error: neg. dep(m) < -10.00

[yilong lyu]

Hi All,

I am trying to run a near global model with 1/2 resolution. When I was in the spin-up stage, I found that the SST would overheat significantly. Therefore, I have made some corrections to the shortwave radiation flux. When I used the corrected forcing field re-drive model, the model prompts neg. dp error after running to the 13th year.

After searching this forum, I found that many people’s problems are caused by batrop and baclin, so I adjusted these parameters and the problem remains. So I used the forcing drive model without correction, model can run stably for more than 20 years, which shows that this is related to my revised shwflx. Therefore, I performed spatial smoothing on my shwflx, but the problem remains, and there is still a problem with the 49th layer. 

I have attached my log files, where no_correct_batrop_50.out is not corrected shwflx, correct_batrop_50.out is corrected shwflx, correct_batrop_45.out is corrected shwflx and batrop is reduced to 45.

I tried many adjustments but still encountered the same error. Can anyone provide a clue about this error and how can I fix it?

Thanks in advance.

Yilong Lyu

[Alan Wallcraft]

If a simulation runs stably for years and then has problems the usual work around is to repeat the bad segment with half the baclin and the usual batrop (or close to it if baclin/batrop is then not an even number of time steps).  This is often enough to get past the problem and you can go back to the usual baclin.  If the problem happens frequently, then reducing baclin by 25% (say) for all segments might make sense.

You should try this, but if half baclin fails (perhaps even after going back for one more segment with the half baclin) then you should look at what is happening near the neg. dp location.  One common cause is loss of all stratification at high latitudes, but see what you get.  Note that hycom_ij2lonlat will tell you where the problem is based on the neg. dp i,j:

gaffney01 167> pwd
/p/home/wallcraf/hycom/GLBt0.72/topo
gaffney01 168> hycom_ij2lonlat 50 50
 109.440E  63.150S

The hycom_profile command will produce a .txt single point profile from a 3-D archive file, and you can also change itest jtest to a neg. dp point to get a .txt profile with a bit more information (and it gets written out every dsurfq and every diagfq days).

Alan.

[yilong lyu]

Dear Alan, Thanks for your reply. Your explanations are very helpful.

But I still encountered the problem of neg. dp(m). My model is near-global, a uniform grid with a resolution of 0.5 degree, ranging from 30 to389.5 and from 75S to 75N. The vertical grid is 50 layers and sigma0 is used. I found that this problem may be as you said, loss of all stratification at high latitudes. The point where the problem occurs is 45.5W, 70.0S. Below is my vertical grid layers, the thickness of the layer near the moment of error and the log file. How should I fix this problem? 

Hope your reply. 

Thanks in advance.

Yilong Lyu

>>>blkdat.input<<<

   35.0   'saln0 ' = initial salinity value (psu), only used for iniflg<2

   0      'locsig' = locally-referenced pot. density for stability (0=F,1=T)

   0      'kapref' = thermobaric ref. state (-1=input,0=none,1,2,3=constant)

   0      'thflag' = reference pressure flag (0=Sigma-0, 2=Sigma-2)

  25.0    'thbase' = reference density (sigma units)

   0      'vsigma' = spacially varying isopycnal target densities (0=F,1=T)

  17.00   'sigma ' = layer  1  density (sigma units)

  18.00   'sigma ' = layer  2  density (sigma units)

  19.00   'sigma ' = layer  3  density (sigma units)

  19.50   'sigma ' = layer  4  density (sigma units)

  20.00   'sigma ' = layer  5  density (sigma units)

  20.50   'sigma ' = layer  6  density (sigma units)

  21.00   'sigma ' = layer  7  density (sigma units)

  21.25   'sigma ' = layer  8  density (sigma units)

  21.50   'sigma ' = layer  9  density (sigma units)

  21.75   'sigma ' = layer 10  density (sigma units)

  22.00   'sigma ' = layer 11  density (sigma units)

  22.25   'sigma ' = layer 12  density (sigma units)

  22.50   'sigma ' = layer 13  density (sigma units)

  22.75   'sigma ' = layer 14  density (sigma units)

  23.00   'sigma ' = layer 15  density (sigma units)

  23.25   'sigma ' = layer 16  density (sigma units)

  23.50   'sigma ' = layer 17  density (sigma units)

  23.75   'sigma ' = layer 18  density (sigma units)

  24.00   'sigma ' = layer 19  density (sigma units)

  24.25   'sigma ' = layer 20  density (sigma units)

  24.50   'sigma ' = layer 21  density (sigma units)

  24.75   'sigma ' = layer 22  density (sigma units)

  25.00   'sigma ' = layer 23  density (sigma units)

  25.25   'sigma ' = layer 24  density (sigma units)

  25.50   'sigma ' = layer 25  density (sigma units)

  25.75   'sigma ' = layer 26  density (sigma units)

  26.00   'sigma ' = layer 27  density (sigma units)

  26.25   'sigma ' = layer 28  density (sigma units)

  26.50   'sigma ' = layer 29  density (sigma units)

  26.75   'sigma ' = layer 30  density (sigma units)

  27.00   'sigma ' = layer 31  density (sigma units)

  27.10   'sigma ' = layer 32  density (sigma units)

  27.20   'sigma ' = layer 33  density (sigma units)

  27.30   'sigma ' = layer 34  density (sigma units)

  27.40   'sigma ' = layer 35  density (sigma units)

  27.50   'sigma ' = layer 36  density (sigma units)

  27.55   'sigma ' = layer 37  density (sigma units)

  27.60   'sigma ' = layer 38  density (sigma units)

  27.65   'sigma ' = layer 39  density (sigma units)

  27.70   'sigma ' = layer 40  density (sigma units)

  27.72   'sigma ' = layer 41  density (sigma units)

  27.74   'sigma ' = layer 42  density (sigma units)

  27.76   'sigma ' = layer 43  density (sigma units)

  27.78   'sigma ' = layer 44  density (sigma units)

  27.80   'sigma ' = layer 45  density (sigma units)

  27.82   'sigma ' = layer 46  density (sigma units)

  27.84   'sigma ' = layer 47  density (sigma units)

  27.86   'sigma ' = layer 48  density (sigma units)

  27.88   'sigma ' = layer 49  density (sigma units)

  27.90   'sigma ' = layer 50  density (sigma units)

>>>Profile of idm=576, jdm=11 at 1968_119_00<<<

##   expt    idm    jdm    kdm   iloc   jloc   lonloc  latloc yrflag year day hr

##     60    720    301     50    576     11  317.500 -70.000      3 1968 119 00

## model-day  srfhgt  surflx     dpbl   dpmixl    tmix    smix   thmix    umix    vmix   ubavg   vbavg  steric  nsterc   oneta         E-P

# 24590.0000 -217.70  -251.8    6.971   11.406 -9.0328 34.2120 27.4235  -25.99   28.31   -1.58   13.59  -12.83 -204.87  0.9995    1608.803

#  k    utot    vtot  p.temp    saln  p.dens    thkns      dpth

   1  -25.99   28.31 -9.0328 34.2120 27.4235    3.000     1.500

   2  -25.59   25.58 -8.9901 34.2122 27.4266    3.375     4.688

   3  -24.43   23.87 -8.9012 34.2132 27.4333    3.797     8.273

   4  -19.35   22.98 -7.3104 34.2436 27.5451    4.271    12.308

   5  -21.01   22.05 -6.3095 34.2663 27.5999    4.805    16.846

   6  -21.19   21.93 -4.8457 34.3060 27.6603    5.406    21.952

   7  -21.38   21.97 -3.7344 34.3550 27.7020    6.082    27.696

   8  -20.81   21.50 -2.9922 34.3966 27.7278    6.842    34.158

   9  -20.53   20.80 -2.4648 34.4353 27.7490    7.697    41.428

  10  -20.96   20.29 -2.0886 34.4739 27.7707    8.660    49.606

  11  -21.28   19.93 -1.8809 34.5011 27.7867    9.742    58.807

  12  -21.23   19.60 -1.7895 34.5127 27.7932   10.960    69.158

  13  -20.80   19.04 -1.7337 34.5176 27.7954   12.000    80.637

  14  -20.13   18.49 -1.6478 34.5217 27.7959   12.000    92.637

  15  -19.28   17.94 -1.5088 34.5315 27.7990   12.000   104.637

  16  -18.48   17.44 -1.3409 34.5436 27.8026   12.000   116.637

  17  -17.45   17.01 -1.1362 34.5568 27.8053   12.000   128.637

  18  -16.72   16.76 -0.9273 34.5718 27.8087   12.000   140.637

  19  -16.52   16.58 -0.7794 34.5853 27.8130   12.000   152.637

  20  -16.99   16.44 -0.7416 34.5927 27.8173   12.000   164.637

  21  -17.83   16.30 -0.7347 34.5949 27.8188   12.000   176.637

  22  -18.39   15.76 -0.7305 34.5953 27.8189   12.000   188.637

  23  -19.50   14.86 -0.7288 34.5955 27.8190   12.000   200.637

  24  -21.96   13.66 -0.7279 34.5963 27.8196   12.000   212.637

  25  -24.03   11.93 -0.7271 34.5999 27.8225   12.000   224.637

  26  -23.83   10.28 -0.7269 34.6167 27.8361   12.000   236.637

  27  -22.65    9.18 -0.7269 34.6220 27.8404   12.000   248.637

  28  -23.86    9.38 -0.7276 34.6222 27.8406   12.000   260.637

  29  -27.86   10.49 -0.7300 34.6223 27.8407   12.000   272.637

  30  -32.33   12.25 -0.7347 34.6223 27.8409   12.000   284.637

  31  -35.40   14.21 -0.7420 34.6223 27.8413   12.000   296.637

  32  -37.24   16.10 -0.7504 34.6223 27.8417   12.000   308.637

  33  -37.18   18.23 -0.7569 34.6223 27.8420   12.000   320.637

  34  -41.87   19.91 -0.7629 34.6223 27.8422   12.000   332.637

  35  -48.51   21.41 -0.7742 34.6223 27.8427   12.000   344.637

  36  -51.44   23.28 -0.7853 34.6222 27.8432   12.000   356.637

  37  -47.93   27.31 -0.7952 34.6221 27.8435   12.000   368.637

  38  -31.51   35.22 -0.8028 34.6214 27.8434   12.000   380.637

  39  -11.28   47.42 -0.8128 34.6205 27.8430   12.000   392.637

  40    6.86   61.52 -0.8227 34.6198 27.8430   12.000   404.637

  41   21.11   75.48 -0.8299 34.6194 27.8429   12.000   416.637

  42   34.17   85.53 -0.8423 34.6185 27.8428   12.000   428.637

  43   47.56   93.49 -0.8479 34.6179 27.8425   12.000   440.637

  44   63.18   96.45 -0.8411 34.6184 27.8426   12.000   452.637

  45   79.33   69.26 -0.8337 34.6187 27.8425   12.000   464.637

  46   93.05  -19.79 -0.8412 34.6187 27.8428   12.000   476.637

  47   26.04    0.01 -0.8431 34.6185 27.8428  953.388   959.331

  48  -54.79 -147.20 -0.9908 34.6249 27.8545    1.754  1436.902

  49    0.80  -42.05 -1.2956 34.6412 27.8800 1370.132  2122.845

  50  -24.00   45.82 -1.3448 34.6635 27.9000  973.510  3294.667

>>>log file<<<

 step  2360640 day  24590.00000 -- archiving completed --

 tau_ewd: day,span,range =  24591.000   1.0000  -1.1603385E+00   6.1683679E-01  

 tau_nwd: day,span,range =  24591.000   1.0000  -6.1652213E-01   8.8126224E-01  

 wnd_spd: day,span,range =  24591.000   1.0000  5.6833960E-03   2.7552412E+01   

  2360682 i,j,k=  577   11 48 neg. dp (m) in loop 19      -0.24     max

  2360685 i,j,k=  576   11 48 neg. dp (m) in loop 19       -1.6     max

  2360688 i,j,k=  574   11 48 neg. dp (m) in loop 19       -7.6     max

  2360691 i,j,k=  573   11 48 neg. dp (m) in loop 19       -7.6     max

  2360694 i,j,k=  573   11 48 neg. dp (m) in loop 19      -0.26     max

  2360700 i,j,k=  575   13 48 neg. dp (m) in loop 19       -3.3     max

  2360703 i,j,k=  575   13 48 neg. dp (m) in loop 19       -1.0     max

  2360736 (1968/120 00) mean      SSH (mm):   97.49  (-2.4E+03 to  1.4E+03)

  2360736 (1968/120 00) mean MontgPot (mm):   97.49

 airtmp: day,span,range =  24592.000   1.0000  -5.2439980E+01   3.7327415E+01   

 vapmix: day,span,range =  24592.000   1.0000  2.4154548E-05   2.0886566E-02    

 precip: day,span,range =  24592.000   1.0000  0.0000000E+00   1.1107768E-06    

  2360739 i,j,k=  577   11 48 neg. dp (m) in loop 19      -0.76E-01 max

  2360742 i,j,k=  576   11 48 neg. dp (m) in loop 19       -13.     max

  2360742 i,j,k=  576   11 48 neg. dp (m) in loop 19       -13.     fatal

error: neg. dp (m) <    -10.00

[Alfatih]

Hi Yilong Lyu,

It seems that your  neg dp occurs close to the open boundary. I have also experience these problem when using boundary nesting. 

It will probably help tuning  your e-folding time and the width of the relaxation zones used to generate the rmu files.

Regards,

Alfatih

[Alan Wallcraft]

For global cases, we typically setup the dp00 (fixed depth) profile to reach at least 2500m, but with dp00x=12 your fixed depth profile reaches about 500m.

I suggest setting dp00x=100.0, which would put interface 49 at 2700m (or deeper).  This changes the vertical grid, so you would need to start again with new relax files.

If you want more resolution near the surface you can explicitly define every dp00 in the blkdat.input file.  This is most easily done using hycom_dp00 from HYCOM-tools, see dp00_50Lnew.csh

Alan.