Hello,
I use the COAWST model to couple SWAN and ROMS, and they exchange variables throuth MCT every 3 hours. I use the SSW_bbl for the wavecurrent interactions.the CPP options are as following:
#define SSW_BBL /* Sherwood et al. BBL closure */
#ifdef SSW_BBL
# define SSW_CALC_ZNOT /* Computing bottom roughness internally */
# undef SSW_LOGINT /* Logarithmic interpolation of (Ur,Vr) */
# undef SSW_CALC_UB /* Computing bottom orbital velocity internally */
# undef SSW_FORM_DRAG_COR /* Activate form drag coefficient */
# undef SSW_ZOBIO /* Biogenic bedform roughness from ripples */
# undef SSW_ZOBL /* Bedload roughness for ripples */
# undef SSW_ZORIP /* Bedform roughness from ripples */#endif
and in my log file, they exchange data normally, for example:
== SWAN grid 1 sent wave data to ROMS grid 1
** ROMS grid 1 recv data from SWAN grid 1
SWANtoROMS Min/Max DISBOT (Wm2): 0.000000E+00 1.156542E05
SWANtoROMS Min/Max DISSURF (Wm2): 0.000000E+00 0.000000E+00
SWANtoROMS Min/Max DISWCAP (Wm2): 0.000000E+00 8.076467E05
SWANtoROMS Min/Max HSIGN (m): 0.000000E+00 6.775821E01
SWANtoROMS Min/Max RTP (s): 0.000000E+00 3.343699E+00
SWANtoROMS Min/Max TMBOT (s): 0.000000E+00 8.335983E+00
SWANtoROMS Min/Max UBOT (ms1): 0.000000E+00 5.487803E02
SWANtoROMS Min/Max DIR (deg): 9.044095E04 6.282655E+00
SWANtoROMS Min/Max WLEN (m): 1.000000E+00 1.068273E+01
** ROMS grid 1 sent data to SWAN grid 1
== SWAN grid 1 recv data from ROMS grid 1
ROMStoSWAN Min/Max DEPTH (m): 5.000000E+00 1.321269E+02
ROMStoSWAN Min/Max WLEV (m): 8.748971E01 1.111308E+00
ROMStoSWAN Min/Max VELX (ms1): 8.722052E01 8.795478E01
ROMStoSWAN Min/Max VELY (ms1): 1.390613E+00 8.895970E01
ROMStoSWAN Min/Max ZO (m): 5.000000E02 5.000000E02
But when I check the wave induced shear stress(bustrw,bvstrw), values are really small, So I made a test that turn on the CPP options
SSW_CALC_UB, as following , and I changed nothing about others. And the result is more reasonable.
#define SSW_BBL /* Sherwood et al. BBL closure */
#ifdef SSW_BBL
# define SSW_CALC_ZNOT /* Computing bottom roughness internally */
# undef SSW_LOGINT /* Logarithmic interpolation of (Ur,Vr) */
# define SSW_CALC_UB /* Computing bottom orbital velocity internally */
# undef SSW_FORM_DRAG_COR /* Activate form drag coefficient */
# undef SSW_ZOBIO /* Biogenic bedform roughness from ripples */
# undef SSW_ZOBL /* Bedload roughness for ripples */
# undef SSW_ZORIP /* Bedform roughness from ripples */#endif
So I want to know if should I turn on the SSW_CALC_UB option when I couple SWAN and ROMS.
A problem about coupled SWAN and ROMS model
Moderators: arango, robertson, rsignell

