an error about coupled SWAN and ROMS model

Discussion on computers, ROMS installation and compiling

Moderators: arango, robertson

Post Reply
Message
Author
Rocky_Wang
Posts: 50
Joined: Tue Mar 01, 2016 1:38 am
Location: the Ocean University of China

an error about coupled SWAN and ROMS model

#1 Unread post by Rocky_Wang »

Hello every,
I try to couple swan and roms using COAWST model, to simulate how wave and tide dominate the sediment transport. I do this work folloeing the Inlet_Test. I made a 3 hour wind file to force the swan model, at the same time roms model only forced by tide. When I run them separately, both of them are successful. But when I coupled them, an error occured after runing several days.
I turned on DEBUG switch and the result as following. I have been confused 2 days, Can you tell me the reason? or give me some clues. Thank you very much!

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+time 20160707.223000 , step 333; iteration 1; sweep 3 grid 1
2014 6 23:50:00 8.099446E-02 2.887666E+02 2.888476E+02 3.176563E+13
(166,231,20) 1.865871E-02 2.403729E-02 1.042405E+00 2.460445E+00
2015 6 23:55:00 8.120892E-02 2.887750E+02 2.888562E+02 3.176748E+13
(164,233,20) 1.729698E-02 2.088922E-02 1.168342E+00 2.442810E+00
2016 7 00:00:00 8.140875E-02 2.887843E+02 2.888657E+02 3.176956E+13
(164,233,20) 1.600080E-02 1.813718E-02 1.183052E+00 2.455510E+00
+time 20160707.223000 , step 333; iteration 1; sweep 4 grid 1
+time 20160707.230000 , step 334; iteration 1; sweep 1 grid 1
+time 20160707.230000 , step 334; iteration 1; sweep 2 grid 1
+time 20160707.230000 , step 334; iteration 1; sweep 3 grid 1
+time 20160707.230000 , step 334; iteration 1; sweep 4 grid 1
+time 20160707.233000 , step 335; iteration 1; sweep 1 grid 1
+time 20160707.233000 , step 335; iteration 1; sweep 2 grid 1
+time 20160707.233000 , step 335; iteration 1; sweep 3 grid 1
+time 20160707.233000 , step 335; iteration 1; sweep 4 grid 1
+time 20160708.000000 , step 336; iteration 1; sweep 1 grid 1
+time 20160708.000000 , step 336; iteration 1; sweep 2 grid 1
+time 20160708.000000 , step 336; iteration 1; sweep 3 grid 1
+time 20160708.000000 , step 336; iteration 1; sweep 4 grid 1
== SWAN grid 1 sent wave data to ROMS grid 1
forrtl: error (65): floating invalid
Image PC Routine Line Source
coawstG 0000000000EBF79C swoexa_ 4169 swanout1.f90
coawstG 0000000000E91137 swouto_ 828 swanout1.f90
coawstG 0000000000E3A3FF swan_cpl_ 731 swanmain.f90
coawstG 0000000000E047CC waves_control_mod 178 waves_control.f90
coawstG 00000000004385D2 MAIN__ 197 master.f90
coawstG 0000000000435ACC Unknown Unknown Unknown
libc.so.6 000000358481ED1D Unknown Unknown Unknown
coawstG 00000000004359C9 Unknown Unknown Unknown
** ROMS grid 1 recv data from SWAN grid 1
forrtl: error (78): process killed (SIGTERM)
Image PC Routine Line Source
mca_btl_sm.so 00007FB692012BAB Unknown Unknown Unknown
libmpi.so.1 00007FB69A8C7E39 Unknown Unknown Unknown
libmpi.so.1 00007FB69A7FA472 Unknown Unknown Unknown
mca_coll_tuned.so 00007FB6911458EE Unknown Unknown Unknown
mca_coll_tuned.so 00007FB69114C918 Unknown Unknown Unknown
libmpi.so.1 00007FB69A8079DA Unknown Unknown Unknown
libmpi_f77.so.1 00007FB69ABB7958 Unknown Unknown Unknown
coawstG 00000000009D006D distribute_mod_mp 3695 distribute.f90
coawstG 000000000044E7E1 ocean_coupler_mod 988 ocean_coupler.f90
coawstG 0000000000449B7F ocean_coupler_mod 777 ocean_coupler.f90
coawstG 0000000000455F1C ocean_coupler_mod 1358 ocean_coupler.f90
coawstG 0000000000489B10 main3d_ 181 main3d.f90
coawstG 0000000000439E4B ocean_control_mod 185 ocean_control.f90
coawstG 00000000004387E8 MAIN__ 211 master.f90
coawstG 0000000000435ACC Unknown Unknown Unknown
libc.so.6 000000358481ED1D Unknown Unknown Unknown
coawstG 00000000004359C9 Unknown Unknown Unknown

lfparejar
Posts: 12
Joined: Sat Oct 05, 2013 5:21 am
Location: Rutgers University

Re: an error about coupled SWAN and ROMS model

#2 Unread post by lfparejar »

Ideally, you might want to have winds in both ROMS and SWAN for the coupled runs. If you provide SWAN with wind forcing, ROMS will not know unless you create a ROMS wind forcing file too. That, or you can set a surface stress in ROMS using ANA_SMFLUX.

Regarding the SWAN error, SWAN outputs PRINT and ERRFILES. What do these say? This could give you a clue of what's missing.

Also, make sure that you have this in your header file

#define ROMS_MODEL
#define SWAN_MODEL
#define MCT_LIB

Post Reply