Custom Query (964 matches)

We need add deallocation of communication array Arec in routine mp_scatter_state of distribute.F:

!
!  Deallocate receive buffer.
!
      IF (allocated(Arecv).and.(MyRank.eq.MyMaster)) THEN
        deallocate (Arecv)
      END IF

The mp_scatter_state is not currently used in ROMS. So this memory leak doesn't affect the ROMS MPI communications.

Also corrected an information type in def_rst.F.

Corrected the output variable long-name for output inert tracers. What ROMS computes in the age concentration and not the concentration mean age.

So we need to have:

    double dye_01(ocean_time, s_rho, eta_rho, xi_rho) ;
                    dye_01:long_name = "dye concentration, type 01" ;
                    dye_01:units = "kilogram meter-3" ;
                    dye_01:time = "ocean_time" ;
                    dye_01:coordinates = "x_rho y_rho s_rho ocean_time" ;
    double dye_01_age(ocean_time, s_rho, eta_rho, xi_rho) ;
                    dye_01_age:long_name = "age concentration, type 01" ;
                    dye_01_age:units = "second kilogram meter-3" ;
                    dye_01_age:time = "ocean_time" ;
                    dye_01_age:coordinates = "x_rho y_rho s_rho ocean_time" ;

instead of:

       double dye_01_age(ocean_time, s_rho, eta_rho, xi_rho) ;
                dye_01_age:long_name = "dye concentration mean age, type 01" ;
                dye_01_age:units = "second" ;
                dye_01_age:time = "ocean_time" ;
                dye_01_age:coordinates = "x_rho y_rho s_rho ocean_time" ;

So we corrected the output metadata for long_name. To compute the concentration mean age, we need to have dye_01_age/dye_01 during post-processing.

Many thanks to John Wilkin for bringing this to my attention.

I also corrected CPP typo in mod_stepping.F. We need to have ADJUST_BOUNDARY instead.

When the NEMURO model (nemuro_inp.h) is activated, ROMS reports the wrong parameter (AlphaZL, VO2S0, KD2N) values to standard output.

Many thanks to Paul Mattern for ​bringing this to our attention.

I also added a local CPP flag DEBUG_VOLCONS to obc_volcons.F to track the parallel global reductions:

# ifdef DEBUG_VOLCONS
          WRITE (150,10) MyRank, iic(ng), iif(ng), bc_area, buffer(1),  &
                         bc_flux, buffer(2), buffer(2)/buffer(1)
  10      FORMAT (i3,1x,i3.3,1x,i3.3,5(1x,1pe23.15))
          CALL my_flush (150)
# endif

This was done to track the round-off of global parallel reductions (big numbers) with different partitions. The default is to have this CPP off.