
Version arps5.1.0 (10/30/2003)
  
  10/30/2003  Note: ARPS I/O now supports only binary and HDF 4 format.
                    GrADS format was supported as a dump format only.

Version ARPS5.0.0IHOP_3

  * splithdf and joinhdf do not work well when hdfcompr >= 4.
                                 -- Fixed in version 5.0.0IHOP_5

Version ARPS5.0.0IHOP_2

  1. liba2io is not installed on paige, so 88d2arps_a2/88d2arps_rt may not
     be compilied on paige.  
     ---- Library a2io now delivered with ARPS since version 5.0.0IHOP_5.

Version ARPS5.0.0WDT

  1. mpi code for FCT scheme is still not complete.

Version ARPS5.0.0Beta7

  1. Error when run 25May1998 real case. 
     >> Maximum velocity component exceeded 150.0 m/s,
     More test needed.                      ----- Done.
  2. Same inconsistent results as (4) in Beta6 exist for periodic boundary
     condition.                             ---- Fixed in Beta8
  3. Beta6 version has a bug which was fixed temporary by change:
      <   CHARACTER (LEN=*  ) :: soiloutfl ! Surface data file name
      ---
      >   CHARACTER (LEN=80  ) :: soiloutfl ! Surface data file name
      at about 2226 line in iolib3d.f90.
                                          ---- Bug fixed permanently

Version ARPS5.0.0Beta6

  1. The current sfcphysics in Beta6 is untested. The work is in progress.
                                                           ----  Done in Beta7
  2. MPI run has problems for Periodic boundary condtion.  ----  Done in Beta7
  3. ADAS updates are in progress.                         ----  Done in Beta7
  4. Inconsistent results exist between MPI run and no MPI run for Open 
     (radiation) boundary condition. The tests were done on CAPS Origin 2000
     and on Origin machine at NCSA. However, the difference will disappear 
     when using "-opt 3" or lower level optimization option.
 

Ming's TODO list (Commented on 10/27/2003 by Yunheng)

  Update ARPGAGR. Working with single grid and May 20 storm test.
  Not with nested grid.

  snowcvr0 changed to snowdpth0                               -- Done

  incrfmt, ternfmt,sfcfmt,soilfmt,hdmpfmt,extdfmt are new.    -- Done

  Find MPI bug when using open LBC (see with may 20 storm test).
                                                              -- Done

  Complete testing, including at least one real data case

  Just want to let you know, when I tried to convert NCEP Reanalysis to
  ARPS, somehow there is a problem in 4.5.2. I got some kind of NAN.
  So I changed back to 4.5.0.2 which I've used it before. and It's no
  problem to convert. You may want to check it.
  Sam Chiao [schiao@unity.ncsu.edu]

  Need to distinquish between Tur64 Unix alpha and Alpha-Linux in makearps.

  Streamline arpssfc                                         -- Partially done

  Merge with Gene's latest version.                          -- Done

  Debug MPI for sadvopt=3,4.                                 -- Partially done

  Add restart test script                                    -- Done

  increase maximum length of filenames for plotting


  smooth9p_nobc does not include MPI call?
    IF ( smthflx == 1 ) THEN
      DO nsmth=1,numsmth
        CALL smooth9p_nobc( usflx, nx,ny,1,nx,  1,ny-1,tem9 )
        CALL smooth9p_nobc( vsflx, nx,ny,1,nx-1,1,ny,  tem9 )
        CALL smooth9p_nobc( ptsflx,nx,ny,1,nx-1,1,ny-1,tem9 )
        CALL smooth9p_nobc( qvsflx,nx,ny,1,nx-1,1,ny-1,tem9 )


TODO list (ARPS5.3.0 src/adas):

-continue updating to handle scalar array version of ARPS
-update cloud analysis to handle multiple categories/moments
-merge with MPI version of ADAS
