GP33A-01 INVITED
Empirical Constraints on Geodynamo Behaviour During Superchrons
The excessive length of superchrons relative to other observed periods of stable polarity makes them highly visible features in the palaeomagnetic record. Here we ask: how anomalous is the geomagnetic field's behaviour in other respects during these periods? Studies of palaeosecular variation (PSV) made using the magnetization recorded in lavas or sediments provide an insight into the extent of "normal" geomagnetic variation of internal origin. The difference in PSV recorded in rocks formed during the most recent superchron (84-125 Myr ago) and rocks from the last 5 Myr has recently been shown to be smaller than previously reported but still significant. Specifically, the angular dispersion of virtual geomagnetic poles (VGPs) recorded in rocks formed at low palaeolatitudes appears to be slightly lower in the Cretaceous superchron than at other times and the magnitude of this dispersion also appears to be more constant through time. These differences are qualitatively similar to those displayed by the Glatzmaier-Roberts dynamo model when made to exhibit the relevant degrees of stability by changes in the heat flux across the core-mantle boundary. The results of new and published PSV studies made using rocks formed during the Permo-Carboniferous Reversed Superchron (262-318 Ma) will also be discussed. Producing reliable measurements of the geomagnetic palaeointensity from igneous rocks is extremely challenging and consequently, significant uncertainty remains over the average geomagnetic intensity during superchrons. That said, there is a growing body of evidence that the field, as observed at the Earth's surface, may have been stronger during the two most recent superchrons than at most other times. This is also consistent with the behaviour of certain numerical dynamo models. However, it should be noted that the potential increase in geomagnetic intensity which occurred during the Cretaceous superchron at least, does not appear larger than the total range of variation observed within the rest of the record away from superchrons. That is, average geomagnetic intensity varied at least as much during times outside of superchrons as it did at times of superchron initiation and termination. This, when taken together with other empirical evidence, does not strongly support the idea that superchrons represent periods when the geodynamo was in a fundamentally different state to that which it was in at all other times during the last few hundreds of Myr. Should superchrons simply be seen as one end-member on a continuous scale representing the geodynamo's tendency to produce polarity reversals? Regardless of the answer to this question, the specific cause(s) of the indisputable time dependence of this tendency for reversal remains as elusive and intriguing as ever.
GP33A-02 INVITED
Cryptochrons and Short Subchrons in the Marine Sedimentary Record
Cryptochrons are short geomagnetic events (less than about 30 k.y. in duration) that are caused by large changes in the direction and/or intensity of the geomagnetic field. As such, they may equate to paleointensity fluctuations, geomagnetic excursions, or full polarity reversals that have durations less than that of a subchron. Their existence was first recognized by Cande and Kent (1992) by the "tiny wiggles" that they caused in marine magnetic anomaly profiles. Since then, our ability to document and study these and other forms of short-term geomagnetic field variability has been greatly enhanced by the recovery of long, continuous sedimentary sections from the World's oceans. The Ocean Drilling Program (ODP) and its successor, the Integrated Ocean Drilling Program (IODP), have played a crucial role in this effort, not only in acquiring the necessary cores but also in supporting the scientific community as they have developed new sampling strategies, stratigraphic analysis tools, and instrumentation. In particular, coring multiple holes at a site has allowed the construction of composite sections that yield stratigraphically complete records. Continuous, high-resolution measurements, made possible by automated-track systems, has provided the necessary quantity and quality of data to resolve the age, duration, and geomagnetic characteristics of several of the cryptochrons that had previously only be observed in marine magnetic anomaly profiles. We will discuss the geomagnetic changes associated with eight cryptochrons that occur in the Miocene and Plio- Pleistocene. We will also examine intervals in the Paleocene, Eocene, and Oligocene where cryptochrons were expected based on marine magnetic anomaly profiles but where marine sedimentary sections fail to record any notable changes in direction. This may indicate that the events are too short-lived to be recorded in the lower sedimentation rate records from these older time periods or that the cryptochrons are due only to fluctuations in the intensity of the field and not directional changes.
