New URL for NEMO forge! http://forge.nemo-ocean.eu

Since March 2022 along with NEMO 4.2 release, the code development moved to a self-hosted GitLab.
This present forge is now archived and remained online for history.

dynnxt.F90 in trunk/NEMOGCM/NEMO/OPA_SRC/DYN – NEMO

Context Navigation

source: trunk/NEMOGCM/NEMO/OPA_SRC/DYN/dynnxt.F90 @ 7698

Last change on this file since 7698 was 7698, checked in by mocavero, 7 years ago
update trunk with OpenMP parallelization
Property svn:keywords set to `Id`
File size: 22.2 KB

Rev	Line
[3]	1	MODULE dynnxt
[1502]	2	!!=========================================================================
[3]	3	!! * MODULE dynnxt *
	4	!! Ocean dynamics: time stepping
[1502]	5	!!=========================================================================
[1438]	6	!! History : OPA ! 1987-02 (P. Andrich, D. L Hostis) Original code
	7	!! ! 1990-10 (C. Levy, G. Madec)
	8	!! 7.0 ! 1993-03 (M. Guyon) symetrical conditions
	9	!! 8.0 ! 1997-02 (G. Madec & M. Imbard) opa, release 8.0
	10	!! 8.2 ! 1997-04 (A. Weaver) Euler forward step
	11	!! - ! 1997-06 (G. Madec) lateral boudary cond., lbc routine
	12	!! NEMO 1.0 ! 2002-08 (G. Madec) F90: Free form and module
	13	!! - ! 2002-10 (C. Talandier, A-M. Treguier) Open boundary cond.
	14	!! 2.0 ! 2005-11 (V. Garnier) Surface pressure gradient organization
	15	!! 2.3 ! 2007-07 (D. Storkey) Calls to BDY routines.
[1502]	16	!! 3.2 ! 2009-06 (G. Madec, R.Benshila) re-introduce the vvl option
[2528]	17	!! 3.3 ! 2010-09 (D. Storkey, E.O'Dea) Bug fix for BDY module
[2723]	18	!! 3.3 ! 2011-03 (P. Oddo) Bug fix for time-splitting+(BDY-OBC) and not VVL
[4292]	19	!! 3.5 ! 2013-07 (J. Chanut) Compliant with time splitting changes
[6140]	20	!! 3.6 ! 2014-04 (G. Madec) add the diagnostic of the time filter trends
[5930]	21	!! 3.7 ! 2015-11 (J. Chanut) Free surface simplification
[1502]	22	!!-------------------------------------------------------------------------
[1438]	23
[1502]	24	!!-------------------------------------------------------------------------
[6140]	25	!! dyn_nxt : obtain the next (after) horizontal velocity
[1502]	26	!!-------------------------------------------------------------------------
[6140]	27	USE oce ! ocean dynamics and tracers
	28	USE dom_oce ! ocean space and time domain
	29	USE sbc_oce ! Surface boundary condition: ocean fields
	30	USE phycst ! physical constants
	31	USE dynadv ! dynamics: vector invariant versus flux form
	32	USE dynspg_ts ! surface pressure gradient: split-explicit scheme
	33	USE domvvl ! variable volume
[7646]	34	USE bdy_oce , ONLY: ln_bdy
[6140]	35	USE bdydta ! ocean open boundary conditions
	36	USE bdydyn ! ocean open boundary conditions
	37	USE bdyvol ! ocean open boundary condition (bdy_vol routines)
	38	USE trd_oce ! trends: ocean variables
	39	USE trddyn ! trend manager: dynamics
	40	USE trdken ! trend manager: kinetic energy
[4990]	41	!
