GP14A-01 INVITED
A Tour of Paleomagnetic Results From Ocean Drilling
After over a decade of working on the paleomagnetism of ocean drill cores, I have had the opportunity to collaborate with many other magnetists as we have tried to extract the paleomagnetic and rock magnetic information stored in the sediments and rocks cored from the ocean basins. In many instances, the cores have acted as high-fidelity magnetic recorders, providing detailed chronstratigraphies and exceptionally detailed records of geomagnetic field behavior, including the occurrence of multiple Brunhes-age excursions and complex changes in the direction and intensity of the field across geomagnetic reversals. They have also retained records of the long-term paleomagnetic direction at numerous sites around the globe, which have been used to constrain plate and hotspot motions. And, although one needs only to tow a magnetometer across the seafloor to map the marine magnetic lineations and date the ocean floor, it is the drill cores that allow us to examine the origin of the marine magnetic anomaly signal and to learn more about magmatic, tectonic, and alteration processes at mid-ocean ridges. In other instances, the paleomagnetic record has been less robust than expected owing to a myriad of natural processes, which provide clues about the environment, microbial activity, tectonic deformation, and fluid flow, or owing to drilling and handling artifacts, which are very important to recognize if the paleomagnetic record is to be interpreted accurately. I will take you through a tour of some of these results and suggest that the paleomagnetic journey has an important path to follow, with a focus on coring high and ultra-high resolution geomagnetic records. The records would become part of a constellation of paleomagnetic stations. Comparable to the modern magnetic observatories that provide data for modeling the present-day geomagnetic field, these stations will provide both the paleomagnetic directions and paleointensities necessary to construct spherical harmonic models of the geomagnetic field back in time.
GP14A-02 INVITED
High Sedimentation Rate Paleomagnetic Records for the Last 70 kyrs From the Chilean Margin (ODP Sites 1233, 1234, 1235)
Sediments that accumulate at around 1-m/kyr or greater preserve a paleomagnetic record that, under favorable conditions, may record the original geomagnetic input with little smoothing. However, such great rates of accumulation come with a price as features of interest are often deeply buried and may only be adequately recovered using drilling technologies. Here we present a full-vector geomagnetic reconstruction for the last 70,000 yrs from ultrahigh resolution records obtained through ODP drilling (Leg 202) on the Chilean Margin. ODP Site 1233 (41.0 S, 74.26 W, water depth 838 m) provides a 135-mcd u-channel derived directional paleomagnetic secular variation (PSV) and relative paleointensity (RPI) records. The chronology is constrained by AMS radiocarbon dates and tuning of alkenone sea surface temperature to Antarctic ice core temperature records back to 70,000 years BP. To the north, ODP Sites 1234 (36.13 S, 73.40W, water depth 1015 m) and 1235 (36.9 S, 73.33 W, water depth 489 m) provide independently dated shipboard and developing u-channel paleomagnetic records that reproduce many of the geomagnetic features observed at Site 1233 including excursions and high amplitude PSV intervals, while providing additional radiocarbon and isotopic constraints for development of a regional master chronology. All three Sites have exceptionally high glacial sedimentation rates that average 2-m/kyr for 1233, 80-cm/kyr for 1234 and 1-m/kyr for 1235. Fortuitously, the Laschamp magnetic excursion at Site 1233 occurs during an interval where sedimentation rates exceed 3-m/kyr. The Site 1233 chronology indicates that the Laschamp event, centered at 41,000 yrs BP, has a duration in reverse polarity of only 600 yrs, with polarity transitions occurring in less than 200 yrs within a 1500 yr long interval of low RPI. The path of virtual geomagnetic pole (VGP) positions for the Laschamp excursion at Site 1233 is generally consistent with the hypothesis of a simple field geometry because the large clockwise VGP loop mimics the path seen for other sites recording the same excursion. Unlike prior observations, this path begins and ends at approximately the same Alaskan location after making a complete loop through the Pacific to high (> 80 degrees) southerly latitudes. Insights into the dynamics PSV and RPI, the resolution of magnetic stratigraphies, the geomagnetic controls on cosmic ray shielding and whether these might affect climate will also be touched upon.