 Posts: 45
 Joined: Tue Mar 01, 2016 1:38 am
 Location: the Ocean University of China
Re: A problem about coupled SWAN and ROMS model
typically when you couple roms with swan, then you should use the bottom orbital velocity from swan. so we typically only use ssw_calc_znot with roms+swan.
If you added ssw_calc_ub, then Ubot is being computed in sswbbl as:
! Set bed wave orbital velocity and excursion amplitude. Use data
! from wave models (SWAN) or use Dean and Dalrymple (1991) 6thdegree
! polynomial to approximate wave number on shoaling water.
Fwave_bot(i,j)=twopi/MAX(Pwave_bot(i,j),0.05_r8)
#ifdef SSW_CALC_UB
Kdh=h(i,j)*Fwave_bot(i,j)**2/g
Kbh2=Kdh*Kdh+ &
& Kdh/(1.0_r8+Kdh*(K1+Kdh*(K2+Kdh*(K3+Kdh*(K4+ &
& Kdh*(K5+K6*Kdh))))))
Kbh=SQRT(Kbh2)
Ab(i,j)=0.5_r8*Hwave(i,j)/SINH(Kbh)+eps
Ub(i,j)=Fwave_bot(i,j)*Ab(i,j)+eps
#else
Ub(i,j)=MAX(Uwave_rms(i,j),0.0_r8)+eps
Ab(i,j)=Ub(i,j)/Fwave_bot(i,j)+eps
#endif
You might want to look at the bottom roughness values that are being sent to swan. See if there is a grain size effect. that might have some effect on the Ub.
j
If you added ssw_calc_ub, then Ubot is being computed in sswbbl as:
! Set bed wave orbital velocity and excursion amplitude. Use data
! from wave models (SWAN) or use Dean and Dalrymple (1991) 6thdegree
! polynomial to approximate wave number on shoaling water.
Fwave_bot(i,j)=twopi/MAX(Pwave_bot(i,j),0.05_r8)
#ifdef SSW_CALC_UB
Kdh=h(i,j)*Fwave_bot(i,j)**2/g
Kbh2=Kdh*Kdh+ &
& Kdh/(1.0_r8+Kdh*(K1+Kdh*(K2+Kdh*(K3+Kdh*(K4+ &
& Kdh*(K5+K6*Kdh))))))
Kbh=SQRT(Kbh2)
Ab(i,j)=0.5_r8*Hwave(i,j)/SINH(Kbh)+eps
Ub(i,j)=Fwave_bot(i,j)*Ab(i,j)+eps
#else
Ub(i,j)=MAX(Uwave_rms(i,j),0.0_r8)+eps
Ab(i,j)=Ub(i,j)/Fwave_bot(i,j)+eps
#endif
You might want to look at the bottom roughness values that are being sent to swan. See if there is a grain size effect. that might have some effect on the Ub.
j

 Posts: 45
 Joined: Tue Mar 01, 2016 1:38 am
 Location: the Ocean University of China
Re: A problem about coupled SWAN and ROMS model
Thanks John,
That is to say, If I want ROMS one way coupled with swan (make the swan output hwave, tmbot wdir into a wave forcing file) , the CPP option define SSW_CALC_UB can be activated. And if I want to ROMS two way coupled with SWAN, There is no need to activate it. Right?
Yes, I want to look at the bottom roughness values that are being sent to swan, but why in my log file, the ZO that roms sent to swan is always be 0.05m? and the attachment is my log file.
ROMStoSWAN Min/Max ZO (m): 5.000000E02 5.000000E02
That is to say, If I want ROMS one way coupled with swan (make the swan output hwave, tmbot wdir into a wave forcing file) , the CPP option define SSW_CALC_UB can be activated. And if I want to ROMS two way coupled with SWAN, There is no need to activate it. Right?
Yes, I want to look at the bottom roughness values that are being sent to swan, but why in my log file, the ZO that roms sent to swan is always be 0.05m? and the attachment is my log file.
ROMStoSWAN Min/Max ZO (m): 5.000000E02 5.000000E02
 Attachments

 log.txt
 (2.66 MiB) Downloaded 96 times
Re: A problem about coupled SWAN and ROMS model
For both choices (uncoupled or coupled) you still have 2 choices: you can save Ub from swan and force roms with that or use the SSW_CALC_UB option and have the code compute Ub.
For the bottom roughness, we have a minimum value of 0.05 set in MAster/mct_roms_swan.h. you can edit that file and change it if you want.
For the bottom roughness, we have a minimum value of 0.05 set in MAster/mct_roms_swan.h. you can edit that file and change it if you want.

 Posts: 45
 Joined: Tue Mar 01, 2016 1:38 am
 Location: the Ocean University of China
Re: A problem about coupled SWAN and ROMS model
Ok, I get the point.Thanks John