1 | subroutine condense_n2(klon,klev,nq,ptimestep, & |
---|
2 | pcapcal,pplay,pplev,ptsrf,pt, & |
---|
3 | pphi,pdt,pdu,pdv,pdtsrf,pu,pv,pq,pdq, & |
---|
4 | picen2,psolaralb,pemisurf, & |
---|
5 | pdtc,pdtsrfc,pdpsrf,pduc,pdvc, & |
---|
6 | pdqc,pdicen2) |
---|
7 | |
---|
8 | use radinc_h, only : naerkind |
---|
9 | use comgeomfi_h |
---|
10 | use comcstfi_mod, only: g, r, cpp, pi |
---|
11 | USE surfdat_h, only: phisfi,kp,p00 |
---|
12 | USE tracer_h, only: noms, igcm_n2, lw_n2 |
---|
13 | USE callkeys_mod, only: fast,ch4lag,latlag,nbsub,no_n2frost,tsurfmax,kmixmin,source_haze,vmrlag |
---|
14 | USE comvert_mod, ONLY: ap,bp |
---|
15 | use geometry_mod, only: latitude |
---|
16 | |
---|
17 | |
---|
18 | implicit none |
---|
19 | |
---|
20 | !================================================================== |
---|
21 | ! Purpose |
---|
22 | ! ------- |
---|
23 | ! Condense and/or sublime N2 ice on the ground and in the |
---|
24 | ! atmosphere, and sediment the ice. |
---|
25 | ! |
---|
26 | ! Inputs |
---|
27 | ! ------ |
---|
28 | ! klon Number of vertical columns |
---|
29 | ! klev Number of layers |
---|
30 | ! pplay(klon,klev) Pressure layers |
---|
31 | ! pplev(klon,klev+1) Pressure levels |
---|
32 | ! pt(klon,klev) Temperature (in K) |
---|
33 | ! ptsrf(klon) Surface temperature |
---|
34 | ! |
---|
35 | ! pdt(klon,klev) Time derivative before condensation/sublimation of pt |
---|
36 | ! pdtsrf(klon) Time derivative before condensation/sublimation of ptsrf |
---|
37 | ! picen2(klon) n2 ice at the surface (kg/m2) |
---|
38 | ! |
---|
39 | ! Outputs |
---|
40 | ! ------- |
---|
41 | ! pdpsrf(klon) \ Contribution of condensation/sublimation |
---|
42 | ! pdtc(klon,klev) / to the time derivatives of Ps, pt, and ptsrf |
---|
43 | ! pdtsrfc(klon) / |
---|
44 | ! pdicen2(klon) Tendancy n2 ice at the surface (kg/m2) |
---|
45 | ! |
---|
46 | ! Both |
---|
47 | ! ---- |
---|
48 | ! psolaralb(klon) Albedo at the surface |
---|
49 | ! pemisurf(klon) Emissivity of the surface |
---|
50 | ! |
---|
51 | ! Authors |
---|
52 | ! ------- |
---|
53 | ! Francois Forget (1996,2013) |
---|
54 | ! Converted to Fortran 90 and slightly modified by R. Wordsworth (2009) |
---|
55 | ! |
---|
56 | ! |
---|
57 | !================================================================== |
---|
58 | |
---|
59 | !----------------------------------------------------------------------- |
---|
60 | ! Arguments |
---|
61 | |
---|
62 | INTEGER klon, klev, nq |
---|
63 | |
---|
64 | REAL ptimestep |
---|
65 | REAL pplay(klon,klev),pplev(klon,klev+1) |
---|
66 | REAL pcapcal(klon) |
---|
67 | REAL pt(klon,klev) |
---|
68 | REAL ptsrf(klon),flu1(klon),flu2(klon),flu3(klon) |
---|
69 | REAL pphi(klon,klev) |
---|
70 | REAL pdt(klon,klev),pdtsrf(klon),pdtc(klon,klev) |
---|
71 | REAL pdtsrfc(klon),pdpsrf(klon) |
---|
72 | REAL picen2(klon),psolaralb(klon),pemisurf(klon) |
---|
73 | |
---|
74 | |
---|
75 | REAL pu(klon,klev) , pv(klon,klev) |
---|
76 | REAL pdu(klon,klev) , pdv(klon,klev) |
---|
77 | REAL pduc(klon,klev) , pdvc(klon,klev) |
---|
