GP41B-01 INVITED 08:00h
Fast changes in the geomagnetic dipole field intensity based on radionuclide records from ice cores
The production rates of cosmogenic radionuclides strongly depend on the geomagnetic shielding of the cosmic rays. Measuring cosmogenic radionuclides in natural archives, such as ice cores, allows the reconstruction of past changes in geomagnetic field intensity. Measurements of radionuclides in ice cores have the advantage to provide well-dated high-resolution records. Therefore, it is possible to study in detail short-term changes. Previous studies showed that 10Be and 36Cl records from Summit in central Greenland reliably record past changes in geomagnetic dipole field intensity. Around 40,000 yr BP these records show an increase in radionuclide deposition of approximately 100 % which corresponds to a geomagnetic minimum which is in the order of 10 % of the present field. This estimate refers to changes on time scales longer than 3000 years. During other periods there is similar variability on shorter time scales. For example, the so-called Mono Lake excursion is visible on time scales of 2000 years or shorter. Here, we will discuss 10Be and 36Cl records from the GRIP and GISP2 ice cores from Greenland. We will analyse the time evolution of the geomagnetic dipole field changes during the last ice age. In particular, the radionuclide changes and their time scales around the Laschamp and Mono Lake excursion will be discussed.
GP41B-02 08:15h
A 300 kyr Record of Geomagnetic Excursions and Paleointensity From the Irminger Basin: Candidates for Mono Lake, Laschamp, Iceland Basin, Jamaica and Pringle Falls?
Sediments recovered at ODP Site 919, off east Greenland, record geomagnetic directional excursions at 33 ka and 40 ka (Mono Lake and Laschamp), and at 187 ka (Iceland Basin), 208 ka (Jamaica?) and at 220 ka (Pringle Falls). U-channel records are augmented by 1-cm discrete samples collected back-to-back alongside the u-channel troughs. Deconvolution of the u-channel records yields records that can be closely matched to the discrete sample data. The age-model based on planktic oxygen isotope data (St. John et al., Marine Geology, in press) is consistent with the relative paleointensity record and the recognition of Ash Layer 2 (55 ka). The results indicate that the Mono Lake and Laschamp excursions, and the Iceland Basin and Pringle Falls (and perhaps also Jamaica), are distinct excursions, rarely recorded together in individual stratigraphic sections. Why are they recorded at ODP Site 919? Mean sedimentation rates are 22 cm/kyr in MIS 3 where Mono Lake/Laschamp are recorded, but sedimentation rates do not appear to be especially high in MIS 7 (13 cm/kyr) where candidates for Iceland Basin/Jamaica/Pringle Falls are recorded.
GP41B-03 08:30h
Field Geometry During the Iceland Basin Event Observed from the North Atlantic Ocean, North Pacific Ocean and the South China Sea
The Iceland Basin event (IBE), which is named for a record from ODP Site 983 in the North Atlantic Ocean (60.5°N), is a focus of growing interest in our community. This geomagnetic excursion is coeval with the marine oxygen isotope (MIS) stage 7/6 boundary at about 190 ka according to the orbitally tuned SPECMAP time scale. We have detailed new records of this excursion at two new sites from the North Atlantic (core MD99-2247; lat. 59°N, long. 31°W and core MD99-2242; lat. 59°N, long. 47°W), at one site from the North Pacific (ODP Site 884; lat. 51.5°N, long. 168.3°E), and in two others from the South China Sea (ODP Site 1146; lat. 19.5°N, long. 116.3°E and ODP Site 1145; lat. 19.6°N; long. 117.6°E). For all five sites, the event is identified at the MIS 7/6 boundary, which confirms its potential as a precise stratigraphic marker. The average sedimentation rate for this portion of the cores is 7.5, 10 and 15 cm/kyr in North Atlantic, North Pacific and South China sea cores, respectively, which allows high-resolution studies in this time interval. The event is characterized by a marked minimum in the relative paleointensity (sometimes with a double feature) and by large swings in inclination and declination. The VGP latitudes reach 78°S and 45°S for ODP sites 1146 and 1145, respectively, 37°S for ODP Site 884, and 74°S and 41°S for cores MD99-2247 and MD99-2242, respectively. The VGP paths for the four records from the North Atlantic and South China Sea cores are highly similar. For the most detailed of these records, the poles first pass over Africa, then they proceed to Antarctica and return northward over Australia. The paths are less well resolved for ODP Site 1145 and core MD99-2242, for which the poles cross the southern Indian Ocean rather than reaching Antarctica. The VGP paths for the two most detailed records (ODP Site 1146 and MD99-2247) are highly similar and are also similar to the path for ODP Site 983, which also has a high sedimentation rate and is precisely dated. In contrast, but in common with other published records of the IBE, the VGP path for ODP Site 884 loops in the opposite sense and passes from North to South America and then over the Pacific Ocean before returning to high northern latitudes. The striking similarities between multiple records for two observed sets of differing VGP paths for the IBE at locations spanning the northern hemisphere provides evidence for a non-dipolar, but relatively simple, geometry of the transitional field.