[6140]	42	USE in_out_manager ! I/O manager
	43	USE iom ! I/O manager library
	44	USE lbclnk ! lateral boundary condition (or mpp link)
	45	USE lib_mpp ! MPP library
	46	USE wrk_nemo ! Memory Allocation
	47	USE prtctl ! Print control
	48	USE timing ! Timing
[2528]	49	#if defined key_agrif
	50	USE agrif_opa_interp
	51	#endif
[3]	52
	53	IMPLICIT NONE
	54	PRIVATE
	55
[1438]	56	PUBLIC dyn_nxt ! routine called by step.F90
	57
[2715]	58	!!----------------------------------------------------------------------
[2528]	59	!! NEMO/OPA 3.3 , NEMO Consortium (2010)
[1438]	60	!! $Id$
[2715]	61	!! Software governed by the CeCILL licence (NEMOGCM/NEMO_CeCILL.txt)
	62	!!----------------------------------------------------------------------
[3]	63	CONTAINS
	64
	65	SUBROUTINE dyn_nxt ( kt )
	66	!!----------------------------------------------------------------------
	67	!! * ROUTINE dyn_nxt *
	68	!!
[5930]	69	!! ** Purpose : Finalize after horizontal velocity. Apply the boundary
	70	!! condition on the after velocity, achieve the time stepping
[1502]	71	!! by applying the Asselin filter on now fields and swapping
	72	!! the fields.
[3]	73	!!
[5930]	74	!! ** Method : * Ensure after velocities transport matches time splitting
	75	!! estimate (ln_dynspg_ts=T)
[3]	76	!!
[1502]	77	!! * Apply lateral boundary conditions on after velocity
	78	!! at the local domain boundaries through lbc_lnk call,
[7646]	79	!! at the one-way open boundaries (ln_bdy=T),
[4990]	80	!! at the AGRIF zoom boundaries (lk_agrif=T)
[3]	81	!!
[1502]	82	!! * Apply the time filter applied and swap of the dynamics
	83	!! arrays to start the next time step:
	84	!! (ub,vb) = (un,vn) + atfp [ (ub,vb) + (ua,va) - 2 (un,vn) ]
	85	!! (un,vn) = (ua,va).
[6140]	86	!! Note that with flux form advection and non linear free surface,
	87	!! the time filter is applied on thickness weighted velocity.
	88	!! As a result, dyn_nxt MUST be called after tra_nxt.
[1502]	89	!!
	90	!! ** Action : ub,vb filtered before horizontal velocity of next time-step
	91	!! un,vn now horizontal velocity of next time-step
[3]	92	!!----------------------------------------------------------------------
	93	INTEGER, INTENT( in ) :: kt ! ocean time-step index
[2715]	94	!
[3]	95	INTEGER :: ji, jj, jk ! dummy loop indices
[6140]	96	INTEGER :: ikt ! local integers
	97	REAL(wp) :: zue3a, zue3n, zue3b, zuf, zcoef ! local scalars
[4990]	98	REAL(wp) :: zve3a, zve3n, zve3b, zvf, z1_2dt ! - -
	99	REAL(wp), POINTER, DIMENSION(:,:) :: zue, zve
	100	REAL(wp), POINTER, DIMENSION(:,:,:) :: ze3u_f, ze3v_f, zua, zva
[1502]	101	!!----------------------------------------------------------------------
[3294]	102	!
[4990]	103	IF( nn_timing == 1 ) CALL timing_start('dyn_nxt')
[3294]	104	!
[6140]	105	IF( ln_dynspg_ts ) CALL wrk_alloc( jpi,jpj, zue, zve)
	106	IF( l_trddyn ) CALL wrk_alloc( jpi,jpj,jpk, zua, zva)
[3294]	107	!
