CONNECTING MARINE AND ON-LAND PALEOINITENSITY STUDIES

Paleomagnetism on marine sediments and land based volcanic rocks have been developed rather independently. Intensity studies are also the same that the relative paleointensity (RPI) sequence by sediment studies are not compared with the conventional absolute paleointensity (API) determinations. It is because the secular variation is, in general, a fast phenomenon that the error in age determination of volcanic rocks is larger than the typical period of secular variation. Here, we present an idea for correlating those two approaches, using wide spread tephra and welded tuff from a single pyroclastic flow.

Some of the volcanoes in Japan have had extremely large eruptions which deposit welded tuffs around the volcanoes, and disperse tephras to distance in the Pacific Ocean. Those huge eruptions provide us connecting land based paleointensity data to marine sediment studies. The welded tuff possesses TRM which allows us API study, while the wide spread tephra gives time marker in marine sediments. The correlation is a little tricky that it is, first, correlated to the sediments close to the Japanese archipelago. However, the sediments are not suitable for RPI study, since the magnetic mineralogy largely changes because of proximity to the land. Those sediments allow the tephra to the oxygen isotope stratigraphy, which can correlate to the relative paleointensity data from an abyssal plane. As a whole, the API is correlated to cyclostratigraphy by the tephra, and the cyclostratogaphy correlate them to the RPI. It is important to emphasize that the correlation is not altered even if the absolute age assignment is changed.

API studies are conducted with 21 welded tuffs of pyroclastic-flow deposits in Japan, utilizing LTD-DHT Show type paleointensity experiments. Out of the 21 tuffs, 10 are correlated to tephras in marine sediments. Combining three API data of excursions and reversals from Unzen, Japan, Auckland, New Zealand, and Tahiti, we can compare API and RPI in 13 stratigraphic level. Comparison with PISO-1500 and SINT-800 in VADM indicate that the API and RPI of both stacks has good linear correlation of high significance level, though PISO-1500 has better correlation coefficient. This result strongly confirms that the NRM-ARM ratio indeed represent the RPI, if sediment materials are carefully chosen.