The Earth's magnetic field varies on a variety of time scales, with perhaps the best known example being the significant changes in the direction of local magnetic north documented in historic times. In addition to these short term variations, there are many documented reversals of the polarity of the field during geologic history. The change in polarity through recent geologic time has been well documented and provides a key method of dating geological events and establishing the past motions of tectonic plates. The Earth's magnetic field also exhibits profound variations in intensity. For example, the intensity of the field has decreased by approximately 10% in historic time and five-fold variations in intensity have been documented over longer time scales. A comprehensive description of magnetic field variations, in direction and intensity and over a wide range of time scales, is desirable as these fluctuations can provide critical constraints on how the magnetic field is generated as well as several other deep Earth processes.
Compared with directional variations, documenting past intensity fluctuations of the Earth's magnetic field is much more difficult and, consequently, relatively few reliable determinations of absolute intensity (paleointensity) are available. This scarcity of paleointensity information is primarily the result of the difficulty in identifying geological materials that contain dominantly the very fine and stable magnetic particles that are required for determining paleointensity. We are investigating the potential of ash flow tuffs, generated by explosive volcanic eruptions, as a potential new material for paleointensity information. Although such deposits have the requisite fine magnetic particles, a variety of post-emplacement processes may potentially affect the ability to recover ancient field intensity information. As a test of this material, we are documenting the post-emplacement thermal history and determining paleointensity for samples from two historical ash flows. The 1980 ash flows at Mt. St. Helens, Washington, and the 1912 flows from Mt. Katmai in the Valley of Ten Thousand Smokes, Alaska provide a natural laboratory for testing measured paleointensities against known field values. Direct temperature measurements at both localities constrain the emplacement temperature, and significant information is available on the conditions of post-emplacement fumarolic activity. Combined with focused studies in the older (0.76 Ma), better-exposed Bishop Tuff, our sampling strategy will allow us to evaluate the suitability of ash flows for paleointensity analysis, as well as the feasibility of identifying (in the field or lab) samples most likely to provide reliable results. The proposed work will provide valuable information to to guide future workers in both the field and the laboratory in the selection of suitable materials for paleointensity analysis. Ash flows (many with high quality radiometric ages) are common worldwide and commonly can be isotopically dated with high precision. Thus, if absolute paleointensities can be determined from ash flows, it may be possible to compile a much more comprehensive database of past field intensity fluctuations of the geomagnetic field.
In addition to the scientific goals of this project, the research is providing training for a graduate student at Scripps Institute of Oceanography and training of two undergraduate students at the University of Minnesota. Additional undergraduate student participation is being encouraged by participation of students from the University of Minnesota?s NSF-Research Experience of Undergraduates (REU) program.
Understanding how Earth’s magnetic field changes through space and time gives us insight into a wealth of geologic processes and history, including processes in Earth’s liquid core; planetary and atmospheric evolution; and how Earth’s tectonic plates have shifted through time. It also allows us to place age constraints on many types of rock sequences and some archeological artifacts. Geologists aim to answer many of these questions by studying the "rock records" of magnetic field variation, such as those contained in volcanic rocks. Both the direction and strength of the field is recorded by volcanic rocks when they erupt and cool, and past variations in the field direction are relatively well known from sampling many thousands of rock units. By contrast, the so-called "paleo"intensity variations have been determined from only a few hundred cooling units, with more than half of these limited to the past 300 thousand years. This scarcity of paleointensity data stems largely from the fact that it is far more difficult to acquire than directional data, for reasons including sample alteration during laboratory protocols and non-ideal magnetic mineralogy. The theory behind all absolute paleointensity studies is strictly valid only for samples with very fine-grained (single-domain) magnetic minerals that carry a magnetization acquired on cooling from high temperatures (typically greater than 250-600°C or 392-1112°F). Because true single-domain grain size is rare in nature, most work has focused on basaltic lava flows and other volcanic products with slightly larger magnetic minerals. The alternative is to find so-called "ideal" paleointensity materials that naturally have a smaller grain size and may be less sensitive to sample alteration. Rapidly cooled volcanic glasses have been shown to be a near ideal paleointensity recorder in some circumstances, and the goal of this study was to examine whether or not a certain type of glassy volcanic flow (ignimbrites) can be reliably used for paleointensity analyses. Ignimbrites (also called "ash flow tuffs") are a type of volcanic flow composed of hot gas, ash, pumice and rock fragments. The ash and pumice fragments are glassy and contain the ideal single-domain magnetic particles. However, ignimbrites tend to have more complicated geological histories than lava flows in ways that may affect their ability to faithfully record field intensity. Ignimbrites may be emplaced over a wide variety of temperatures. Magnetic mineralogy can vary widely. After deposition, flows may be altered when hot water percolates through the flows, or parts of some flows may become "welded" when the individual ash particles deform under heat and pressure, resulting in a more dense rock This project was designed to assess the reliability of ignimbrites for recording the intensity of Earth’s magnetic field at the time they are emplaced. To do this, we sampled two historical ignimbrites: Mt. St. Helens, erupted in Washington State in 1980, and Novarupta, erupted in Alaska in 1912. For these two flows, we know what the field intensity should be, so we can compare our laboratory estimates of paleointensity with the known value and with the variables of interest (emplacement temperature, degree of welding, hydrothermal alteration). We also sampled the Bishop Tuff, erupted about 760 thousand years ago, which has much more variability in welding than the two historical flows. Key outcomes from the project include the documentation of substantial variation in magnetic properties and geomagnetic intensity estimates that is apparently correlated with hydrothermal or vapor-phase alteration in the large volume Bishop and Novarupta ignimbrites. By contrast, small volume flows (such as at Mt. St. Helens) may more reliably record geomagnetic field intensity. In the Novarupta and Mt. St. Helens samples, standard field or laboratory protocols were able to identify and reject samples that gave incorrect paleointensity estimates. The same was not true throughout the Bishop tuff, however, and in future paleointensity work great care should be taken to identify and avoid sections of ignimbrite that have undergone hydrothermal alteration. An unexpected outgrowth of this project was the identification of an atomic-scale process within crystals of a common magnetic mineral (titanomagnetite). The way in which the atoms are arranged within the crystal structure depends on the thermal history of the sample and results in measurable changes to magnetic properties of titanomagnetite-bearing rocks. This phenomenon is the subject of ongoing work, but we believe it can be developed into a tool to determine emplacement temperature and cooling rate of some types of volcanic flows, which has implications for volcanic hazard assessment. This project contributed to the training and development of two graduate students and two undergraduate students, who participated in field work, laboratory work, or both. The two undergraduate students supported on this grant gained a basic background in the field of rock magnetism and paleomagnetism, and were involved in sample preparation and generation of data, as well as data processing and interpretation.
