Investigating the utility of a high-temperature Thellier-style paleointensity experimental protocol

Abstract

AbstractHigh-quality data are vital to the research field of paleointensity, which has long suffered from poor-quality and/or sparse data. Previous paleointensity work has established that repeatedly heating specimens increases the opportunity for thermochemical alteration to decrease the reliability of paleointensity data. In addition, recent work has shown that repeatedly heating specimens in paleointensity experiments can also exaggerate the effects of non-ideal, non-single domain grains. Arai plots resulting from paleointensity experiments containing such grains are often curvilinear (two-slope) across most of the specimen’s unblocking temperature spectrum, except in the temperature range nearest to the grains’ Curie temperature. This study tests the following strategy to mitigate these effects: that of performing a Thellier paleointensity experiment using fewer temperature steps that are also concentrated at higher temperatures. For this purpose, we use natural specimens with well-constrained rock magnetic data from the Hawaiian Scientific Observation Hole 1 (SOH1) drill core in paleointensity experiments with starting temperatures ranging from 200 °C to 500+ °C. Those experiments that focused in on the portion of the unblocking temperature spectrum nearest the Curie temperature of the specimen (HiTeCT) had an exceptionally low success rate, whereas those with initial temperatures at comparatively moderate temperatures (200–400 °C) had high success rates (~ 70%). Thermochemical alteration was minimized with a start temperature of 400 °C, but the curvature of the Arai plots had no clear dependance on start temperature. We conclude herein that increasing the start temperature can help avoid the effects of low-temperature alterations. Additionally, we found that the approach of focusing in on the highest temperature range is not a feasible one to apply in paleointensity experiments, in general, and consider this likely to be a result of a lack of intermediate-temperature checks for alteration and insufficient independence of temperature steps. Graphical abstract