Magnetization of Greenland ice and its relationship with dust content 

Journal of Geophysical Research, Vol. 109, D09104, doi:10.1029/2003JD004433, 2004 

L. Lanci
Istituto di Dinamica Ambientale, Università di Urbino, Urbino, Italy and Department of Geological Sciences, Rutgers University, Piscataway, New Jersey, USA.
D.V. Kent
Department of Geological Sciences, Rutgers University, Piscataway, New Jersey, USA and Lamont-Doherty Earth Observatory of Columbia University, Palisades, New York, USA
P.E. Biscaye
Lamont-Doherty Earth Observatory of Columbia University, Palisades, New York, USA
J.P. Steffensen
Ice and Climate, The Niels Bohr Institute, Juliane Maries Vej 30, DK-2100 Copenhagen, Denmark

ABSTRACT.
We estimate the concentration of fine magnetic particles in ice samples from the North Greenland Ice Core Project core from the central Greenland ice sheet, using low-temperature (77K) isothermal remanent magnetization (IRM) analysis and compare it with the mass concentration of aerosol dust. Samples were taken from six climatic intervals, spanning the time from the Holocene (Preboreal) back to the Last Glacial Dansgaard/Oeschger cycle 5. The mean IRM intensity of the ice varies by a factor of 3 from glacial to interglacial stages, being lower during interglacials. The IRM acquisition curves of the ice do not quite saturate at the maximum available field of 0.8 T and show a relatively broad coercivity, which is compatible with a mixture of maghemite or magnetite and hematite. Comparison of the IRM intensity and total dust mass shows a remarkably good correlation but also reveals a large background magnetization, which may be essentially constant over the different climatic stages. IRM suggests that the dust properties are independent of the background signal and that the dust aerosol has a magnetization within about 30% of pristine loess from the Chinese Loess Plateau, which is considered to have the same source in the same east Asian deserts as dust in Greenland ice. Ice contamination and the flux of extraterrestrial dust particles were considered in order to explain the origin of the background magnetization. Nevertheless, we could not find a convincing explanation for this signal, which represents a considerable part of the IRM signal and is the dominant component during interglacial intervals, without invoking the presence of undetected dust mass. The alternative hypothesis of a varying magnetization of the ice dust at different climatic periods would suggest that different sources of aerosol are active during different climatic periods. This, however, has not proven to be the case so far for studies of the provenance of dust in Greenland ice.