GP41B-04 INVITED 08:45h
The Anatomy of a Geomagnetic Excursion: The Iceland Basin Excursion (188 ka)
The origin of geomagnetic excursions has been debated ever since their discovery. Several testable mechanisms for excursions have been suggested: (i) if an excursion represents a significant departure of the dipole field from the rotation axis, it would be expected to have a global manifestation with similar VGP paths; (ii) if excursions occur when the non-dipole to dipole field ratio becomes large, a global manifestation would also be expected, with different VGP paths (and possibly different apparent polarity) at varying locations on the globe, and (iii) if excursions represent relatively localized perturbations in the Earth's outer core, excursions would have restricted rather than global geographic extent. Few geomagnetic excursions are represented by high-quality data from a broad geographic coverage of well-dated sites, which has prevented adequate testing of these hypotheses. The Iceland Basin excursion (IBE; 188 ka) is now an exception. It occurred at the boundary between marine oxygen isotope stages 7 and 6 and is therefore easily dated in marine sediments. It also coincides with a clear paleointensity minimum, so it is also easily identifiable in continuous relative paleointensity time series. We have compiled more than 20 records of the IBE from a broad distribution of northern hemisphere sites, including several of our own new records. The level of detail of these records is variable, but many are from relatively rapidly deposited sediments (>10 cm/kyr) that provide detailed representation of field behaviour during the excursion. The VGP paths are highly consistent and can be split into two groups. One group has VGP paths that loop counter-clockwise from high northern latitudes over Africa to high southern latitudes and that then return to high northern latitudes through Australia/east Asia. The other group has VGP paths that loop clockwise from high northern latitudes over the North Atlantic/eastern North America to South America and then to high southern latitudes and that return to high northern latitudes through the Pacific and over Kamchatka. The widespread distribution of sites that record the IBE, and the fact that many of the records have high southern latitude VGPs, indicate that the IBE is a global feature and that it does not represent a localized perturbation of fluid flow in the outer core. The fact that a single VGP path is not observed for all excursions indicates that the excursional field is not a simple dipolar field. However, the dominance of only two VGP pathways indicates that the excursional field had a relatively simple geometry. We therefore prefer the hypothesis that geomagnetic excursions occur globally when the non-dipole to dipole field ratio becomes large at times when the dipole field is weak.
