EVALUATION OF A MULTI-SPECIMEN APPROACH TO PALEOINTENSITY DETERMINATION

Hoffman et al.'s multi-specimen variation of the Thellier palaeointensity method has been applied to a number of historically dated lavas from Etna (Sicily), Vesuvius (Italy), and the Canary Islands. The majority of experiments resulted in palaeointensity estimates which were in good agreement with values obtained by other palaeointensity methods and geomagnetic field models. A lava which produced an overestimate of the actual field value using the Rolph and Shaw method also failed to give a reasonable intensity with the Hoffman method. A Miocene lava from New Zealand which had given a very high field intensity using the Rolph and Shaw method gave a lower value with Hoffman, but it is uncertain whether this value is a better estimate of the actual field intensity. Detailed rock magnetic studies were performed on every sub-specimen of six Miocene lavas from the Vogelsberg (Germany): bulk susceptibility and IRM acquisition curves were measured, and the magnetic minerals were observed using reflected light microscopy. Changes in susceptibility were often observed, but they do not correspond with changes in TRM capacity as observed on the Arai plots. Very similar susceptibility versus heating temperature plots were observed from sister samples which had been used in modified Thellier experiments, suggesting that the majority of thermal alteration occurs on the first heating to a particular temperature. The IRM and optical observations showed no obvious trends which could be related to thermal alteration.