamo processes
During the past 2.6 million years Earth's outer core geodynamo has produced at least 18 geomagnetic excursions and 5 full polarity reversals. This record has been compiled from terrestrial volcanic rocks, including mainly basaltic lava flow sequences, but also two silicic ash beds, that have been analyzed using modern paleomagnetic techniques and dated using the 40Ar/39Ar method. Several brief periods of field instability associated with excursions correlate with lows in paleointensity or directional changes recorded in marine sediments, for example in the SINT2000 or GLOPIS75 composite records, or the more detailed records found at ODP site 919, that are dated using astronomically-forced oxygen isotope signals or ice layer counting. However, the lack of correlation of several excursions between marine and terrestrial records indicates that neither sediments, nor lava flows, are ideal recording media. Another factor complicating correlation is that some excursions may be geographically localized and not expressed globally. Despite decades of observation, these records remain fragmentary, especially when periods of millions of years are considered. Recent 40Ar/39Ar dating in our laboratory, that includes age determinations for the Mono Lake, Laschamp, Blake, Pringle Falls, Big Lost, West Eifel, and Agua Nova excursions, as well as the Halawa (C2r.2r-1) cryptochron, prompt us to critically review the terrestrial record of geodynamo instability and propose a GITS for the entire Quaternary period. Both the ca. 4:1 ratio of excursions to reversals during the past 2.6 Ma as well as the temporal pattern of occurrence of these events provide fundamental input as to the long-term behavior and, possibly, the structure of the core dynamo. On the one hand, intervals of significant temporal clustering of excursions have highlighted a relatively stable period of high field strength lasting >250 ka in the middle of the Brunhes chron during which time few, or no, excursions took place. On the other hand, successive paleomagnetic excursion records may be critical in regard to understanding the behaviors and interactions between the mantle-influenced field source in the shallow core (the hypothesized SCOR-field) and the deeper-held source of the axial dipole. If in fact a successful reversal attempt requires the axial dipole field source to be weakened below some threshold strength for substantial duration (ca. 10- 20 ka), times of grouped excursions may also be the most probable times for a change in polarity.
GP21A-0776
Further Evidence for a new Brunhes Chron Excursion Recorded in Lava Flows at Agua Nova, Santo Antao, Cape Verde
New paleodirectional and initial paleointensity results from a re-sampled Brunhes age sequence of 34 basaltic lava flows at Agua Nova, Santo Antão, Cape Verde are linked with precise ages determined using the 40Ar/39Ar incremental heating method on groundmass separated from oriented hand samples originally studied by Knudsen et al. (Earth Planet. Sci. Lett., 206, 199-214, 2003). Alternating field demagnetization experiments on one sample per flow by Knudsen et al. (2003) suggested that two directional excursions may have been recorded. Five new 40Ar/39Ar isochron ages, which bracket and include the two excursional flows, are indistinguishable from one another and imply that a single complex period of excursional behavior took place at about 480 ± 13 ka. This excursion is the first to be revealed in a terrestrial volcanic sequence of this age and could correspond to cryptochron C1n-1 (504 to 493 ka) in the marine magnetic anomaly record (Cande and Kent, J. Geophys. Res., 100, 6093-6095, 1995). It is too young to be associated with the widely recorded Big Lost excursion, which is 40Ar/39Ar dated at 579 ± 6 ka, or any of the five West Eifel excursions dated between 722 and 528 ka (Singer et al., Phys. Earth Planet. Inter., in press, 2008). Neither is it clearly correlative with significant directional or paleointensity features observed in several high sedimentation rate ODP cores. Accurate paleodirections and paleointenities will enable us to define the full vector behavior of the geomagnetic field during this brief period of geomagnetic instability and will help us to better understand the processes within the Earth's core that give rise to such events.
GP21A-0777
Ar/Ar Dates From Tarrafal, Santo Antao, Cape Verde: Re-evaluation of two Brunhes age Excursions
We present new 40Ar/39Ar ages from a 500 m thick sequence of basaltic lava flows on the island of Santo Antão, Cape Verde, and assess their implications for two previously reported, fully reversed excursions in the Brunhes chron (Knudsen et al., Earth Planet. Sci. Lett., 206, 199-214, 2003). Paleodirectional experiments on a sequence of 52 lava flows at Tarrafal, Santo Antão, coupled with two argon dates, one from the bottom and one from near the top of the sequence, suggested that the majority of the volcanic stratigraphy was emplaced between 0.54 ± 0.09 Ma and 0.38 ± 0.05 Ma and that the two excursions could be related to geomagnetic instabilities within the Brunhes Chron that had not previously been observed in terrestrial volcanic records. New 40Ar/39Ar isochron ages, determined using the incremental heating method at the University of Wisconsin-Madison for nine groundmass separates from hand samples collected by Knudsen et al. (2003), range from 5.04 ± 0.10 Ma and 2.57 ± 0.02 Ma and indicate that that the stratigraphy of the Tarrafal section is more complex than originally thought. This more complicated stratigraphy can be explained by infilling of a flank collapse structure with younger Brunhes age lava flows and subsequent erosion of the sequence. The lower "excursion" reflects sampling across two normally magnetized Brunhes age flows at the base of the section and an older sequence of flows encompassing both reversed and normal polarity. The older reversed flows correspond to part of the Gilbert chron between 5.04 ± 0.10 Ma and 4.53 ± 0.09 Ma; the normal flows could relate to a normal polarity chron or subchron between 4.53 ± 0.09 Ma and 2.8 ± 0.02 Ma. The upper "excursion" marks the change from a volcanic event of reversed polarity (3.56 ± 0.29 Ma) to the normal polarity flows of the Brunhes chron (0.41 ± 0.02 Ma). Our findings further highlight the need for precise age determinations and a high sampling density of dated flows when using long lava sequences to reconstruct paleofield behaviour.