78 | REAL pq(klon,klev,nq),pdq(klon,klev,nq) |
---|
79 | REAL pdqc(klon,klev,nq) |
---|
80 | |
---|
81 | !----------------------------------------------------------------------- |
---|
82 | ! Local variables |
---|
83 | |
---|
84 | INTEGER l,ig,ilay,it,iq,ich4_gas |
---|
85 | |
---|
86 | REAL*8 zt(klon,klev) |
---|
87 | REAL tcond_n2 |
---|
88 | REAL pcond_n2 |
---|
89 | REAL glob_average2d ! function |
---|
90 | REAL zqn2(klon,klev) ! N2 MMR used to compute Tcond/zqn2 |
---|
91 | REAL ztcond (klon,klev) |
---|
92 | REAL ztcondsol(klon),zfallheat |
---|
93 | REAL pdicen2(klon) |
---|
94 | REAL zcondicea(klon,klev), zcondices(klon) |
---|
95 | REAL zfallice(klon,klev+1) |
---|
96 | REAL zmflux(klev+1) |
---|
97 | REAL zu(klev),zv(klev) |
---|
98 | REAL zq(klev,nq),zq1(klev) |
---|
99 | REAL ztsrf(klon) |
---|
100 | REAL ztc(klev), ztm(klev+1) |
---|
101 | REAL zum(klev+1) , zvm(klev+1) |
---|
102 | REAL zqm(klev+1,nq),zqm1(klev+1) |
---|
103 | LOGICAL condsub(klon) |
---|
104 | REAL subptimestep |
---|
105 | Integer Ntime |
---|
106 | real masse (klev),w(klev+1) |
---|
107 | real wq(klon,klev+1) |
---|
108 | real vstokes,reff |
---|
109 | real dWtotsn2 |
---|
110 | real condnconsn2(klon) |
---|
111 | real nconsMAXn2 |
---|
112 | ! Special diagnostic variables |
---|
113 | real tconda1(klon,klev) |
---|
114 | real tconda2(klon,klev) |
---|
115 | real zdtsig (klon,klev) |
---|
116 | real zdtlatent (klon,klev) |
---|
117 | real zdt (klon,klev) |
---|
118 | ! REAL albediceF(klon) |
---|
119 | ! SAVE albediceF |
---|
120 | INTEGER nsubtimestep,itsub !number of subtimestep when calling vl1d |
---|
121 | REAL subtimestep !ptimestep/nsubtimestep |
---|
122 | REAL dtmax |
---|
123 | |
---|
124 | REAL zplevhist(klon) |
---|
125 | REAL zplevnew(klon) |
---|
126 | REAL zplev(klon) |
---|
127 | REAL zpicen2(klon) |
---|
128 | REAL ztsrfhist(klon) |
---|
129 | REAL zdtsrf(klon) |
---|
130 | REAL globzplevnew |
---|
131 | |
---|
132 | real,dimension(:),save,allocatable :: vmrn2 |
---|
133 | !$OMP THREADPRIVATE(vmrn2) |
---|
134 | REAL stephan |
---|
135 | DATA stephan/5.67e-08/ ! Stephan Boltzman constant |
---|
136 | SAVE stephan |
---|
137 | !----------------------------------------------------------------------- |
---|
138 | ! Saved local variables |
---|
139 | |
---|
140 | ! REAL latcond |
---|
141 | REAL ccond |
---|
142 | REAL cpice ! for atm condensation |
---|
143 | SAVE cpice |
---|
144 | ! SAVE latcond,ccond |
---|
145 | SAVE ccond |
---|
146 | |
---|
147 | LOGICAL firstcall |
---|
148 | SAVE firstcall |
---|
149 | REAL SSUM |
---|
150 | EXTERNAL SSUM |
---|
151 | |
---|
152 | ! DATA latcond /2.5e5/ |
---|
153 | ! DATA latcond /1.98e5/ |
---|
154 | DATA cpice /1300./ |
---|
155 | DATA firstcall/.true./ |
---|
156 | |
---|
157 | INTEGER,SAVE :: i_n2ice=0 ! n2 ice |
---|
158 | CHARACTER(LEN=20) :: tracername ! to temporarily store text |
---|
159 | logical olkin ! option to prevent N2 ice effect in the south |
---|
160 | DATA olkin/.false./ |
---|
161 | save olkin |
---|
162 | |
---|
163 | CHARACTER(len=10) :: tname |
