Wrong array type passed to four NetCDF read routines
In the file 'main_src/cx_netcdf4_mod.f90', four subroutines (read_and_unscale_netcdf_*d) are using the wrong array type to read in data. They state that they "Read and Unscale [x]D arrays Two-Byte Integers", but they pass a 4-byte floating-point array to the NetCDF read routine. Fixed by passing a 2-byte integer array to the NetCDF read routine. This is not a good thing to do in general, causes a segfault when using the Intel Fortran compiler, and may be silently causing memory and/or other issues when using the gfortran compiler.
My code changes that reflect the above can be found in the "develop/refine_build_process" branch (see the recent relevant commit to that branch).