IN037-13
Visualization and Analysis of 3D Data in the Geosciences Using the yt Platform
Tuesday, 15 December 2020: 09:06
Virtual
Christopher Havlin1, Matthew Turk1, Benjamin K Holtzman2, Leigh Orf3, Kelton Halbert4, John B Naliboff5, Kacper Kowalik6, Madicken Munk6 and Sam Walkow6, (1)University of Illinois at Urbana Champaign, Urbana, IL, United States, (2)Lamont Doherty Earth Observatory, Columbia Univ., Palisades, NY, United States, (3)University of Wisconsin, Cooperative Institute for Meteorological Satellite Studies, Madison, WI, United States, (4)University of Wisconsin Madison, Cooperative Institute for Meteorological Satellite Studies, Madison, United States, (5)University of California Davis, Davis, United States, (6)University of Illinois at Urbana Champaign, Urbana, United States
Abstract:
We present a number of advances to the
yt platform for visualization and analysis of 3D observational and model data in the geosciences.
yt is an open source, python-based package originally designed for analysis of astrophysical datasets.
yt enables units-aware analysis and visualization of large 3D datasets, including 2D slices and projections, particle path analysis and 3D volume rendering. A number of recent efforts have focused on transforming
yt into a more domain-agnostic tool for use beyond astrophysics through new implementations within
yt itself and through connections to the wider open source ecosystem.
In this presentation, we highlight advances in using yt within the geoscience domains of atmospheric dynamics, seismic tomography and computational geodynamics. Within atmospheric dynamics, we demonstrate a new yt-native front end for CF-compliant netcdf files using 3D visualizations of supercell thunderstorms creating violent tornadoes. Within seismic tomography, we demonstrate using yt to generate 3D volume renderings of IRIS Earth Model Collaboration netcdf files that explore seismic tomography models at different lengthscales, furthering the goal of robust feature interpretation that is less dependent on hidden user choices in rendering parameters. And within computational geodynamics we demonstrate initial advances in visualizing both 2D and 3D output from the ASPECT mantle convection and lithospheric dynamics code using yt. Additionally, we present a number of new features including interactive visualization and shapefile support that help make yt a cross-domain tool for analysis and visualization in the geosciences.