[3]	108	IF( kt == nit000 ) THEN
	109	IF(lwp) WRITE(numout,*)
	110	IF(lwp) WRITE(numout,*) 'dyn_nxt : time stepping'
	111	IF(lwp) WRITE(numout,*) '~~~~~~~'
	112	ENDIF
	113
[5930]	114	IF ( ln_dynspg_ts ) THEN
	115	! Ensure below that barotropic velocities match time splitting estimate
	116	! Compute actual transport and replace it with ts estimate at "after" time step
[7698]	117	!$OMP PARALLEL
	118	!$OMP DO schedule(static) private(jj, ji)
	119	DO jj = 1, jpj
	120	DO ji = 1, jpi
	121	zue(ji,jj) = e3u_a(ji,jj,1) * ua(ji,jj,1) * umask(ji,jj,1)
	122	zve(ji,jj) = e3v_a(ji,jj,1) * va(ji,jj,1) * vmask(ji,jj,1)
	123	END DO
	124	END DO
[5930]	125	DO jk = 2, jpkm1
[7698]	126	!$OMP DO schedule(static) private(jj,ji)
	127	DO jj = 1, jpj
	128	DO ji = 1, jpi
	129	zue(ji,jj) = zue(ji,jj) + e3u_a(ji,jj,jk) * ua(ji,jj,jk) * umask(ji,jj,jk)
	130	zve(ji,jj) = zve(ji,jj) + e3v_a(ji,jj,jk) * va(ji,jj,jk) * vmask(ji,jj,jk)
	131	END DO
	132	END DO
[1502]	133	END DO
[7698]	134	!$OMP DO schedule(static) private(jk,jj,ji)
[1502]	135	DO jk = 1, jpkm1
[7698]	136	DO jj = 1, jpj
	137	DO ji = 1, jpi
	138	ua(ji,jj,jk) = ( ua(ji,jj,jk) - zue(ji,jj) * r1_hu_a(ji,jj) + ua_b(ji,jj) ) * umask(ji,jj,jk)
	139	va(ji,jj,jk) = ( va(ji,jj,jk) - zve(ji,jj) * r1_hv_a(ji,jj) + va_b(ji,jj) ) * vmask(ji,jj,jk)
	140	END DO
	141	END DO
[592]	142	END DO
[7698]	143	!$OMP END PARALLEL
[6140]	144	!
	145	IF( .NOT.ln_bt_fw ) THEN
[5930]	146	! Remove advective velocity from "now velocities"
	147	! prior to asselin filtering
	148	! In the forward case, this is done below after asselin filtering
	149	! so that asselin contribution is removed at the same time
[7698]	150	!$OMP PARALLEL DO schedule(static) private(jk,jj,ji)
[5930]	151	DO jk = 1, jpkm1
[7698]	152	DO jj = 1, jpj
	153	DO ji = 1, jpi
	154	un(ji,jj,jk) = ( un(ji,jj,jk) - un_adv(ji,jj) + un_b(ji,jj) )*umask(ji,jj,jk)
	155	vn(ji,jj,jk) = ( vn(ji,jj,jk) - vn_adv(ji,jj) + vn_b(ji,jj) )*vmask(ji,jj,jk)
	156	END DO
	157	END DO
	158	END DO
	159
[5930]	160	ENDIF
[4292]	161	ENDIF
	162
[1502]	163	! Update after velocity on domain lateral boundaries
	164	! --------------------------------------------------
[5930]	165	# if defined key_agrif
	166	CALL Agrif_dyn( kt ) !* AGRIF zoom boundaries
	167	# endif
	168	!
[1502]	169	CALL lbc_lnk( ua, 'U', -1. ) !* local domain boundaries
	170	CALL lbc_lnk( va, 'V', -1. )
	171	!
	172	! !* BDY open boundaries
[7646]	173	IF( ln_bdy .AND. ln_dynspg_exp ) CALL bdy_dyn( kt )
	174	IF( ln_bdy .AND. ln_dynspg_ts ) CALL bdy_dyn( kt, dyn3d_only=.true. )
[3294]	175
	176	!!$ Do we need a call to bdy_vol here??
	177	!
[4990]	178	IF( l_trddyn ) THEN ! prepare the atf trend computation + some diagnostics
	179	z1_2dt = 1._wp / (2. * rdt) ! Euler or leap-frog time step
	180	IF( neuler == 0 .AND. kt == nit000 ) z1_2dt = 1._wp / rdt
	181	!