---|
164 | |
---|
165 | !----------------------------------------------------------------------- |
---|
166 | |
---|
167 | ! Initialisation |
---|
168 | IF (firstcall) THEN |
---|
169 | ccond=cpp/(g*lw_n2) |
---|
170 | print*,'In condense_n2cloud: ccond=',ccond,' latcond=',lw_n2 |
---|
171 | |
---|
172 | ! calculate global mean surface pressure for the fast mode |
---|
173 | IF (.not. ALLOCATED(kp)) ALLOCATE(kp(klon)) |
---|
174 | DO ig=1,klon |
---|
175 | kp(ig) = exp(-phisfi(ig)/(r*38.)) |
---|
176 | ENDDO |
---|
177 | IF (fast) THEN |
---|
178 | p00=glob_average2d(kp) ! mean pres at ref level |
---|
179 | ENDIF |
---|
180 | |
---|
181 | ALLOCATE(vmrn2(klon)) |
---|
182 | vmrn2(:) = 1. |
---|
183 | !IF (ch4lag) then |
---|
184 | ! DO ig=1,klon |
---|
185 | ! if (latitude(ig)*180./pi.ge.latlag) then |
---|
186 | ! vmrn2(ig) = vmrlag |
---|
187 | ! endif |
---|
188 | ! ENDDO |
---|
189 | !ENDIF |
---|
190 | IF (no_n2frost) then |
---|
191 | DO ig=1,klon |
---|
192 | if (picen2(ig).eq.0.) then |
---|
193 | vmrn2(ig) = 1.e-15 |
---|
194 | endif |
---|
195 | ENDDO |
---|
196 | ENDIF |
---|
197 | firstcall=.false. |
---|
198 | ENDIF |
---|
199 | |
---|
200 | !----------------------------------------------------------------------- |
---|
201 | ! Calculation of n2 condensation / sublimation |
---|
202 | |
---|
203 | ! Variables used: |
---|
204 | ! picen2(klon) : amount of n2 ice on the ground (kg/m2) |
---|
205 | ! zcondicea(klon,klev): condensation rate in layer l (kg/m2/s) |
---|
206 | ! zcondices(klon) : condensation rate on the ground (kg/m2/s) |
---|
207 | ! zfallice(klon,klev) : amount of ice falling from layer l (kg/m2/s) |
---|
208 | ! zdtlatent(klon,klev): dT/dt due to phase changes (K/s) |
---|
209 | |
---|
210 | ! Tendencies initially set to 0 |
---|
211 | zcondices(1:klon) = 0. |
---|
212 | pdtsrfc(1:klon) = 0. |
---|
213 | pdpsrf(1:klon) = 0. |
---|
214 | ztsrfhist(1:klon) = 0. |
---|
215 | condsub(1:klon) = .false. |
---|
216 | pdicen2(1:klon) = 0. |
---|
217 | zfallheat=0 |
---|
218 | pdqc(1:klon,1:klev,1:nq)=0 |
---|
219 | pdtc(1:klon,1:klev)=0 |
---|
220 | pduc(1:klon,1:klev)=0 |
---|
221 | pdvc(1:klon,1:klev)=0 |
---|
222 | zfallice(1:klon,1:klev+1)=0 |
---|
223 | zcondicea(1:klon,1:klev)=0 |
---|
224 | zdtlatent(1:klon,1:klev)=0 |
---|
225 | zt(1:klon,1:klev)=0. |
---|
226 | |
---|
227 | !----------------------------------------------------------------------- |
---|
228 | ! Atmospheric condensation |
---|
229 | |
---|
230 | ! Condensation / sublimation in the atmosphere |
---|
231 | ! -------------------------------------------- |
---|
232 | ! (calcul of zcondicea , zfallice and pdtc) |
---|
233 | |
---|
234 | zt(1:klon,1:klev)=pt(1:klon,1:klev)+ pdt(1:klon,1:klev)*ptimestep |
---|
235 | if (igcm_n2.ne.0) then |
---|
236 | zqn2(1:klon,1:klev) = 1. ! & temporaire |
---|
237 | ! zqn2(1:klon,1:klev)= pq(1:klon,1:klev,igcm_n2) + pdq(1:klon,1:klev,igcm_n2)*ptimestep |
---|
238 | else |
---|
239 | zqn2(1:klon,1:klev) = 1. |
---|
240 | end if |
---|
241 | |
---|
242 | if (.not.fast) then |
---|
