Modeling Viscosity of Volcanic Melts With Artificial Neural Networks
DOI: https://doi.org/10.1029/2022GC010673
Persistent URL: http://resolver.sub.uni-goettingen.de/purl?gldocs-11858/11145
Persistent URL: http://resolver.sub.uni-goettingen.de/purl?gldocs-11858/11145
Supplement: https://share.streamlit.io/domlang/visc_calc/main/final_script.py, https://doi.org/10.5281/zenodo.7317803
Langhammer, D.; Di Genova, D.; Steinle‐Neumann, G., 2022: Modeling Viscosity of Volcanic Melts With Artificial Neural Networks. In: Geochemistry, Geophysics, Geosystems, Band 23, 12, DOI: 10.1029/2022GC010673.
|
View/
|
Viscosity is of great importance in governing the dynamics of volcanoes, including their eruptive style. The viscosity of a volcanic melt is dominated by temperature and chemical composition, both oxides and water content. The changes in melt structure resulting from the interactions between the various chemical components are complex, and the construction of a physical viscosity model that depends on composition has not yet been achieved. We therefore train an artificial neural network (ANN) on a large database of measured compositions, including water, and viscosities that spans virtually the entire chemical space of terrestrial magmas, as well as some technical and extra‐terrestrial silicate melts. The ANN uses composition, temperature, a structural parameter reflecting melt polymerization and the alkaline ratio as input parameters. It successfully reproduces and predicts measurements in the database with significantly higher accuracy than previous global models for volcanic melt viscosities. Viscosity measurements are restricted to low and high viscosity ranges, which exclude typical eruptive temperatures. Without training data at such conditions, the ANN cannot reliably predict viscosities for this important temperature range. To overcome this limitation, we use the ANN to create synthetic viscosity data in the high and low viscosity range and fit these points using a physically motivated, temperature‐dependent viscosity model. Our study introduces a synthetic data approach for the creation of a physically motivated model predicting volcanic melt viscosities based on ANNs. Plain Language Summary:
Magma viscosity is a key parameter that controls the style of a volcanic eruption, whether it will be effusive or explosive. For this reason, any volcanic hazard mitigation plan requires detailed knowledge of this property. Melt viscosity can vary by up to 15 orders of magnitude (a factor of a quadrillion) with temperature and composition. Unfortunately, it is not possible to perform measurements over this range continuously in the laboratory, but only in two distinct temperature regimes, termed high and low viscosity ranges. In order to obtain a model to predict how composition and temperature control viscosity, we use machine learning and train an artificial neural network on a large viscosity database. This allows us to calculate high‐ and low‐temperature viscosity data that we call synthetic. Since most magmas are erupted at temperatures between the high‐ and low‐temperature ranges, we combine the synthetic data and a physically motivated equation to describe the dependence of viscosity on temperature. This model can compute viscosities in the region without measurements, including typical eruption temperatures of volcanoes. Our model serves the scientific community studying volcanic eruption mechanisms and its future prediction on a data driven basis. Key Points:
We train an artificial neural network that calculates temperature‐ and composition‐dependent viscosity of volcanic melts.
The neural network reproduces and predicts experimental viscosity significantly better than previous global models.
A synthetic data approach based on the neural network is combined with a physical model to predict viscosity at eruptive temperatures.
Statistik:
View StatisticsCollection
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.