	182	! ! Kinetic energy and Conversion
	183	IF( ln_KE_trd ) CALL trd_dyn( ua, va, jpdyn_ken, kt )
	184	!
	185	IF( ln_dyn_trd ) THEN ! 3D output: total momentum trends
[7698]	186	!$OMP PARALLEL DO schedule(static) private(jk, jj, ji)
	187	DO jk = 1, jpk
	188	DO jj = 1, jpj
	189	DO ji = 1, jpi
	190	zua(ji,jj,jk) = ( ua(ji,jj,jk) - ub(ji,jj,jk) ) * z1_2dt
	191	zva(ji,jj,jk) = ( va(ji,jj,jk) - vb(ji,jj,jk) ) * z1_2dt
	192	END DO
	193	END DO
	194	END DO
[4990]	195	CALL iom_put( "utrd_tot", zua ) ! total momentum trends, except the asselin time filter
	196	CALL iom_put( "vtrd_tot", zva )
	197	ENDIF
	198	!
[7698]	199	!$OMP PARALLEL DO schedule(static) private(jk, jj, ji)
	200	DO jk = 1, jpk
	201	DO jj = 1, jpj
	202	DO ji = 1, jpi
	203	zua(ji,jj,jk) = un(ji,jj,jk) ! save the now velocity before the asselin filter
	204	zva(ji,jj,jk) = vn(ji,jj,jk) ! (caution: there will be a shift by 1 timestep in the
	205	! ! computation of the asselin filter trends)
	206	END DO
	207	END DO
	208	END DO
[4990]	209	ENDIF
	210
[1438]	211	! Time filter and swap of dynamics arrays
	212	! ------------------------------------------
[1502]	213	IF( neuler == 0 .AND. kt == nit000 ) THEN !* Euler at first time-step: only swap
[7698]	214	!$OMP PARALLEL
	215	!$OMP DO schedule(static) private(jk,jj,ji)
[1502]	216	DO jk = 1, jpkm1
[7698]	217	DO jj = 1, jpj
	218	DO ji = 1, jpi
	219	un(ji,jj,jk) = ua(ji,jj,jk) ! un <-- ua
	220	vn(ji,jj,jk) = va(ji,jj,jk)
	221	END DO
	222	END DO
[1438]	223	END DO
[7698]	224	!$OMP END DO NOWAIT
[6140]	225	IF(.NOT.ln_linssh ) THEN
[7698]	226	!$OMP DO schedule(static) private(jk,jj,ji)
[4292]	227	DO jk = 1, jpkm1
[7698]	228	DO jj = 1, jpj
	229	DO ji = 1, jpi
	230	e3t_b(ji,jj,jk) = e3t_n(ji,jj,jk)
	231	e3u_b(ji,jj,jk) = e3u_n(ji,jj,jk)
	232	e3v_b(ji,jj,jk) = e3v_n(ji,jj,jk)
	233	END DO
	234	END DO
[6140]	235	END DO
[4292]	236	ENDIF
[7698]	237	!$OMP END PARALLEL
[1502]	238	ELSE !* Leap-Frog : Asselin filter and swap
[2528]	239	! ! =============!
[6140]	240	IF( ln_linssh ) THEN ! Fixed volume !
[2528]	241	! ! =============!
[7698]	242	!$OMP PARALLEL DO schedule(static) private(jk, jj, ji, zuf, zvf)
[1502]	243	DO jk = 1, jpkm1
[592]	244	DO jj = 1, jpj
[1502]	245	DO ji = 1, jpi
[4990]	246	zuf = un(ji,jj,jk) + atfp * ( ub(ji,jj,jk) - 2._wp * un(ji,jj,jk) + ua(ji,jj,jk) )
	247	zvf = vn(ji,jj,jk) + atfp * ( vb(ji,jj,jk) - 2._wp * vn(ji,jj,jk) + va(ji,jj,jk) )
[1502]	248	!
