Opened 3 years ago

Closed 3 years ago

#1950 closed help (worksforme)

aerosol dry depositions zero everywhere in regional outputs

Reported by: ggxmy Owned by: um_support
Component: UKCA Keywords: dry deposition STASH
Cc: Platform: MONSooN
UM Version: 10.3

Description

Dear CMS,

I requested UKCA aerosol dry depositions, STASH items 38-214~228 in global (glm_um) and regional (um) models in UM vn10.3 nested suite. They appear fine in global outputs but have value zero everywhere in all regional outputs. These fields have value zero everywhere even in regional dump files on which I have no control.

But if I add print statements, I can see these are not zero in ukca_aero_ctl.F90. So I guess this is likely to be a problem in STASH. I tried quite a few things in the STASH settings but with almost all of these settings, I got dry deposition diagnostics OK in global outputs but got values zero in the regional outputs.

Have you heard a problem like this? Could you suggest anything I can try?

Masaru

Change History (7)

comment:1 Changed 3 years ago by grenville

Masaru

Please tell us the suite id.

Grenville

comment:2 Changed 3 years ago by ggxmy

Hi Grenville,

It is u-af172. u-af349 is also the same but this one contains print statements in ukca_aero_ctl.F90 to check the values of dry depositions. They are found not to be zero in the computation.

Thanks,
Masaru

comment:3 Changed 3 years ago by grenville

Masaru

I have been in contact with Mohit about this. He will try to look at the problem. He also told me

"almost all the knowhow related to (coding of) the Section-38/GLOMAP diagnostics is at Leeds" — have you contacted your Leeds colleagues for help?

Grenville

comment:4 Changed 3 years ago by ggxmy

Hi Grenville,

I had asked Mohit, Graham, and Leeds aerosol group email list for help before I consulted here. Especially I have been working closely with Mohit. For all that, I couldn't find a solution for this.

The print statements I added in ukca_aero_ctl.F90 of u-af349 output positive aerosol dry depositions. I guess this means the problem is likely to be related to STASH rather than the codes of the section 38?

Thanks.
Masaru

comment:5 Changed 3 years ago by grenville

Masaru

could you check that mode_diags has data you expect (this is in ukca_main1-ukca_main1.F90)?

IF (sf(item,section) .AND. item > item1_mode_diags+12) THEN

! DEPENDS ON: copydiag_3d
CALL copydiag_3d (stashwork38(si(item,section,im_index)), &

mode_diags(:,:,:,icnt), &
row_length,rows,model_levels,0,0,0,0, at_extremity, &
stlist(1,stindex(1,item,section,im_index)),len_stlist, &
stash_levels,num_stash_levels+1, &
atmos_im,section,item,icode,cmessage)

Grenville

comment:6 Changed 3 years ago by ggxmy

Thanks for your suggestion. I added a print statement here;

DO l=1,nmax_mode_diags
  IF (UkcaD1codes(imode_first+l-1)%item /= imdi) THEN
    icnt = icnt + 1
    item = UkcaD1codes(imode_first+l-1)%item
    section = MODE_diag_sect
    IF (sf(item,section) .AND. item > item1_mode_diags+12) THEN
      ! DEPENDS ON: copydiag_3d
      print*,'icnt,item,maxval(mode_diags(:,:,:,icnt)): '   & !MY added
             ,icnt,item,maxval(mode_diags(:,:,:,icnt))        !MY added
      CALL copydiag_3d (stashwork38(si(item,section,im_index)),   &
        mode_diags(:,:,:,icnt),                                   &
        row_length,rows,model_levels,0,0,0,0, at_extremity,       &
        stlist(1,stindex(1,item,section,im_index)),len_stlist,    &
        stash_levels,num_stash_levels+1,                          &
        atmos_im,section,item,icode,cmessage)

In the output file;

/home/myosh/cylc-run/u-af349/log/job/20130803T0000Z/COPE_1p0_L70_ukv_um_fcst_002/01/job.out

positive values seem to have been written out;

 icnt,item,maxval(mode_diags(:,:,:,icnt)):  10,  214,  1.45635344735002055E-13
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  11,  215,  1.01277454923448151E-5
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  12,  216,  1.94692268071352821E-4
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  13,  217,  2.48245580621182094E-5
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  14,  218,  9.1874087253134959E-6
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  15,  219,  7.09801292117544051E-3
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  16,  220,  5.23523425526045236E-5
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  17,  221,  3.20578711029127116E-5
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  18,  222,  2.61509258566083155E-7
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  19,  223,  3.73180526641611358E-5
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  20,  224,  4.67547891900096829E-9
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  21,  225,  9.25077267356176348E-5
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  22,  226,  1.26867793206920872E-4
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  23,  227,  5.92244737348536686E-6
 icnt,item,maxval(mode_diags(:,:,:,icnt)):  24,  228,  7.5603229830484113E-5

I think these are aerosol dry depositions. Lines below are copied from ukca_aero_ctl.f90;

      ! ddep SU -- 214-217
      ! ddep SS -- 218-219
      ! ddep BC -- 220-223
      ! ddep OC -- 224-228
      ! ddep SO -- 229-232
      ! ddep DU -- 233-236

Thanks,
Masaru

comment:7 Changed 3 years ago by ggxmy

  • Resolution set to worksforme
  • Status changed from new to closed

Hi,

For a scientific reason, I changed the code so that aerosol depositions are calculated at every model time step instead of chemistry time step. I also changed time averaging method in STASH so the output data are sampled at every time step. Now the outputs contain positive dry depositions.

So even though I couldn't pinpoint the problem, I can work around this issue. I'm closing this case now. If this comes back as an issue in the future, I may create a new ticket for the same topic.

Thank you for your help.
Masaru

Note: See TracTickets for help on using tickets.