Iron-Oxide Geochronology to Constrain the Formation of Soils and Paleosols

Citation

Hofmann, Florian (2019) Iron-Oxide Geochronology to Constrain the Formation of Soils and Paleosols. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/PHV4-Y279. https://resolver.caltech.edu/CaltechTHESIS:05312019-131059095

Abstract

In this thesis, I show how iron-oxide geochronology can be applied to soils and paleosols and I explore new applications of the (U-Th)/He and cosmogenic ³He dating methods. In Chapters II and III, I apply the (U-Th)/He method to goethite pisoliths in paleosols and fissure fillings of the Bohnerz deposits, which are a widespread erosional lag deposit in Central Europe. I show that this deposit formed between ~55 Ma and 2 Ma and not in the late Cretaceous-early Eocene, as was previously claimed. A map compiled from published sources shows the paleo-extent of the Bohnerz deposit, which were developed on every limestone plateau in Central Europe. Concentrations of cosmogenic ³He measured on pisoliths in paleosols, demonstrate that ancient cosmic-ray exposure occurred over at least 5 Ma, possibly 10-20 Ma. This shows that the Bohnerz deposits represent surfaces, which were stable for tens of millions of years before burial. Even today, these surfaces experience extremely low erosion rates of ~0.1 m/Ma, comparable to some of the Earth's most stable surfaces in arid environments. The hiatus represented by the Bohnerz unconformity lasts 125-150 Ma, yet only 30-40% of that duration is recorded in Bohnerz deposits. Pisoliths of the Bohnerz deposits are a continuous record of Central European continental climate for most of the Cenozoic.

In Chapter IV, I develop a laboratory technique to enable (U-Th)/He dating of hematite samples by the single-aliquot method. Highly retentive hematite samples have to be heated to >1000 °C to fully degas helium, but U is lost from the sample at around 980 °C. Through infrared spectroscopy and trace element analysis of heated samples, I show that U-loss correlates with a phase change from hematite to magnetite. Delaying this phase change to higher temperatures extends the usable range of temperatures to which samples can be safely heated without U-loss. This is achieved by degassing samples in a partial oxygen pressure of around 100 mbar, which permits degassing of samples up to 1150 °C without loss of U. I demonstrate that precise and accurate (U-Th)/He ages can be obtained for hematite samples, which agree with established two-aliquot ages. I show how this method can be implemented and automated.

In Chapter V, I extend the use of ³He cosmogenic dating to fine-grained iron-oxide particles, which are abundant in most types of modern soils. Diffusion modeling predicts that hematite particles down to ~10 nm should quantitatively retain helium for at least 1 Ma at Earth-surface conditions. In order to test whether pedogenic iron-oxides can be used for geochronology, I study a vertical profile of a relict soil developed on a fanglomerate terrace at Whitewater Hill, California. Profiles of ¹⁰Be and ²⁶Al in detrital quartz agree well with an exponential decrease in cosmogenic nuclide production and they yield an exposure age of 52.4±2.2 ka assuming no erosion. The vertical profile of ³He concentrations in pedogenic iron-oxides shows a decrease with depth, but concentrations are higher than expected in the 40-100 cm depth range. This indicates vertical movement of iron-oxides in the soil, which is a well-known soil formation process. These observations are more consistent with a soil age of 208±44 ka. This approach yields information on both age constraints and the formation and migration of pedogenic iron-oxides in the soil column.

Thesis Files