Custom Query (964 matches)

The is a problem when compiling ini_hmixcoef.F under _OPENMP. The following include file is missing:

#include "set_bounds.h"

Many thanks to Jose Alves for ​reporting this problem.

The observation impact operator for W4DPSAS and W4DVAR was further developed to additionally compute the contributions due to initial conditions, surface forcing, and boundary conditions. A new C-preprocessing option, OBS_IMPACT_SPLIT, is introduced to achieve this. It requires additional integrations of the tangent linear model. This option needs to be used at the same time with OBS_IMPACT and W4DPSAS_SENSITIVITY or W4DVAR_SENSITIVITY.

If OBS_IMPACT_SPLIT is defined as well, then the contribution to the observation sensitivities for the increments in the initial conditions, surface forcing, and boundary conditions are all computed separately. This is a very nice feature and is very revealing.

Many thanks to Andy Moore for coding and testing this new option.

I corrected few bugs:

• Added to get_state.F and checkvars.F logic to process additional fields for perfect restart. Many thanks to Greg Gerbi for reporting this problem. We need the following conditional instead:

        IF (((model.eq.0).or.(model.eq.iNLM)).and.(nrrec(ng).ne.0)) THEN
  

• Modified step2d_LF_AM3.h and its TLM, RPM, and ADM versions to include a different logic for point mass sources. Eliminated the need for using local array Dnew and used zeta+h instead. There is still an adjoint bug in this routine when UV_PSOURCE is activated. We are still looking for it.

        DO is=1,Nsrc
          i=Isrc(is)
          j=Jsrc(is)
          IF (((IstrR.le.i).and.(i.le.IendR)).and.                        &
       &      ((JstrR.le.j).and.(j.le.JendR))) THEN
            IF (INT(Dsrc(is)).eq.0) THEN
              cff=1.0_r8/(on_u(i,j)*                                      &
       &                  0.5_r8*(zeta(i-1,j,knew)+h(i-1,j)+              &
       &                          zeta(i  ,j,knew)+h(i  ,j)))
              ubar(i,j,knew)=Qbar(is)*cff
  #  ifdef SOLVE3D
              DU_avg1(i,j)=Qbar(is)
  #  endif
            ELSE
              cff=1.0_r8/(om_v(i,j)*                                      &
       &                  0.5_r8*(zeta(i,j-1,knew)+h(i,j-1)+              &
       &                          zeta(i,j  ,knew)+h(i,j  )))
              vbar(i,j,knew)=Qbar(is)*cff
  #  ifdef SOLVE3D
              DV_avg1(i,j)=Qbar(is)
  #  endif
            END IF
          END IF
        END DO
  

• Corrected a parallel bug in ana_grid.h when computing bathymetry h. Many thanks to Aurelien Ponte for ​reporting this problem. This was a clear shared- and distribute-memory bug.

        DO j=JstrR,JendR
          DO i=IstrR,IendR
            val1=(xr(i,j)-0.5_r8*Xsize)/40000.0_r8
            val2=(yr(i,j)-0.5_r8*Ysize)/40000.0_r8
            h(i,j)=depth-4500.0_r8*EXP(-(val1*val1+val2*val2))
          END DO
        END DO
  

• Added additional optional argument Lupdate to routine netcdf_close to specify if we want to update the bio_file global attribute. This attribute can only be updated in those NetCDF opened for writting.