243 | ! Forecast the atmospheric frost temperature 'ztcond' with function tcond_n2 |
---|
244 | DO l=1,klev |
---|
245 | DO ig=1,klon |
---|
246 | ztcond (ig,l) = tcond_n2(pplay(ig,l),zqn2(ig,l)) |
---|
247 | ENDDO |
---|
248 | ENDDO |
---|
249 | |
---|
250 | DO l=klev,1,-1 |
---|
251 | DO ig=1,klon |
---|
252 | pdtc(ig,l)=0. ! final tendancy temperature set to 0 |
---|
253 | |
---|
254 | IF((zt(ig,l).LT.ztcond(ig,l)).or.(zfallice(ig,l+1).gt.0))THEN |
---|
255 | condsub(ig)=.true. !condensation at level l |
---|
256 | IF (zfallice(ig,l+1).gt.0) then |
---|
257 | zfallheat=zfallice(ig,l+1)*& |
---|
258 | (pphi(ig,l+1)-pphi(ig,l) +& |
---|
259 | cpice*(ztcond(ig,l+1)-ztcond(ig,l)))/lw_n2 |
---|
260 | ELSE |
---|
261 | zfallheat=0. |
---|
262 | ENDIF |
---|
263 | zdtlatent(ig,l)=(ztcond(ig,l) - zt(ig,l))/ptimestep |
---|
264 | zcondicea(ig,l)=(pplev(ig,l)-pplev(ig,l+1))& |
---|
265 | *ccond*zdtlatent(ig,l)- zfallheat |
---|
266 | ! Case when the ice from above sublimes entirely |
---|
267 | ! """"""""""""""""""""""""""""""""""""""""""""""" |
---|
268 | IF ((zfallice(ig,l+1).lt.-zcondicea(ig,l)) & |
---|
269 | .AND. (zfallice(ig,l+1).gt.0)) THEN |
---|
270 | |
---|
271 | zdtlatent(ig,l)=(-zfallice(ig,l+1)+zfallheat)/& |
---|
272 | (ccond*(pplev(ig,l)-pplev(ig,l+1))) |
---|
273 | zcondicea(ig,l)= -zfallice(ig,l+1) |
---|
274 | END IF |
---|
275 | |
---|
276 | zfallice(ig,l) = zcondicea(ig,l)+zfallice(ig,l+1) |
---|
277 | |
---|
278 | END IF |
---|
279 | |
---|
280 | ENDDO |
---|
281 | ENDDO |
---|
282 | endif ! not fast |
---|
283 | |
---|
284 | !----------------------------------------------------------------------- |
---|
285 | ! Condensation/sublimation on the ground |
---|
286 | ! (calculation of zcondices and pdtsrfc) |
---|
287 | |
---|
288 | ! Adding subtimesteps : in fast version, pressures too low lead to |
---|
289 | ! instabilities. |
---|
290 | IF (fast) THEN |
---|
291 | IF (pplev(1,1).gt.0.3) THEN |
---|
292 | nsubtimestep= 1 |
---|
293 | ELSE |
---|
294 | nsubtimestep= nbsub !max(nint(ptimestep/dtmax),1) |
---|
295 | ENDIF |
---|
296 | ELSE |
---|
297 | nsubtimestep= 1 ! if more then kp and p00 have to be calculated |
---|
298 | ! for nofast mode |
---|
299 | ENDIF |
---|
300 | subtimestep=ptimestep/float(nsubtimestep) |
---|
301 | |
---|
302 | do itsub=1,nsubtimestep |
---|
303 | ! first loop : getting zplev, ztsurf |
---|
304 | IF (itsub.eq.1) then |
---|
305 | DO ig=1,klon |
---|
306 | zplev(ig)=pplev(ig,1) |
---|
307 | ztsrfhist(ig)=ptsrf(ig) + pdtsrf(ig)*ptimestep |
---|
308 | ztsrf(ig)=ptsrf(ig) + pdtsrf(ig)*subtimestep !! |
---|
309 | zpicen2(ig)=picen2(ig) |
---|
310 | ENDDO |
---|
311 | ELSE |
---|
312 | ! additional loop : |
---|
313 | ! 1) pressure updated |
---|
314 | ! 2) direct redistribution of pressure over the globe |
---|
315 | ! 3) modification pressure for unstable cases |
---|
316 | ! 4) pressure update to conserve mass |
---|
317 | ! 5) temperature updated with radiative tendancies |
---|
318 | DO ig=1,klon |
---|
319 | zplevhist(ig)=zplev(ig) |
---|
