HR: 16:30h
AN: GP24A-03    [Abstracts]
TI: Fall in Earth's Dipole Moment is Intermittent
AU: * Gubbins, D
EM: gubbins@earth.leeds.ac.uk
AF: School of Earth and Environment, University of Leeds, Leeds, LS29JT United Kingdom
AB: The present fall in the Earth's dipole moment of about 5% per century and is well documented in the historical record. The paleomagnetic record suggests the fall began about 2,000 years ago. Although global intensity measurements did not begin until AD1840, good directional data is available from the 17^th century, which has been used to create a global model (gufm1 of Jackson et al., 2000) of the geomagnetic field for the entire interval 1590-1995. Prior to 1840 intensity in the model is normalised by assuming an arbitrary fixed rate of fall of the dipole moment equal to the average for the later period, an increase of 15~nT/yr in the Gauss coefficient g₁⁰. Paleointensity data are too inaccurate to be worth including in the model because their uncertainty exceeds the plausible secular variation of two centuries, but here we use the directional model to convert each paleointensity to an estimate of g₁⁰(t) at the time, t, of the measurement (which in most cases is uncertain). The Earth Ref Digital Archive provides 315 paleointensity measurements with assigned errors on both intensity and date. We converted measurements to estimates of g₁⁰(t) and corresponding errors and found the best-fit straight line through the data and the global model value for AD1840. This resulted in a slope of 2.28± 2.72~nT/yr for the interval 1590-1839, indistinguishable from zero and significantly different from the later value of 15~nT/yr. The χ² goodness-of-fit parameter shows that a straight line fits the data well (with 99% confidence) and a Student t-test indicates that the slope differs from the later value at the 99.9996% confidence level. The present fall in dipole moment is directly linked to growth and drift of reverse flux patches on the core-mantle boundary in the Southern Hemisphere: this activity is imaged by directional data alone and appears to have started in the late 18^th century, adding support to this new result of a steady dipole moment during the earlier period. We conclude that the activity responsible for the present fall in dipole moment is intermittent and that the long term fall may have taken place by a series of such events rather than by a continuous process.
DE: 1503 Archeomagnetism
DE: 1510 Dynamo: theories and simulations
DE: 1521 Paleointensity
DE: 1522 Paleomagnetic secular variation
DE: 1560 Time variations: secular and longer
SC: Geomagnetism and Paleomagnetism [GP]
MN: Fall Meeting 2005