BOUT-dev VS HAIL-CAESAR

BOUT++: a framework for writing fluid and plasma simulations in curvilinear geometry

At least in my field (computational plasma physics), the majority of software is (in descending order) Fortran (e.g., SOLPS-ITER), Python (e.g., IPS; OMFIT; UEDGE), and C/C++ (e.g., BOUT++; Exascale Computing Project tools).

I don't want to paint too bleak a picture of my field - there are definitely big, well-supported projects that invest in code quality! For example, Bout++ is a fluid solver whose devs care a lot about best practices, and the folks at the Exascale Computing Project are doing great work with tools like Kokkos (GPU acceleration of HPCcodes), ADIOS (I/O for HPC), AMREX (meshes for HPC), etc.

EDIT: Already proposed solutions are " CAESAR-Lisflood" ( https://github.com/dvalters/HAIL-CAESAR ) and "SIBERIA" ( https://csdms.colorado.edu/wiki/Model:SIBERIA ), but both are difficult to use and have steep learning curves. I've also found the GRASS module "r.landscape.evol" ( https://grass.osgeo.org/grass78/manuals/addons/r.landscape.evol.html ), but this one seems even more complicated to use. So I'm open to easier alternatives. It would also be interesting if someone with the experiencein this could compare the proposed solutions (advantages/disadvantages, compare difficulty of use, ...).

Any idea which model would be best? For CAESAR-Lisflood there exists a derived HAIL-CAESAR model that has an easier to use GUI ( https://github.com/dvalters/HAIL-CAESAR ), and there seems to be at east some info and support available online. But for SIBERIA I can't find anything besides this: https://csdms.colorado.edu/wiki/Model:SIBERIA.