	249	ub(ji,jj,jk) = zuf ! ub <-- filtered velocity
	250	vb(ji,jj,jk) = zvf
	251	un(ji,jj,jk) = ua(ji,jj,jk) ! un <-- ua
	252	vn(ji,jj,jk) = va(ji,jj,jk)
	253	END DO
	254	END DO
	255	END DO
[2528]	256	! ! ================!
	257	ELSE ! Variable volume !
	258	! ! ================!
[4292]	259	! Before scale factor at t-points
	260	! (used as a now filtered scale factor until the swap)
	261	! ----------------------------------------------------
[6140]	262	IF( ln_dynspg_ts .AND. ln_bt_fw ) THEN ! No asselin filtering on thicknesses if forward time splitting
[7698]	263	!$OMP PARALLEL DO schedule(static) private(jj,ji)
	264	DO jj = 1, jpj
	265	DO ji = 1, jpi
	266	e3t_b(ji,jj,1:jpkm1) = e3t_n(ji,jj,1:jpkm1)
	267	END DO
	268	END DO
[4292]	269	ELSE
[7698]	270	!$OMP PARALLEL DO schedule(static) private(jk,jj,ji)
[6140]	271	DO jk = 1, jpkm1
[7698]	272	DO jj = 1, jpj
	273	DO ji = 1, jpi
	274	e3t_b(ji,jj,jk) = e3t_n(ji,jj,jk) + atfp * ( e3t_b(ji,jj,jk) - 2._wp * e3t_n(ji,jj,jk) + e3t_a(ji,jj,jk) )
	275	END DO
	276	END DO
[6140]	277	END DO
[4292]	278	! Add volume filter correction: compatibility with tracer advection scheme
	279	! => time filter + conservation correction (only at the first level)
[6140]	280	zcoef = atfp * rdt * r1_rau0
	281	IF ( .NOT. ln_isf ) THEN ! if no ice shelf melting
[7698]	282	!$OMP PARALLEL DO schedule(static) private(jj,ji)
	283	DO jj = 1, jpj
	284	DO ji = 1, jpi
	285	e3t_b(ji,jj,1) = e3t_b(ji,jj,1) - zcoef * ( emp_b(ji,jj) - emp(ji,jj) &
	286	& - rnf_b(ji,jj) + rnf(ji,jj) ) * tmask(ji,jj,1)
	287	END DO
	288	END DO
[5643]	289	ELSE ! if ice shelf melting
[7698]	290	!$OMP PARALLEL DO schedule(static) private(jj,ji,ikt)
[6140]	291	DO jj = 1, jpj
	292	DO ji = 1, jpi
	293	ikt = mikt(ji,jj)
	294	e3t_b(ji,jj,ikt) = e3t_b(ji,jj,ikt) - zcoef * ( emp_b (ji,jj) - emp (ji,jj) &
	295	& - rnf_b (ji,jj) + rnf (ji,jj) &
	296	& + fwfisf_b(ji,jj) - fwfisf(ji,jj) ) * tmask(ji,jj,ikt)
[5643]	297	END DO
	298	END DO
	299	END IF
[4292]	300	ENDIF
[2528]	301	!
[6140]	302	IF( ln_dynadv_vec ) THEN ! Asselin filter applied on velocity
	303	! Before filtered scale factor at (u/v)-points
	304	CALL dom_vvl_interpol( e3t_b(:,:,:), e3u_b(:,:,:), 'U' )
	305	CALL dom_vvl_interpol( e3t_b(:,:,:), e3v_b(:,:,:), 'V' )
[7698]	306	!$OMP PARALLEL DO schedule(static) private(jk, jj, ji, zuf, zvf)
[4292]	307	DO jk = 1, jpkm1
	308	DO jj = 1, jpj
[2528]	309	DO ji = 1, jpi
[4292]	310	zuf = un(ji,jj,jk) + atfp * ( ub(ji,jj,jk) - 2._wp * un(ji,jj,jk) + ua(ji,jj,jk) )
	311	zvf = vn(ji,jj,jk) + atfp * ( vb(ji,jj,jk) - 2._wp * vn(ji,jj,jk) + va(ji,jj,jk) )
[2528]	312	!
