Paleomagnetism of Aleutian Volcanic Rocks:
Testing the Geocentric Axial Dipole (GAD) Hypothesis and Paleosecular Variation of the Geomagnetic Field.

Principal Investigator: David Stone

To learn about the behavior of the geomagnetic field with time, direct observatory records may be used for the last century of so, ancient writings often give more clues, as do the logs from mariners of old. To extend the record much further back in time requires other techniques. The most useful is paleomagnetism and the sub-discipline of archeomagnetism. Paleomagnetism relies on the ability of many rocks, pottery shards, kilns et cetera to record the magnetic field surrounding them at the time they were formed or heated. There are many ways in which the field can be recorded ranging from the physical alignment of very small magnetic particles in sediments as they are deposited to the cooling of magnetic material from above the temperature at which they loose their magnetism. This temperature is known as the magnetic blocking temperature. It has been demonstrated many times that the magnetization acquired on cooling can accurately record the ambient field.

To model the geomagnetic field on a global scale requires that reliable data sets are available from a global distribution of sites. At the present time the global distribution of reliable paleomagnetic sites is marginal, so a group of universities, including the University of Alaska, put together a joint project to make the best possible measurements of volcanic rocks younger than five million years. It was decided to concentrate sampling along a line of longitude that roughly follows the western edge of the Americas, going as far north and south as possible. This longitude was chosen because it forms part of the margin of the Pacific ocean basin, and there is some evidence that when the geomagnetic field changes polarity a main component of the field sweeps along this, or the equivalent Asian margin. If this is real, it implies that the earth deep beneath the Pacific at the core-mantle boundary is different from the non-Pacific hemisphere. Sites along the western margin of the Americas offers the possibility of detailed studies of the way the earth’s magnetic field varies with time as well as glimpses into the mechanisms involved in reversals.

The aim of our part of the overall project was to re-look at paleomagnetic data from the volcanic rocks of the Aleutian Island Arc. The first measurements of these samples were made in the late 1960’s and early 1970’s, when the paleomagnetic techniques available were not as capable as they are today. To obtain results that are acceptable by today’s standards for inclusion in the databases needed for modeling the geomagnetic field through time required new measurements. To be acceptable, all samples have to be progressively demagnetized using multiple steps. This allows an assessment of the stability of the magnetization recorded in the rock. Since the volcanic rocks sampled were all lava flows, they acquired their record of the ambient magnetic field as they cooled, thus the preferred demagnetizing technique was to heat them to successively higher temperatures, and cool them in a zero field chamber and remeasure them. This heat-cool cycle effectively removes any magnetization that is unstable at the highest temperature reached by any given heating step. By heating to successively higher temperatures, the magnetization acquired during any given temperature interval can be determined, and an estimate of the overall stability of the record established. A similar technique, using an alternating magnetic field that can be reduced to zero, give additional information on the magnetic characteristics of the samples.

Our results are based on samples taken from six locations on Unalaska, Umnak and Kanaga islands. In general we took about six 2.5cm diameter cores with a gasoline powered diamond drill form each lava flow. These cores were then subdivided into at least two 1 or 2 cm tall cylinders. One sample from each core was demagnetized by alternating magnetic fields, and the other thermally demagnetized. For the thermal studies we commonly used temperature steps of 25oC above 250oC up to 600oC. Almost all samples showed very clean demagnetization curves, with little change in direction of the recorded field with increasing demagnetization levels.

Samples were also selected for geochronologic studies. It is obviously important to know the age of each of the lava sequence sampled, and also to know the time-span represented by a sequence. Radiometric age determinations were made for selected samples from each of the flow sequences using Ar/Ar techniques. These give us reliable ages for the sequences, but determining the time-span of a sequence is often at the limits of the technique.

Since each lava flow records the magnetic field at an instant in geologic time, each sequence of lavas tells us something about sequential changes with time. All the sequences combined tell us about the longer-term behavior of the field. The results of our study indicate that overall time-variations seen over the two million years of our record look very similar to those seen in records from the last few centuries, and that the inferred location of the time-averaged geomagnetic dipole is very close to that expected from a dipole oriented along the spin axis of the earth. This latter point is important inasmuch as continental reconstructions based on paleomagnetism assume that the geographic (spin) axis and the time-averaged magnetic dipole axis always coincide.

Now that the results from our multi-university study are coming together, we can work towards getting a much clearer picture of at least one portion of the global variations in the geomagnetic field.

Current news |Research disciplines| People|Current Research Topics|
Student Research Opportunities|Analytical Facilities| Publications | Home Page