GP14A-03 INVITED
Paleomagnetic and Mineral-Magnetic Results From the Lake Bosumtwi Drilling Project
Lake Bosumtwi is an impact crater lake located at 6.5ºN and 1.5ºW in Ghana. The Bosumtwi impact event and the Ivory Coast tektite strewn field generated by the event have an age of 1.07 Myr. The lake contains an approximately 300m sedimentary section that provides a nearly continuous stratigraphy of this interval. In addition, the majority of the section is annually laminated. Advantages of magnetic studies of Lake Bosumtwi sediments include: (1) duplicate drill holes; (2) high sedimentation rates; (3) laminations allow assessmet of core disturbance; and (4) sediments record polarity transitions, excusions, and possibly paleointensity at a low-latitude site; and (5) Milankovitch and sub- Milankovitch scale paleoclimate variations. Challenges include: (1) low-latitude site location makes it difficult to recognize excursions and polarity transitions; (2) anoxia that produces annual laminations also causes some reductive diagenesis; and (3) some intervals are weakly magnetic and have noisy data. Overall, with respect to its magnetic record, Lake Bosumtwi is neither a "silk purse," nor a "sows ear." We will detail our efforts to maximize the former and minimize the latter.
GP14A-04 INVITED
The Relative Timing of Directional and Intensity Changes During Geomagnetic Polarity Transitions: Constraints on Transitional Field Geometries
Improvements in our ability to obtain relative paleointensity records from deep-sea sediments are moving us closer to the goal of obtaining widely distributed, full-vector records of polarity reversals. Coupled with high- resolution oxygen isotope data, recent transition records provide insights into the chronology of the field changes that occur as the field reverses polarity. These advances warrant examining if there exist first-order relationships between the timing of the directional and the intensity changes that might provide constraints on reversal models. A recent compilation of transition records indicated that the duration of the directional change is dependent upon the site latitude and that this is to be expected from simple geometric models of reversals. The same geometric models, however, do not indicate a site-dependence of the timing or duration of the intensity decrease that occurs during a reversal. I present here results of additional geometric reversal models and demonstrate that not only is the duration of the directional change dependent upon site latitude, but that the directional change occurs at significantly different times within the intensity low. Depending upon the site latitudes, the difference may exceed 4 kyr. The time-transgressive nature of the directional changes differ significantly for models in which a reversal is initiated at low-latitudes from those in which the reversal is initiated at high-latitudes. This means that the age of the directional change may be expected to vary at different locations in a predictable and testable manner. Spatial variation in the ages of the directional changes therefore may provide constraints on the reversal geometry.
GP14A-05
Paleomagnetism of Basaltic Rocks Cored from the Walvis Ridge, South Atlantic and Implications for Hotspot Paleolatitude
Walvis Ridge, a volcanic ridge in the southeastern Atlantic Ocean basin, is considered one of the Earth's few primary hotspot tracks and thus it can provide important information about the formation and evolution of hotspots. Deep Sea Drilling Project (DSDP) Leg 74 cored ~226 m of basaltic rocks from three sites (525, 527 and 528) in close proximity to one another on a transect about mid- way along the chain. These rocks gave a radiometric age of ~79 Ma and paleomagnetic data of this age from the Atlantic-bordering continents imply that the paleolatitude of this site may have been ~8° higher than the present-day hotspot location. To examine this possible shift, we made paleomagnetic measurements on 158 samples from 18 flow units cored from the three sites. Thermal demagnetization was used on most samples to isolate the characteristic remanence direction and a number of magnetic tests were performed to understand the grains that carry the magnetization. Although magnetic properties imply that the magnetic carrier is fine-grained titanomagnetite (normally a reliable recorder of the magnetic field), interpretation is complicated because the rocks show evidence of self-reversal (i.e., acquiring some part of the magnetization in a direction opposite to the ambient field). This likely explains why Site 525 shows 6 zones of alternating polarity that do not apparently respect flow unit boundaries. Magnetic measurements produced 18 independent paleocolatitude groups, with 6 from Site 525, 3 from Site 527, and 9 from Site 528. Combined, the data from all three sites imply a paleolatitude of 36.2° S +7.8°/-7.6° (95% confidence), which is statistically indistinguishable from the 37°-41° S latitude of the current hotspot. However, this result may be biased by high paleolatitude values from several units cored at Site 528, where flow unit paleocolatitudes show an unusually high degree of between-unit scatter that suggests these results could be affected by a factor other than secular variation. Excluding the Site 528 data, a paleolatitude determined from the 9 remaining units is 27.3° S +7.7°/+6.7°. Although this result implies that this part of the ridge acquired its magnetization significantly north of the present-day hotspot, the small number of magnetic units and the resulting potential for a misleading result owing to sampling bias indicates that this result should be interpreted with caution. Overall, the paleomagnetic results from Leg 74 sites on Walvis Ridge are intriguing, but show some of the problems and uncertainties of trying to determine a reliable paleolatitude using a limited number of lava flows from poorly-understood volcanic terrain.