	313	ub(ji,jj,jk) = zuf ! ub <-- filtered velocity
	314	vb(ji,jj,jk) = zvf
	315	un(ji,jj,jk) = ua(ji,jj,jk) ! un <-- ua
	316	vn(ji,jj,jk) = va(ji,jj,jk)
	317	END DO
	318	END DO
	319	END DO
	320	!
[6140]	321	ELSE ! Asselin filter applied on thickness weighted velocity
	322	!
	323	CALL wrk_alloc( jpi,jpj,jpk, ze3u_f, ze3v_f )
	324	! Before filtered scale factor at (u/v)-points stored in ze3u_f, ze3v_f
	325	CALL dom_vvl_interpol( e3t_b(:,:,:), ze3u_f, 'U' )
	326	CALL dom_vvl_interpol( e3t_b(:,:,:), ze3v_f, 'V' )
[7698]	327	!$OMP PARALLEL
	328	!$OMP DO schedule(static) private(jk, jj, ji, zue3a, zve3a, zue3n, zve3n, zue3b, zve3b, zuf, zvf)
[4292]	329	DO jk = 1, jpkm1
	330	DO jj = 1, jpj
[4312]	331	DO ji = 1, jpi
[6140]	332	zue3a = e3u_a(ji,jj,jk) * ua(ji,jj,jk)
	333	zve3a = e3v_a(ji,jj,jk) * va(ji,jj,jk)
	334	zue3n = e3u_n(ji,jj,jk) * un(ji,jj,jk)
	335	zve3n = e3v_n(ji,jj,jk) * vn(ji,jj,jk)
	336	zue3b = e3u_b(ji,jj,jk) * ub(ji,jj,jk)
	337	zve3b = e3v_b(ji,jj,jk) * vb(ji,jj,jk)
[2528]	338	!
[3294]	339	zuf = ( zue3n + atfp * ( zue3b - 2._wp * zue3n + zue3a ) ) / ze3u_f(ji,jj,jk)
	340	zvf = ( zve3n + atfp * ( zve3b - 2._wp * zve3n + zve3a ) ) / ze3v_f(ji,jj,jk)
[2528]	341	!
[3294]	342	ub(ji,jj,jk) = zuf ! ub <-- filtered velocity
[2528]	343	vb(ji,jj,jk) = zvf
[3294]	344	un(ji,jj,jk) = ua(ji,jj,jk) ! un <-- ua
[2528]	345	vn(ji,jj,jk) = va(ji,jj,jk)
	346	END DO
	347	END DO
	348	END DO
[7698]	349	!$OMP DO schedule(static) private(jj, ji)
	350	DO jj = 1, jpj
	351	DO ji = 1, jpi
	352	e3u_b(ji,jj,1:jpkm1) = ze3u_f(ji,jj,1:jpkm1) ! e3u_b <-- filtered scale factor
	353	e3v_b(ji,jj,1:jpkm1) = ze3v_f(ji,jj,1:jpkm1)
	354	END DO
	355	END DO
	356	!$OMP END PARALLEL
[6140]	357	!
	358	CALL wrk_dealloc( jpi,jpj,jpk, ze3u_f, ze3v_f )
[2528]	359	ENDIF
	360	!
[3]	361	ENDIF
[2528]	362	!