GP33A-03 INVITED
Tracking the long-term evolution of geomagnetic secular variation from tiny wiggles in marine magnetic profiles
Knowledge of the ancient geomagnetic secular variation is principally restricted to the past few million years. As a consequence, it has been very difficult to assess a connection between the secular variation regime and the magnetic reversal frequency as initially suggested by McFadden et al. (1991). Marine magnetic measurements exhibit coherent short wavelength anomalies (or "tiny wiggles") that are superimposed on the broader polarity-interval anomalies. Recent high-resolution magnetic measurements acquired near the seafloor demonstrated that most tiny wiggles do reflect paleointensity fluctuations recorded by the oceanic crust (e.g. Gee et al., 2000; Pouliquen et al., 2001, Bowles et al., 2003). Although these studies were limited to a few areas and to short periods of time, they show that the sequence of tiny wiggles offers a unique way for constraining the long-term evolution of secular variation. Hence, the purpose of our study was to perform an exhaustive investigation of tiny wiggles over a long period of time in order to study their temporal distribution as a proxy for secular variation. To this end, we performed a careful inspection of sea-surface marine magnetic profiles selected from worldwide databases within the Indian and Pacific Oceans for the period 83-41 Ma. Many tiny wiggles were isolated by comparing stacks of profiles computed within widespread study areas. Modelling of those anomalies confirms that most tiny wiggles are likely ascribed to past fluctuations of the paleointensity rather than undetected short polarity events. We observe that tiny wiggles are ubiquitous and uniformly distributed throughout this long period of time, which may indicate a nearly constant secular variation regime while the magnetic reversal frequency markedly varied from zero during the Cretaceous Normal Superchron (~118-83 Ma) to about 2-3 reversals per Myr at ~40 Ma. These results motivate testing if the pattern of variation continues throughout the Cretaceous Normal Superchron which is characterized by the absence of magnetic reversal during ~35 Myr. To answer these questions, we are acquiring new high-resolution magnetic data across the Cretaceous Quiet Zone on the eastern flank of the Mid-Atlantic ridge (cruises Magofond3 held during Summer 2005 and Magofond3bis planned for Fall 2008).
GP33A-04 INVITED
Simulated geomagnetic reversals: How realistic are they and what can we learn?
Self-consistent numerical dynamo simulations have been used for more than a decade now to simulate geomagnetic field reversals. The computer models, in theory, allow to explore the underlying dynamics which is largely inaccessible to paleomagnetic exploration. However, the complex spatial and temporal nature of the dynamo mechanism complicates the numerical computation as well as the interpretation. Therefore typically rather simple numerical models have been analyzed that are quite remote from realistic parameters. It is thus essential to test their behavior against paleomagnetic inferences. On the other hand, the models may nevertheless provide valuable help in interpreting and understanding paleomagnetic records. We have analyzed several reversing models to infer some common characteristics. Numerical reversals are a three stage process: First, plume-like outflows in the equatorial region or at higher latitudes produce significant inverse field and step-wise degrade the predominant dipole. The second stage is a period where the field is dominated by higher harmonics and changes rather fast. Finally, the dipole grows again with opposite polarity. The slow degradation is also observed in paleomagnetism, but the second low-dipole period seems not a common feature. Also, in paleomagnetic data the dipole recovers typically faster after the reversal than it had decrease before while both processes have a comparable duration in the simulations. Though the overall reversal process tends to last longer in the computer models than in the geodynamo, the latitudinal dependence of durations agrees with paleomagnetic findings. The simulations also suggest some additional features that can be tested with paleomagnetic data: Reversals as well as excursion are rarely simple dipole swings but typically involve several polarity changes. Also, excursions tend to be more pronounced at low and high latitudes where the upwellings produce more inverse field, but excursions may be hard to discern from the background variation at mid latitudes. The simulations thus predict that fewer excursions are reported from mid-latitude regions. We have also analyzed the statistics of chron durations in our simulations and find that a log-normal or power-law probability distribution function is most likely. This is somewhat surprising, since these distributions require a long-term memory we have no explanation for. However, a log-normal statistics has also been suggested for the paleomagnetic record.