Hi, and sorry about this very late response.
We have a working version of the MPAS dynamical core in the Community Atmosphere Model (CAM) that is the atmospheric component of the Community Earth System Model (CESM), and this is part of the System for Integrated Modeling of the Atmosphere (SIMA) project. We have implemented the CAM energy fixer, and tests indicate that MPAS/CAM is viable for traditional climate applications. We are now testing MPAS/CAM using MPAS variable-resolution atmospheric meshes to get down to convection-permitting resolution (i.e. around 3 km cell spacing). I think, optimistically, we might release this to the community by the end of this calendar year. The current issue we are trying to resolve is the memory usage required to run high-resolution meshes; the memory usage does not scale linearly with problem size/processor count, and this becomes an issue for high resolution meshes (e.g., for uniform meshes < 15 km cell spacing on cheyenne).
We do not intend to bring the CAM physics over to the stand-alone MPAS release. We are, however, intending to continue to evolve and support the stand-alone MPAS release that uses physics shared with WRF.
For more information on SIMA, see https://sima.ucar.edu/
You might also find interesting another project that is using MPAS in CESM, that is developing a full GPU implementation of MPAS Atmosphere and Ocean in CESM. The NSF-funded project is called EarthWorks and the webpage (which is a work in progress) is
http://hogback.atmos.colostate.edu/earthworks/
and the initial announcement: https://www.atmos.colostate.edu/2020/09/faculty-ncar-partner-on-5m-nsf-project-to-bolster-earth-system-modeling-capabilities/