[6140]	363	IF( ln_dynspg_ts .AND. ln_bt_fw ) THEN
[4312]	364	! Revert "before" velocities to time split estimate
	365	! Doing it here also means that asselin filter contribution is removed
[7698]	366	!$OMP PARALLEL
	367	!$OMP DO schedule(static) private(jj, ji)
	368	DO jj = 1, jpj
	369	DO ji = 1, jpi
	370	zue(ji,jj) = e3u_b(ji,jj,1) * ub(ji,jj,1) * umask(ji,jj,1)
	371	zve(ji,jj) = e3v_b(ji,jj,1) * vb(ji,jj,1) * vmask(ji,jj,1)
	372	END DO
	373	END DO
[4990]	374	DO jk = 2, jpkm1
[7698]	375	!$OMP DO schedule(static) private(jj, ji)
	376	DO jj = 1, jpj
	377	DO ji = 1, jpi
	378	zue(ji,jj) = zue(ji,jj) + e3u_b(ji,jj,jk) * ub(ji,jj,jk) * umask(ji,jj,jk)
	379	zve(ji,jj) = zve(ji,jj) + e3v_b(ji,jj,jk) * vb(ji,jj,jk) * vmask(ji,jj,jk)
	380	END DO
	381	END DO
[4370]	382	END DO
[7698]	383	!$OMP DO schedule(static) private(jk,jj,ji)
[4370]	384	DO jk = 1, jpkm1
[7698]	385	DO jj = 1, jpj
	386	DO ji = 1, jpi
	387	ub(ji,jj,jk) = ub(ji,jj,jk) - (zue(ji,jj) * r1_hu_n(ji,jj) - un_b(ji,jj)) * umask(ji,jj,jk)
	388	vb(ji,jj,jk) = vb(ji,jj,jk) - (zve(ji,jj) * r1_hv_n(ji,jj) - vn_b(ji,jj)) * vmask(ji,jj,jk)
	389	END DO
	390	END DO
[4292]	391	END DO
[7698]	392	!$OMP END PARALLEL
[4292]	393	ENDIF
	394	!
	395	ENDIF ! neuler =/0
[4354]	396	!
	397	! Set "now" and "before" barotropic velocities for next time step:
	398	! JC: Would be more clever to swap variables than to make a full vertical
	399	! integration
	400	!
[4370]	401	!
[6140]	402	IF(.NOT.ln_linssh ) THEN
[7698]	403	!$OMP PARALLEL
	404	!$OMP DO schedule(static) private(jj, ji)
	405	DO jj = 1, jpj
	406	DO ji = 1, jpi
	407	hu_b(ji,jj) = e3u_b(ji,jj,1) * umask(ji,jj,1)
	408	hv_b(ji,jj) = e3v_b(ji,jj,1) * vmask(ji,jj,1)
	409	END DO
	410	END DO
[6140]	411	DO jk = 2, jpkm1
[7698]	412	!$OMP DO schedule(static) private(jj, ji)
	413	DO jj = 1, jpj
	414	DO ji = 1, jpi
	415	hu_b(ji,jj) = hu_b(ji,jj) + e3u_b(ji,jj,jk) * umask(ji,jj,jk)
	416	hv_b(ji,jj) = hv_b(ji,jj) + e3v_b(ji,jj,jk) * vmask(ji,jj,jk)
	417	END DO
	418	END DO
[4354]	419	END DO
[7698]	420	!$OMP DO schedule(static) private(jj, ji)
	421	DO jj = 1, jpj
	422	DO ji = 1, jpi
	423	r1_hu_b(ji,jj) = ssumask(ji,jj) / ( hu_b(ji,jj) + 1._wp - ssumask(ji,jj) )
	424	r1_hv_b(ji,jj) = ssvmask(ji,jj) / ( hv_b(ji,jj) + 1._wp - ssvmask(ji,jj) )
	425	END DO
	426	END DO
	427	!$OMP END PARALLEL
[4354]	428	ENDIF
	429	!