GP14A-06
Paleolatitude Records of the Western Pacific as Determined From DSDP/ODP Basaltic Cores
We report here the new paleomagnetic, rock magnetic, and Ar-Ar geochronologic results of our recent completed project, which aims to determine the Cretaceous paleomagnetic paleolatitude record and the architecture of the volcanic basins in the western Pacific Ocean. The new results, in concert with our paleomagnetic research on ODP rocks recovered from the Ontong Java Plateau (OJP), suggest that various plateaus and basins in the western Pacific had similar plate-tectonic setting (paleolatitude) and ages with that of OJP at time of emplacement (~120 Ma). Basalts sampled from Deep Sea Drilling Project (DSDP) and Ocean Drilling Program (ODP) sites of the greater OJP as well as from obducted sections in the Solomon Islands of Malaita and Santa Isabel are strikingly uniform in petrologic and geochemical characteristics. Many of these cores, especially those from DSDP sites, have not been well-studied paleomagnetically and hence underutilized for tectonic study. We carefully re-sampled and systematic demagnetized and analyzed 925 basaltic cores from 15 sites drilled by10 DSDP/ODP Legs in the western and central Pacific, which represents a unique possibility for averaging out secular variation to obtain a well-defined paleolatitude estimate. The most important findings from this study include: (1). most basins formed during the Cretaceous long normal magnetic period with similar Ar-Ar ages as the OJP; (2) East Mariana, Pigafetta, the upper flow unit in the Nauru basin and Mid-Pacific Guyots all yielded similar paleolatitudes as those for OJP, suggesting the volcanic eruptions of flows in these basins are likely related to the emplacement of the OJP; and (3) the lower flow unit in the Nauru basin yields a paleolatitude that is ~10° further south and the age is more than 10 m.y. older than these of the OJP.
GP14A-07
Paleomagnetic and Geochemical Properties of Archeological Carbonates: Potential as Recorders of Archeomagnetic Variations and Environmental Changes
Carbonate deposits along roman water canalizations have been reported in many sites of the Roman Empire. Such deposits are characterized by laminated sequences sometimes directly related to an annual cycle. They might provide high resolution records of paleoenvironmental proxies and magnetic data continuously archived by the flowing water. Three oriented large cores were drilled along the 'Pont du Gard' adduction system in Southern France. The paleomagnetic signal and isotope records archived in these deposits is investigated. The magnetic mineralogy analyses show that magnetite at very low concentration is the main magnetic carrier. The magnetic signal is very weak but archeomagnetic field variations were successfully isolated. Comparisons with know archeomagnetic curves for France (Buccur, 1994) show that our data fits well with a period going from the 1st until the 4th century AD allowing to precisely date the cores. Oxygen and carbon isotope records made on the same deposits show perturbations of the Karst system at several periods. The potential of such deposit for magnetic and environmental reconstructions will be discussed.
GP14A-08
Rock Magnetic Characterization Through an Intact Sequence of Oceanic Crust, IODP Hole 1256D
One goal of drilling a complete oceanic crust section is to determine the source of marine magnetic anomalies. For crust generated by fast seafloor spreading, is the signal dominated by the upper extrusive layer, do the sheeted dikes play a role, what role do the gabbros play relative to slow spreading centers, and what is the timing of acquisition of the magnetization? To address these questions, we are conducting a comprehensive set of rock magnetic and paleomagnetic measurements that extend through the intervals drilled on Leg 206 and Expeditions 309 and 312. Recent drilling in the Eastern Pacific Ocean in Hole 1256D reached gabbro within seismic layer 2, 1157 meters into crust formed at a superfast spreading rate (i.e. up to 200mm/year full rate) on the Cocos-Pacific plate boundary between 19 and 12 million years ago. Sampling an intact sequence of oceanic crust through lavas, dikes, and gabbros is necessary to advance the understanding of the formation and evolution of crust formed at mid-ocean ridges, but it has been an elusive goal of scientific ocean drilling for decades. Continuous downhole variations in magnetic grain size, coercivity, mass-normalized susceptibility, Curie temperatures, and composition have been mapped. Compositionally, we have found that the iron oxides vary from being titanium-rich (TM60) to titanium-poor magnetite as determined semi-quantitatively from Curie temperature analyses. Magnetic grain sizes vary from few Single Domain (SD), to the majority of them being Pseudo Single Domain (PSD) and some on the Multi Domain (MD) area of the Day diagram. The low-Ti magnetite or stoichiometric magnetite is present mainly in the lowest part of the section and is associated with higher Curie temperatures (550°C to near 580°C) and higher coercivities than in the extrusive basalts. Skeletal titanomagnetites with varying degrees of alteration is the most common magnetic mineral throughout the section and is often bordered by large iron sulfide grains. Last but not least, absolute paleointensity experiments have been determined on several samples, although the success rate is low as has been found in other studies of oceanic basalts.