GP41B-05 09:00h
Age and Structure of the Laschamp Geomagnetic Excursion
The age of the Laschamp geomagnetic excursion has been recently re-investigated using unspiked K/Ar and Ar/Ar techniques (Guillou et al., Session V01, this conference). The new age determination of 40.4 +/- 2.0 ka (2 sigma) is more precise than those previously reported in the literature and agrees precisely with that deduced from the GLOPIS-75 sedimentary paleointensity stack calibrated against the GISP2 ice core chronology. Two of the North Atlantic cores used in GLOPIS-75 (MD95-2034 and PS2644-5) yield rather detailed transitional VGP paths. In the two cases the paths show large similarities, with the VGP initially descending along mid-western Pacific, then returning to normal polarities with a large clockwise loop over Africa and Europe. Differences in the highest southern latitudes reached by the VGP can be explained assuming more different degrees of smearing of the paleomagnetic record due to differences in sedimentation rate in the two cores. In the most detailed record, MD95-2034 , two smaller loops are present preceding the main excursion. In the two cores, the excursion is characterized by a significant drop in intensity. The reversal paths observed for the Laschamp event are very close in position to those reported for the Icelandic Basin Event (IBE) from sites in the North Atlantic and the South China Sea (Laj et al., this conference) but differ in the sense of looping: while a clockwise loop is observed here, a counterclockwise loop is observed for the IBE. Despite this difference, the similarity of the transitional records tends to suggest that a similar, relatively simple, geometry has dominated the two excursions and therefore that similar dynamo mechanisms have prevailed during the reversal process.
GP41B-06 INVITED 09:15h
VGP-Paths of Extremely High-Resolution Records of the Laschamp and Mono Lake Geomagnetic Excursions in Sub-glacially Deposited Cave Sediments, W. Norway
High-amplitude variations of paleomagnetic directions retained in clay/silt-sediments deposited inside three wave abraded coastal caves have been interpreted to represent incomplete records of the Skjong (Laschamp) and Valderhaug (Mono Lake) excursions1,2,3. The clay-silt sediments deposited in calm waters during glacial periods when the caves were filled with fresh-water (lakes) dammed by advancing ice. Ages of these biologically barren sediments are constrained by C14 and U-Th dates of CaCO3 precipitates and bones of a rich fauna of birds and mammals. Differential post-depositional compaction of these these laminated (annual?) resulted in N-NW dips of 20°-40° in the Skjonghelleren cave sediments. High coercivities of remanent magnetisation (MDF: 50-70 mT) are carried by partially maghemitized magnetite, probably strained during glacier-abrasive erosion. VGP paths of the Laschamp excursion from the three caves partly overlap and define three near-equatorial clusters (90°W, 20°E and 160°E). A secular variation signal apparently prevails during the Skjong excursion (Laschamp), and It is speculated that these loops reflect genuine high-resolution records of the complex behaviour of the geomagnetic field during the development of excursions on decadal time scales. The high-resolution paleomagnetic records have been retained due to the absence of major post-depositional modification during deposition-consolidation of these temporal, sub-terranean `lakes'. 1Lovlie R. and Sandnes A., 1987. Palaeomagnetic excursions recorded in mid-Weichselian cave sediments from Skjonghelleren, Valderoy, W. Norway. Physics of the Earth and Planetary Interiors, 48, 337-348. 2Valen V. Larsen E. & Mangerud J., 1995. High-resolution paleomagnetic correlation of Middle-Weichselian ice-dammed lake sediments in two coastal caves, western Norway, Boreas, 24, 141-153. 3Mangerud J., Lovlie R., Gulliksen S., Hufthammer A-K., Larsen E. & Valen V., 2003. Paleomagnetic correlations between Scandinavian Ice-Sheet fluctuations and Greenland Dansgaard-Oeschger events, 45,000-25,000 yr B.P., Quaternary Research, 59, 213-222.
