Reducing epistemic and model uncertainty in ionic inter-diffusion chronology: A 3D observation and dynamic modeling approach using olivine from Piton de la Fournaise, La Réunion

Abstract

The modeling of ionic diffusion in natural crystals has been developed over the last three decades to calculate timescales of geological processes. As the number of studies and the size of datasets have expanded, improvements in precision of the general technique are needed to resolve temporal patterns that would otherwise be masked by large uncertainties. This contribution examines fundamental aspects of timescale calculation uncertainty using Mg-Fe zonation in olivine crystals from a Piton de la Fournaise oceanite erupted in 2002CE. First, we quantitatively consider the role of geometric uncertainty in datasets, from the perspectives of sectioning angle, crystal shape and crystal agglomeration. Second, we assess how crystal growth and changing boundary conditions during diffusion pose problems for simplistic, 1-D, diffusiononly modeling. An initial database of 104 timescales (7-45 days) was generated using typical, 1-D, isothermal diffusion-only methods for profiles taken from 30 compositionally-and texturally-zoned crystals of olivine. The initial simplistic modeling yields poor model fits and imprecise timescales; prior to this work we would have rejected >60% of these data. Universal-stage measurements of crystal boundary angles and three-dimensional (3D) X-ray microcomputed tomography observations of crystal shape address geometric uncertainties. Ustage measurements show that, contrary to expectations of random sectioning, most boundaries modelled initially were close to the ideal sectioning plane. Assessment of crystal morphology from 2D thin sections suggests olivine crystals are dominantly euhedral, however, 3D imaging reveals that they are significantly subhedral and often exist as agglomerates, an observation which underscores both the potential for diverse crystal interactions through time in the magma This is the peer-reviewed, final accepted version for American Mineralogist, published by the Mineralogical Society of America. The published version is subject to change. Cite as Authors (Year) Title. American Mineralogist, in press.

Using Lunar Granulites to Constrain Re‐Equilibration Timescales of Contact Thermal Metamorphism on the Moon

“…For instance, zoning profiles can commonly display non‐flat asymptotes at either end of the sigmoidal curve, potentially reflecting relict growth zoning. Zoning profiles formed by a combination of diffusion and growth are challenging to model, since the interplay between diffusion and growth is also dependent on crystal growth rate and cooling rate (Chamberlain et al., 2014; Couperthwaite et al., 2021). For the diffusion modeling reported here, we only selected zoning profiles that display the best sigmoidal profiles.…”

Modelling ancient magma plumbing systems through clinopyroxene populations: a case study from Middle Triassic volcanics (Dolomites, Italy)

“…In these cases, mixing between the two melts was slower than the growth rate, which was nearly instantaneous, thus enabling the preservation of a double plateau in the external diopsidic-augitic compositional boundary. To avoid biases in the timescale calculations (see Couperthwaite et al 2021), we excluded these types from the diffusion chronometry modelling.…”

Supplemental Material: Deep magma mobilization years before the 2021 CE Fagradalsfjall eruption, Iceland

“…180 time estimates from modelling the diffusive relaxation of Fe-Mg zoning profiles in 123 olivine crystals have been obtained using Autodiff, an objective, rapid-modelling approach that exploits the self-similarity of one-dimensional diffusion profiles with similar boundary conditions to allow timescales to be determined quickly (Couperthwaite et al, 2021 and2020;Pankhurst et al, 2018;Allan et al, 2017;Hartley et al, 2016). The number of independent estimates (n=180) is larger than the number of crystals modelled (n=123) because some crystals record multi-stage events, which were modelled stepwise (Fig.…”

“…Timescale of deep magma mobilization. Timescale distributions for olivine (green)and plagioclase (blue) macrocrysts modeled using Autodiff(Couperthwaite et al, 2021 and2020; Pankhurst et al, 2018;Allan et al, 2017; Hartley et al, 2016) and DFENS(Mutch et al, 2021(Mutch et al, , 2019a(Mutch et al, , 2019b. (A and B) Kernel density estimates (KDEs) for olivine and plagioclase populations.…”