WHAT IS NEEDED FOR ABSOLUTE PALEOINTENSITY?

Abstract:

Many alternative approaches to the Thellier and Thellier technique for absolute paleointensity have been proposed during the past twenty years. One reason is the time consuming aspect of the experiments. Another reason is to avoid uncertainties in determinations of the paleofield which are mostly linked to the presence of multidomain grains. Despite great care taken by these new techniques, there is no indication that they always provide the right answer and in fact sometimes fail. We are convinced that the most valid approach remains the original double heating Thellier protocol provided that natural remanence is controlled by pure magnetite with a narrow distribution of small grain sizes, mostly single domains. The presence of titanium, even in small amount generates biases which yield incorrect field values. Single domain grains frequently dominate the magnetization of glass samples, which explains the success of this selective approach. They are also present in volcanic lava flows but much less frequently, and therefore contribute to the low success rate of most experiments. However the loss of at least 70% of the magnetization at very high temperatures prior to the Curie point appears to be an essential prerequisite that increases the success rate to almost 100% and has been validated from historical flows and from recent studies. This requirement can easily be tested by thermal demagnetization while low temperature experiments can document the detection of single domain magnetite using the δFC/δZFC parameter as suggested (Moskowitz et al, 1993) for biogenic magnetite.