GP41B-07 09:30h
Correlation of Late Pleistocene Terrestrial Climate Variation From Mono Lake, USA, With Global Records Using Relative Paleointensity
In order to assess different models of global climate variation, it is crucial to be able to accurately correlate terrestrial climate records with each other and with marine climate records. This problem is especially challenging in intervals older than 30 kyr, when problems with accuracy and precision of 14C ages become significant. Recently published stacks of global, high-resolution variation in intensity of Earth's past magnetic field (North and South Atlantic PaleoIntensity Stacks, NAPIS and SAPIS) enable correlation of high-quality terrestrial records of paleointensity with the GISP2 timescale. The lacustrine sediments of the Wilson Creek Formation (Mono Basin, CA) are known to be excellent recorders of Pleistocene climate and geomagnetic field variation, and are the type locality for the Mono Lake paleomagnetic excursion (MLE). Here we present rock magnetic analyses showing that the sediments also fit the criteria required for good recorders of paleomagnetic intensity, with a magnetic fraction dominated by fine-grained magnetite with concentration variation <3. Both the type section and South Shore cliffs were sampled continuously at 2 cm resolution, and susceptibility and Natural, Anhysteretic, and Isothermal Remnant Magnetizations (NRM, ARM, and IRM) were measured on all samples. IRM was chosen to normalize the NRM for paleointensity, though NRM/ARM produces a similar curve. The resulting records are similar both to each other and to the NAPIS and SAPIS curves, allowing correlation of the Wilson Creek sediments to the GISP2 timescale. We have used two independent age constraints to frame our correlation to NAPIS and SAPIS; first, carbonate 14C and tephra 40Ar/39Ar ages agree to 32 ka, which is thus used as an upper tie point. Second, the lakes of the Great Basin have been shown to be strongly controlled by the 100 ka cycle, and so we infer lake transgression over the Wilson Creek site at the M.I.S. 5/4 boundary, fixing the maximum age of sediment deposition. We then use distinctive features of the Mono and global paleo-intensity records to refine the correlation, resulting in an age of 68 ka for the oldest sediments, and 40 ka for the MLE, coeval with the Laschamp excursion. Finally, we compare climatic proxies measured at Mono to regional and global records, with major changes in glacial proxies at M.I.S. 4/3 boundary and evidence for low lake levels at the time of Heinrich events.
GP41B-08 09:45h
A paleomagnetic record of the last 640 kyr from an eastern Mediterranean piston core and a review of geomagnetic excursions in the Brunhes
Core KC01 (25.93 m long) was taken from a small ridge on the lower slope of the southern Calabrian Ridge (Pisano Plateau, 36deg15.25'N, 17deg44.34'E, 3643m water depth) during cruise MD69 of the French R/V Marion Dufresne in June-July 1991. The sediment consists of hemipelagic sediments with intercalated sapropel and tephra layers. They form an alternation of grey, greenish, olive-coloured, yellowish, white and beige shades. Sapropel layers are black to dark green. Paleomagnetic samples (6.4 cm3 cubes) were taken from the half split cores and measurements of the natural remanent magnetization was conducted at Utrecht University with a DC SQUID magnetometer (2G Enterprises model 740-R). Stepwise alternating field demagnetization was done at 8-11 steps up to 60-80 mT on each sample. Core KC01 (37.04 m long) was taken as a companion core from core KC01B at the same locality, earlier subjected to magnetostratigraphic work (Langereis et al., 1997). Langereis et al. (1997) established an age model based on the matching of (ghost-)sapropels with insolation minima. They applied the 65degN summer insolation calculated from the astronomical solution La90 (Laskar 1990; Laskar et al., 1993) as target curve and included a time lag of 3-kyr - based on the age difference between the radiocarbon dated midpoint of S1 at 8.5 ka and the insolation maximum at 11.5 ka following the method by Lourens et al. (1996). Lourens (2004) modified the astronomical chronology established by Langereis et al. (1997) and investigated the sapropel chronology on KC01B and KC01 based on high resolution colour correlation with ODP Site 964, which is 1 km away, and constructed an improved age model, which gives a better estimate for these two cores. Langereis et al. (1997) reported four excursions (CR0, CR1, CR2, CR3) with ages of 261, 318, 515, and 573 ka based on the chronology of sapropels on Core KC01B. Lourens (2004) revised the chronology of Core KC01B and redated the excursions as 260, 319, 543, and 593 ka. In this study, these four excursion records were confirmed and dated as 258-263, 313-319, 541-542, 592-594 ka based on the new chronology. Also Laschamp, Norwegian-Greenland Sea, Iceland Basin and Jamaica excursions were found at ages of 41-43, 77, 193-194 and 212-213 ka. Finally, we conducted a review of published excursion records and identified a maximum of 23 excursions and a minimum of 16 excursions in the Brunhes.