SUBROUTINE inflow_turbulence !------------------------------------------------------------------------------! ! Current revisions: ! ----------------- ! ! ! Former revisions: ! ----------------- ! $Id: inflow_turbulence.f90 623 2010-12-10 08:52:17Z heinze $ ! ! 622 2010-12-10 08:08:13Z raasch ! optional barriers included in order to speed up collective operations ! ! 222 2009-01-12 16:04:16Z letzel ! Bugfix for nonparallel execution ! ! Initial version (2008/03/07) ! ! Description: ! ------------ ! Imposing turbulence at the respective inflow using the turbulence ! recycling method of Kataoka and Mizuno (2002). !------------------------------------------------------------------------------! USE arrays_3d USE control_parameters USE cpulog USE grid_variables USE indices USE interfaces USE pegrid IMPLICIT NONE INTEGER :: i, imax, j, k, ngp_ifd, ngp_pr REAL, DIMENSION(1:2) :: volume_flow_l, volume_flow_offset REAL, DIMENSION(nzb:nzt+1,5) :: avpr, avpr_l REAL, DIMENSION(nzb:nzt+1,nys-1:nyn+1,5) :: inflow_dist CALL cpu_log( log_point(40), 'inflow_turbulence', 'start' ) ! !-- Carry out horizontal averaging in the recycling plane avpr_l = 0.0 ngp_pr = ( nzt - nzb + 2 ) * 5 ngp_ifd = ngp_pr * ( nyn - nys + 3 ) ! !-- First, local averaging within the recycling domain IF ( recycling_plane >= nxl ) THEN imax = MIN( nxr, recycling_plane ) DO i = nxl, imax DO j = nys, nyn DO k = nzb, nzt+1 avpr_l(k,1) = avpr_l(k,1) + u(k,j,i) avpr_l(k,2) = avpr_l(k,2) + v(k,j,i) avpr_l(k,3) = avpr_l(k,3) + w(k,j,i) avpr_l(k,4) = avpr_l(k,4) + pt(k,j,i) avpr_l(k,5) = avpr_l(k,5) + e(k,j,i) ENDDO ENDDO ENDDO ENDIF ! WRITE (9,*) '*** averaged profiles avpr_l' ! DO k = nzb, nzt+1 ! WRITE (9,'(F5.1,1X,F5.1,1X,F5.1,1X,F6.1,1X,F7.2)') avpr_l(k,1),avpr_l(k,2),avpr_l(k,3),avpr_l(k,4),avpr_l(k,5) ! ENDDO ! WRITE (9,*) ' ' #if defined( __parallel ) ! !-- Now, averaging over all PEs IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) CALL MPI_ALLREDUCE( avpr_l(nzb,1), avpr(nzb,1), ngp_pr, MPI_REAL, MPI_SUM, & comm2d, ierr ) #else avpr = avpr_l #endif avpr = avpr / ( ( ny + 1 ) * ( recycling_plane + 1 ) ) ! WRITE (9,*) '*** averaged profiles' ! DO k = nzb, nzt+1 ! WRITE (9,'(F5.1,1X,F5.1,1X,F5.1,1X,F6.1,1X,F7.2)') avpr(k,1),avpr(k,2),avpr(k,3),avpr(k,4),avpr(k,5) ! ENDDO ! WRITE (9,*) ' ' ! !-- Calculate the disturbances at the recycling plane i = recycling_plane #if defined( __parallel ) IF ( myidx == id_recycling ) THEN DO j = nys-1, nyn+1 DO k = nzb, nzt+1 inflow_dist(k,j,1) = u(k,j,i+1) - avpr(k,1) inflow_dist(k,j,2) = v(k,j,i) - avpr(k,2) inflow_dist(k,j,3) = w(k,j,i) - avpr(k,3) inflow_dist(k,j,4) = pt(k,j,i) - avpr(k,4) inflow_dist(k,j,5) = e(k,j,i) - avpr(k,5) ENDDO ENDDO ENDIF #else DO j = nys-1, nyn+1 DO k = nzb, nzt+1 inflow_dist(k,j,1) = u(k,j,i+1) - avpr(k,1) inflow_dist(k,j,2) = v(k,j,i) - avpr(k,2) inflow_dist(k,j,3) = w(k,j,i) - avpr(k,3) inflow_dist(k,j,4) = pt(k,j,i) - avpr(k,4) inflow_dist(k,j,5) = e(k,j,i) - avpr(k,5) ENDDO ENDDO #endif ! !