Introduction

yt is a Open Source Python package for analysis and visualization of volumetric data. It was originally written for analysis of multi-resolution astrophysical simulation outputs and has expanded its use cases into additional domains such as geodynamics, geophysics, weather simulation and engineering. yt can ingest data from a wide range of data structures including AMR grid patches, Octree structures, smoothed-particle hydrodynamic output and unstructured meshes. Additionally, a large portion of the methods in yt are parallelized with MPI and commonly used in HPC systems for analysis of simulation data.

Recent efforts to improve the use of yt in geoscience domains have focused on improving documentation (Havlin et al., 2020, 2021) and improving interoperability with other Python packages which has resulted in the yt_xarray package. xararay is a popular metadata-preserving array library with support for a large number of file formats including traditional formats like netCDF and HDF but also newer cloud optimized storage solutions like Zarr arrays.

The work presented here describes recent improvements to yt_xarray, most notably, the introduction of a coordinate transformation framework to simplify the steps required to utilize any of the methods in yt that rely on ray tracing (such as Volume Rendering). We also describe current efforts to leverage Zarr within the yt framework, both indirectly through the xarray backend exposed by yt_xarray and more directly within yt to access chunked particle data and multi-resolution grid structures.