[7698]	430	!$OMP PARALLEL
	431	!$OMP DO schedule(static) private(jj, ji)
	432	DO jj = 1, jpj
	433	DO ji = 1, jpi
	434	un_b(ji,jj) = e3u_a(ji,jj,1) * un(ji,jj,1) * umask(ji,jj,1)
	435	ub_b(ji,jj) = e3u_b(ji,jj,1) * ub(ji,jj,1) * umask(ji,jj,1)
	436	vn_b(ji,jj) = e3v_a(ji,jj,1) * vn(ji,jj,1) * vmask(ji,jj,1)
	437	vb_b(ji,jj) = e3v_b(ji,jj,1) * vb(ji,jj,1) * vmask(ji,jj,1)
	438	END DO
	439	END DO
[6140]	440	DO jk = 2, jpkm1
[7698]	441	!$OMP DO schedule(static) private(jj, ji)
	442	DO jj = 1, jpj
	443	DO ji = 1, jpi
	444	un_b(ji,jj) = un_b(ji,jj) + e3u_a(ji,jj,jk) * un(ji,jj,jk) * umask(ji,jj,jk)
	445	ub_b(ji,jj) = ub_b(ji,jj) + e3u_b(ji,jj,jk) * ub(ji,jj,jk) * umask(ji,jj,jk)
	446	vn_b(ji,jj) = vn_b(ji,jj) + e3v_a(ji,jj,jk) * vn(ji,jj,jk) * vmask(ji,jj,jk)
	447	vb_b(ji,jj) = vb_b(ji,jj) + e3v_b(ji,jj,jk) * vb(ji,jj,jk) * vmask(ji,jj,jk)
	448	END DO
	449	END DO
[4354]	450	END DO
[7698]	451	!$OMP DO schedule(static) private(jj, ji)
	452	DO jj = 1, jpj
	453	DO ji = 1, jpi
	454	un_b(ji,jj) = un_b(ji,jj) * r1_hu_a(ji,jj)
	455	vn_b(ji,jj) = vn_b(ji,jj) * r1_hv_a(ji,jj)
	456	ub_b(ji,jj) = ub_b(ji,jj) * r1_hu_b(ji,jj)
	457	vb_b(ji,jj) = vb_b(ji,jj) * r1_hv_b(ji,jj)
	458	END DO
	459	END DO
	460	!$OMP END PARALLEL
[4354]	461	!
[6140]	462	IF( .NOT.ln_dynspg_ts ) THEN ! output the barotropic currents
	463	CALL iom_put( "ubar", un_b(:,:) )
	464	CALL iom_put( "vbar", vn_b(:,:) )
	465	ENDIF
[4990]	466	IF( l_trddyn ) THEN ! 3D output: asselin filter trends on momentum
[7698]	467	!$OMP PARALLEL DO schedule(static) private(jk, jj, ji)
	468	DO jk = 1, jpkm1
	469	DO jj = 1, jpj
	470	DO ji = 1, jpi
	471	zua(ji,jj,jk) = ( ub(ji,jj,jk) - zua(ji,jj,jk) ) * z1_2dt
	472	zva(ji,jj,jk) = ( vb(ji,jj,jk) - zva(ji,jj,jk) ) * z1_2dt
	473	END DO
	474	END DO
	475	END DO
[4990]	476	CALL trd_dyn( zua, zva, jpdyn_atf, kt )
	477	ENDIF
	478	!
[1438]	479	IF(ln_ctl) CALL prt_ctl( tab3d_1=un, clinfo1=' nxt - Un: ', mask1=umask, &
	480	& tab3d_2=vn, clinfo2=' Vn: ' , mask2=vmask )
[6140]	481	!
	482	IF( ln_dynspg_ts ) CALL wrk_dealloc( jpi,jpj, zue, zve )
	483	IF( l_trddyn ) CALL wrk_dealloc( jpi,jpj,jpk, zua, zva )
[2715]	484	!
[3294]	485	IF( nn_timing == 1 ) CALL timing_stop('dyn_nxt')
	486	!
[3]	487	END SUBROUTINE dyn_nxt
	488
[1502]	489	!!=========================================================================
[3]	490	END MODULE dynnxt

Note: See TracBrowser for help on using the repository browser.

Download in other formats: