blow up the model

[kobl1201]

Hello

i have some problem, when i worked the ROMS model.

I applicated a boundary file every 1 day, and atmospheric forcing every 1 hour.

Starting time is 2019-02-07 00:00:00

ROMS model worked well until 2019-02-07 10:32:00, and then the model status had blow-up.

When i checked the log file, avgpe had NaN.

I check the almost input files, which are boundary, initial, atmospheric forcing files, grd file and so on.

I think they did't have strange values, but i don't understand why the model has blow-up at 10:32:00.

When i change the DT, VISC2 and so on, it's not useful.

Please tell me the reason.

Thank you~

[kate]

You need to investigate some more. Have you tried a shorter timestep? Have you looked at where the thing died? Is it surface? Boundary? Middle or what?

[kobl1201]

Hello Kate.

I tried to change the shorter time for DT, but it is not useful as well.

And i also changed the bathmetry, but not useful.

I printed the values as follows.


TIME-STEP YYYY-MM-DD hh:mm:ss.ss u2v2 POTEN_ENRG my_avgpe NET_VOLUME
C => (i,j,k) Cu Cv Cw Max Speed


1242 2019-02-07 10:21:00.00 0.000000E+00 8.902596E+03 7.934201E+10 2.232791E+14
(201,102,41) 4.653923E-04 2.943751E-03 6.282829E-01 1.319888E+00
1244 2019-02-07 10:22:00.00 0.000000E+00 8.902596E+03 7.934201E+10 2.232791E+14
(201,102,41) 2.571881E-04 2.907016E-03 6.263029E-01 1.323136E+00
1246 2019-02-07 10:23:00.00 0.000000E+00 8.902596E+03 7.934201E+10 2.232791E+14
(200,102,41) 3.051231E-04 1.685668E-03 5.623060E-01 1.325877E+00
1248 2019-02-07 10:24:00.00 0.000000E+00 8.902596E+03 7.934201E+10 2.232791E+14
(179,272,41) 1.289501E-03 1.541460E-04 4.953033E-01 1.326764E+00
1250 2019-02-07 10:25:00.00 0.000000E+00 8.902596E+03 7.934201E+10 2.232791E+14
(131,326,41) 1.084630E-03 1.168492E-03 5.457481E-01 1.325491E+00
1252 2019-02-07 10:26:00.00 0.000000E+00 8.902596E+03 7.934201E+10 2.232791E+14
(199,103,41) 2.318173E-03 1.087109E-04 5.156771E-01 1.322848E+00
1254 2019-02-07 10:27:00.00 0.000000E+00 8.902595E+03 7.934201E+10 2.232791E+14
(110,353,41) 1.216523E-03 7.193629E-04 4.409253E-01 1.320071E+00
1256 2019-02-07 10:28:00.00 0.000000E+00 8.902595E+03 7.934201E+10 2.232791E+14
(110,353,41) 1.294690E-03 7.477884E-04 5.284653E-01 1.317983E+00
1258 2019-02-07 10:29:00.00 0.000000E+00 8.902595E+03 7.934201E+10 2.232791E+14
(109,354,41) 7.118062E-04 1.313774E-03 5.210776E-01 1.316565E+00
1260 2019-02-07 10:30:00.00 0.000000E+00 8.902595E+03 7.934201E+10 2.232791E+14
(109,354,41) 4.494276E-04 8.878287E-04 4.329849E-01 1.315157E+00
1262 2019-02-07 10:31:00.00 0.000000E+00 8.902595E+03 7.934201E+10 2.232791E+14
(199,103,41) 5.588593E-04 1.999664E-03 4.282117E-01 1.312994E+00
1264 2019-02-07 10:32:00.00 0.000000E+00 NaN 7.934201E+10 2.232791E+14
(199,103,41) 2.872736E-04 2.096221E-03 4.320387E-01 1.309631E+00


Actually, i can't find where occurred the blow-up...

When i checked the re-start file, the values didn't show the unbalanced values at all layers.

Thank you~

[kate]

So save a record right before it blows up and look at it. I can't do it from here.