Seeking Help: The Ideal Models Blow Up

[tao]

Hello friends, I have created a new ideal model by imitating Upwelling. I have added wind stress and sponge layers to it. In this model, there is always a significant speed increase near the north-south boundaries, causing blowing up.
following is what I set in ana.smflux

Code: Select all

#elif defined SCSS
      windamp = 4.0E-02_r8 / rho0
      DO j=JstrT,JendT
        DO i=IstrP,IendT
          sustr(i,j) = windamp * COS(f(i,j) * time(ng))
# ifdef TL_IOMS
          tl_sustr(i,j) = sustr(i,j)
# endif
        END DO
      END DO
      DO j=JstrT,JendT
        IF ((j.le.INT(5.0_r8)).or.(j.gt.Mm(ng)-INT(5.0_r8))) THEN
          DO i=IstrP,IendT
            sustr(i,j) = 0.0_r8
          END DO
        END IF
      END DO
      DO j=JstrP,JendT
        DO i=IstrT,IendT
          svstr(i,j) = 0.0_r8
# ifdef TL_IOMS
          tl_svstr(i,j) = 0.0_r8
# endif
        END DO
      END DO

and following is I set in ana.sponge

Code: Select all

#if defined SCSS
      cff=10.0_r8                       
      width=5.0_r8                    
      factor(IminS:ImaxS,JminS:JmaxS)=1.0_r8
      DO j=JstrT,MIN(INT(width),JendT)
        DO i=IstrT,IendT
          factor(i,j)=1.0_r8+(cff-1.0_r8)*(width-REAL(j,r8))/width
        END DO
      END DO
      DO j=MAX(JstrT,Mm(ng)+1-INT(width)),JendT
        DO i=IstrT,IendT
          val=1.0_r8+(cff-1.0_r8)*REAL(j-(Mm(ng)+1-INT(width)),r8)/width
          factor(i,j)=MAX(factor(i,j), val)
        END DO
      END DO
# if defined UV_VIS2
      IF (LuvSponge(ng)) THEN
        DO i=IstrT,IendT
          DO j=JstrT,JendT
            MIXING(ng) % visc2_r(i,j)=ABS(factor(i,j))* &
     &                                MIXING(ng) % visc2_r(i,j)
          END DO
        END DO
        DO j=JstrP,JendT
          DO i=IstrP,IendT
            MIXING(ng) % visc2_p(i,j)=0.25_r8*ABS(factor(i-1,j-1)+      &
     &                                            factor(i  ,j-1)+      &
     &                                            factor(i-1,j  )+      &
     &                                            factor(i  ,j  ))* &
     &                                MIXING(ng) % visc2_p(i,j)
          END DO
        END DO
      END IF
# endif
# if defined TS_DIF2
      DO itrc=1,NT(ng)
        IF (LtracerSponge(itrc,ng)) THEN
          DO i=IstrT,IendT
            DO j=JstrT,JendT
              MIXING(ng) % diff2(i,j,itrc)=ABS(factor(i,j))* &
     &                                     MIXING(ng) % diff2(i,j,itrc)
            END DO
          END DO
        END IF
      END DO
# endif
# endif

the last ,this is the rst.nc

roms_rst.nc

(48.63 MiB) Downloaded 16 times

I used a translation software to assist me in completing the above text. If it has caused any inconvenience to you, I'm truly sorry.
Welcome any suggestion

[tao]

Hi friends, good news, I finally solved the blow up by setting temp as constant and use salt to change the rho. But now I have another problem: the rho did not change as I expect. The salt vary but rho is still constant. Which step went wrong? thanks any suggestion.