Interrogating Commonly Applied Initial Condition Assumptions in Geospeedometry using NanoSIMS

Abstract:

The geologically short (days to centuries) timescales associated with thermochemical changes in magma chambers during the prelude to eruption are typically beyond the resolution of long-lived radioisotopic geochronometers but can be resolved by “geospeedometry” that quantifies the relatively rapid diffusional relaxation of compositional zoning in igneous phenocrysts. When combined with absolute dating, geospeedometry can reveal long-term chronologies of compositional and thermal oscillations in magma chambers. The ability to accurately constrain timescales via geospeedometry is limited in part by the spatial resolution of commonly used analytical techniques and the derivative chemical profiles, especially in the case of very short timescales or very slow diffusing elements, where the chemical profile is often approximated as a step-function. We present geospeedometry of chemical profiles collected with the NanoSIMS ion microprobe in multi-collection mode with sub-micron resolution (e.g., 0.3 micron spacing). This data facilitates a comparison of how well the timescales calculated with the different experimentally-determined diffusivities and collected along the same profile agree and an interrogation of commonly made model assumptions in geospeedometry. For example, electron probe profiles across internal sanidine zone boundaries from a ca. 250 ka rhyolite lava from Yellowstone Caldera reveal a step function in anorthite content at 10 micron spacing and evidence for earlier dissolution prior to new zone growth, yet we find no observable difference in the width of Ba, Sr and Mg diffusion profiles collected via NanoSIMS for the same profile. These observations support the hypothesis that very little to no diffusive relaxation has affected the initial concentration profile that was produced by crystal growth during magma mixing. Our results highlight the need to quantitatively constrain initial conditions when applying geospeedometry to intermediate to silicic magmatic systems where the timescales of mixing between the intruding magma and that in the existing reservoir may not be approximated as instantaneous compared with the timescales of crystal growth.