MY_SRC update for v4.2

[jdha]

Files to be updated (tick when done):

• bdyini.F90 [@jdha]
• domzgr.F90 [@jdha]
• icerst.F90 [@jdha]
• istate.F90 [@jdha]
• par_oce.F90 [@jdha]
• sbcrnf.F90 [@atb299 ]
• traadv_mus.F90 [@atb299]
• domain.F90 [@jdha]
• dtatsd.F90 [@jdha]
• in_out_manager.F90 [@atb299 ]
• sbcblk.F90 [@atb299 ]

lapack.F90 splines.F90 types.F90 utils.F90 are used for the usr_def routines - let us leave these for now
usr_def v3.6 dirs may be useful ... v3.6 has a bunch of TRC routines that might relate to the branch you were working on @atb299?

[jdha]

update to istate.F90 not required as MY_SRC version now includes additional: CALL day_init

[jdha]

update to bdyini.F90 not require as
Cyclic or symmetric condition
already removed

[atb299]

I think update to sbcrnf.F90 not required. As far as I can tell it only adds TRIM( cn_dir )// to enable the runoff files to reside in a different directory.

[jdha]

re domzgr.F90

Dave put in the following to make sure XIOS didn't hang in the land region of Antarctica - I've added it to v4.2 (just in case), but commented it out as I think that issue has now been solved in 4.2 and XIOS

! DRM 07/08/17 - Modify the top_level (ztop) and bottom_level (zbot) arrays to mask fake ocean points in
!                Antarctica. Need to convert the indices to the local values.
         k_top( mi0(5), mj0(5):mj0(405) ) = 0
         k_bot( mi0(5), mj0(5):mj0(405) ) = 0

[atb299]

re domzgr.F90

Dave put in the following to make sure XIOS didn't hang in the land region of Antarctica - I've added it to v4.2 (just in case), but commented it out as I think that issue has now been solved in 4.2 and XIOS

! DRM 07/08/17 - Modify the top_level (ztop) and bottom_level (zbot) arrays to mask fake ocean points in
!                Antarctica. Need to convert the indices to the local values.
         k_top( mi0(5), mj0(5):mj0(405) ) = 0
         k_bot( mi0(5), mj0(5):mj0(405) ) = 0

Yes, that issue is resolved in v4.2.

[atb299]

sbcblk.F90 change was related to output:

! DRM, 11/07/18 - move the beginning of the IF loop to get the diagnostics I want.
!      IF ( nn_ice == 0 ) THEN
         CALL iom_put( "qlw_oce" ,   zqlw )                 ! output downward longwave heat over the ocean
         CALL iom_put( "qsb_oce" , - zqsb )                 ! output downward sensible heat over the ocean
         CALL iom_put( "qla_oce" , - zqla )                 ! output downward latent   heat over the ocean
         CALL iom_put( "qemp_oce",   qns-zqlw+zqsb+zqla )   ! output downward heat content of E-P over the ocean
         CALL iom_put( "qns_oce" ,   qns  )                 ! output downward non solar heat over the ocean
         CALL iom_put( "qsr_oce" ,   qsr  )                 ! output downward solar heat over the ocean
         CALL iom_put( "qt_oce"  ,   qns+qsr )              ! output total downward heat over the ocean
      IF ( nn_ice == 0 ) THEN
         tprecip(:,:) = sf(jp_prec)%fnow(:,:,1) * rn_pfac * tmask(:,:,1) ! output total precipitation [kg/m2/s]
         sprecip(:,:) = sf(jp_snow)%fnow(:,:,1) * rn_pfac * tmask(:,:,1) ! output solid precipitation [kg/m2/s]
         CALL iom_put( 'snowpre', sprecip )                 ! Snow
         CALL iom_put( 'precip' , tprecip )                 ! Total precipitation
      ENDIF

The code has changed a lot between v4.0.6 and v4.2. Suggest sticking to v4.2 version for now.

[atb299]

Only change in in_out_manager.F90 is the inclusion of:

LOGICAL       ::   ln_reset_ts      !: use TS from initial condition file? (F) no or (T) yes

[jdha]

dropping

splines.F90
types.F90
utils.F90
lapack.F90

into the usr_def dir to deal with at a later date