-- For parallel runs, send the disturbances to the respective inflow PE #if defined( __parallel ) IF ( myidx == id_recycling .AND. myidx /= id_inflow ) THEN CALL MPI_SEND( inflow_dist(nzb,nys-1,1), ngp_ifd, MPI_REAL, & id_inflow, 1, comm1dx, ierr ) ELSEIF ( myidx /= id_recycling .AND. myidx == id_inflow ) THEN inflow_dist = 0.0 CALL MPI_RECV( inflow_dist(nzb,nys-1,1), ngp_ifd, MPI_REAL, & id_recycling, 1, comm1dx, status, ierr ) ENDIF #endif ! !-- Add the disturbance at the inflow IF ( nxl == 0 ) THEN DO j = nys-1, nyn+1 DO k = nzb, nzt+1 ! WRITE (9,*) 'j=',j,' k=',k ! WRITE (9,*) 'mean_u = ', mean_inflow_profiles(k,1), ' dist_u = ',& ! inflow_dist(k,j,1) ! WRITE (9,*) 'mean_v = ', mean_inflow_profiles(k,2), ' dist_v = ',& ! inflow_dist(k,j,2) ! WRITE (9,*) 'mean_w = 0.0', ' dist_w = ',& ! inflow_dist(k,j,3) ! WRITE (9,*) 'mean_pt = ', mean_inflow_profiles(k,4), ' dist_pt = ',& ! inflow_dist(k,j,4) ! WRITE (9,*) 'mean_e = ', mean_inflow_profiles(k,5), ' dist_e = ',& ! inflow_dist(k,j,5) u(k,j,0) = mean_inflow_profiles(k,1) + & inflow_dist(k,j,1) * inflow_damping_factor(k) v(k,j,-1) = mean_inflow_profiles(k,2) + & inflow_dist(k,j,2) * inflow_damping_factor(k) w(k,j,-1) = inflow_dist(k,j,3) * inflow_damping_factor(k) pt(k,j,-1) = mean_inflow_profiles(k,4) + & inflow_dist(k,j,4) * inflow_damping_factor(k) e(k,j,-1) = mean_inflow_profiles(k,5) + & inflow_dist(k,j,5) * inflow_damping_factor(k) e(k,j,-1) = MAX( e(k,j,-1), 0.0 ) ENDDO ENDDO ENDIF ! !-- Conserve the volume flow at the inflow in order to avoid generation of !-- waves in the stable layer ! IF ( conserve_volume_flow .AND. inflow_l ) THEN ! volume_flow(1) = 0.0 ! volume_flow_l(1) = 0.0 ! i = 0 ! DO j = nys, nyn ! !-- Sum up the volume flow through the south/north boundary ! DO k = nzb_2d(j,i) + 1, nzt ! volume_flow_l(1) = volume_flow_l(1) + u(k,j,i) * dzu(k) ! ENDDO ! ENDDO #if defined( __parallel ) ! IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) ! CALL MPI_ALLREDUCE( volume_flow_l(1), volume_flow(1), 1, MPI_REAL, & ! MPI_SUM, comm1dy, ierr ) #else ! volume_flow = volume_flow_l #endif ! volume_flow_offset(1) = ( volume_flow_initial(1) - volume_flow(1) ) & ! / volume_flow_area(1) ! DO j = nys-1, nyn+1 ! DO k = nzb_v_inner(j,i) + 1, nzt ! u(k,j,i) = u(k,j,i) + volume_flow_offset(1) ! ENDDO ! ENDDO ! ENDIF CALL cpu_log( log_point(40), 'inflow_turbulence', 'stop' ) END SUBROUTINE inflow_turbulence