GP14A-09
Magnetic Mineralogy of a Complete Oceanic Crustal Section (IODP hole 1256D)
Oceanic crust is the carrier of the marine magnetic anomalies and is therefore a valuable archive of geomagnetic information. Hole 1256D, which was drilled during ODP leg 206 and IODP expeditions 309 and 312, was the first to retrieve a core comprising of an entire sequence of oceanic crust down to the gabbro. This provides a unique opportunity to study the carriers of the marine magnetic anomalies. We used reflected light microscopy, scanning electron microscopy and energy dispersive x-ray analysis in addition to rock magnetic measurements to study the grain size, morphology, composition, and alteration state of the magnetic minerals. This comprehensive data set allows us not only to understand the magnetic stability of the minerals but also the mode and timing of remanence acquisition. The extrusive layer contains dendritic, low-temperature (LT) oxidized titanomagnetites (TMs) typical for mid- ocean ridge basalts (MORBs). The initial composition of these is close to previously reported values for MORB TMs with an ulvöspinel content of about 60%. The degree of LT oxidation remains fairly constant across the whole extrusive part of the section with an oxidation parameter z=0.6. Therefore, the increase in Curie temperature from 200C at the top to about 500C at the bottom of the extrusives cannot be accounted for by LT oxidation alone. Instead, we favor a model involving submicron inversion of LT oxidized TMs to an intergrowth of TMs and nonmagnetic phases, where the Ti-content of the TM phase is continuously decreasing with depth due to higher inversion temperatures. In the underlying sheeted dykes and gabbros, TMs precipitated as the primary opaque phase, as well. Due to the slower cooling rate, these particles are in most cases oxy-exsolved and form lamellar intergrowths of Ti- poor TMs and ilmenite. After emplacement, these minerals were altered to a much higher degree than the extrusive lavas. Secondary minerals frequently replace the original TMs, and the lower part of the sheeted dykes witnesses the onset of hydrothermal alteration. In the gabbroic part of the section, grain sizes of TMs reach values of up to several mm. These findings lead to the conclusion that the different parts of the section had acquired their remanent magnetisation by different mechanisms: The extrusive part carries a TRM, the intensity of which was later influenced by LT oxidation and inversion. The sheeted dyke part is likely to carry a CRM acquired during hydrothermal alteration, and the underlying gabbro acquired a TCRM at a time significantly after emplacement due to slow cooling at this depth.
GP14A-10
Magnetic Properties of Fast Spreading Ocean Crust Drilled by (I)ODP Hole 1256D Into the Gabbro Zone Revealed by Fuzzy c-Means Cluster Analysis
Despite its importance for plate tectonics the nature of ocean crust is relatively poorly constrained primarily related to the extreme difficulty of direct sampling. Here we report on a magnetic mineral analysis of the entire oceanic crust down to in the gabbro zone at Site 1256 (~15Ma) that was drilled during three (I)ODP Legs (206, 309 and 312) in a super-fast spreading zone in the Pacific Ocean on the Cocos Plate. Mineral magnetically, sheet flows and massive basalts are characterized by low Mrs/Ms ratios of ~0.1 with sheet flows tending to be slightly higher. Sheeted dikes are typified by Mrs/Ms ratios between 0.1 and 0.2 and the underlying gabbros have < 0.1. Sheeted dikes have highest Bc values of ~15 mT, gabbros are 5-10 mT, very similar to sheet flows and massive basalts. Bcr/Bc ratios increase with depth in the sheet flows and massive basalts from ~1.5 to ~2.0. Sheeted dykes are remarkable constant at ~2.0 while gabbros range 2.0-4.0 for the coercivity ratio. Curie temperatures appeared to be very distinctive, gradually increasing with depth from 200-300°C in the top of the sheet flows and massive basalts to 500°C at the base of the extrusive rock package. Sheeted dikes and gabbros have Curie temperatures of 580°C. Fuzzy c-means cluster analysis (333 samples with 5 variables: Mrs/Ms, Bcr/Bc, Bc, low-field susceptibility, and Curie temperature) enables a distinction between gabbros, sheeted dikes and upper crustal rocks on the basis of their combined magnetic properties that is very consistent with lithological observation. Individual magnetic parameters were less discriminative. On the non-linear mapping (NLM) plot sheeted dikes and gabbros on the one hand and sheet flows and massive basalts on the other, follow different trends. Quenched glass samples appear to have a very distinct position on this plot. Most clusters show signs of thermochemical overprinting leading to impure TRMs so they are less suited for paleointensity analysis. The most pristine samples have fairly low Mrs/Ms ratios of ~0.1-0.15 and Bcr/Bc ratios of ~1.6-2.0.