320 | zplevnew(ig)=zplev(ig)+pdpsrf(ig)*subtimestep ! 1) |
---|
321 | !IF (zplevnew(ig).le.0.0001) then |
---|
322 | ! zplevnew(ig)=0.0001*kp(ig)/p00 |
---|
323 | !ENDIF |
---|
324 | ENDDO |
---|
325 | ! intermediaire de calcul: valeur moyenne de zplevnew (called twice in the code) |
---|
326 | globzplevnew=glob_average2d(zplevnew) |
---|
327 | DO ig=1,klon |
---|
328 | zplev(ig)=kp(ig)*globzplevnew/p00 ! 2) |
---|
329 | ENDDO |
---|
330 | DO ig=1,klon ! 3) unstable case condensation and sublimation: zplev=zplevhist |
---|
331 | IF (((pdpsrf(ig).lt.0.).and.(tcond_n2(zplev(ig),zqn2(ig,1)).le.ztsrf(ig))).or. & |
---|
332 | ((pdpsrf(ig).gt.0.).and.(tcond_n2(zplev(ig),zqn2(ig,1)).ge.ztsrf(ig)))) then |
---|
333 | zplev(ig)=zplevhist(ig) |
---|
334 | ENDIF |
---|
335 | zplevhist(ig)=zplev(ig) |
---|
336 | ENDDO |
---|
337 | zplev=zplev*globzplevnew/glob_average2d(zplevhist) ! 4) |
---|
338 | DO ig=1,klon ! 5) |
---|
339 | zdtsrf(ig)=pdtsrf(ig) + (stephan/pcapcal(ig))*(ptsrf(ig)**4-ztsrf(ig)**4) |
---|
340 | ztsrf(ig)=ztsrf(ig)+pdtsrfc(ig)*subtimestep+zdtsrf(ig)*subtimestep |
---|
341 | zpicen2(ig)=zpicen2(ig)+pdicen2(ig)*subtimestep |
---|
342 | ENDDO |
---|
343 | ENDIF ! (itsub=1) |
---|
344 | |
---|
345 | DO ig=1,klon |
---|
346 | ! forecast of frost temperature ztcondsol |
---|
347 | !ztcondsol(ig) = tcond_n2(zplev(ig),zqn2(ig,1)) |
---|
348 | ztcondsol(ig) = tcond_n2(zplev(ig),vmrn2(ig)) |
---|
349 | |
---|
350 | ! Loop over where we have condensation / sublimation |
---|
351 | IF ((ztsrf(ig) .LT. ztcondsol(ig)) .OR. & ! ground cond |
---|
352 | ((ztsrf(ig) .GT. ztcondsol(ig)) .AND. & ! ground sublim |
---|
353 | (zpicen2(ig) .GT. 0.))) THEN |
---|
354 | condsub(ig) = .true. ! condensation or sublimation |
---|
355 | |
---|
356 | ! Condensation or partial sublimation of N2 ice |
---|
357 | if (ztsrf(ig) .LT. ztcondsol(ig)) then ! condensation |
---|
358 | ! Include a correction to account for the cooling of air near |
---|
359 | ! the surface before condensing: |
---|
360 | zcondices(ig)=pcapcal(ig)*(ztcondsol(ig)-ztsrf(ig)) & |
---|
361 | /((lw_n2+cpp*(zt(ig,1)-ztcondsol(ig)))*subtimestep) |
---|
362 | else ! sublimation |
---|
363 | zcondices(ig)=pcapcal(ig)*(ztcondsol(ig)-ztsrf(ig)) & |
---|
364 | /(lw_n2*subtimestep) |
---|
365 | end if |
---|
366 | |
---|
367 | pdtsrfc(ig) = (ztcondsol(ig) - ztsrf(ig))/subtimestep |
---|
368 | |
---|
369 | ! partial sublimation of N2 ice |
---|
370 | ! If the entire N_2 ice layer sublimes |
---|
371 | ! (including what has just condensed in the atmosphere) |
---|
372 | IF((zpicen2(ig)/subtimestep).LE. & |
---|
373 | -zcondices(ig))THEN |
---|
374 | zcondices(ig) = -zpicen2(ig)/subtimestep |
---|
375 | pdtsrfc(ig)=(lw_n2/pcapcal(ig))* & |
---|
376 | (zcondices(ig)) |
---|
377 | END IF |
---|
378 | |
---|
379 | ! Changing N2 ice amount and pressure |
---|
380 | |
---|
381 | pdicen2(ig) = zcondices(ig) |
---|
382 | pdpsrf(ig) = -pdicen2(ig)*g |
---|
383 | ! pdpsrf(ig) = 0. ! OPTION to check impact N2 sub/cond |
---|
384 | IF (fast.and.(zplev(ig)+pdpsrf(ig)*subtimestep.le.0.0000001)) then |
---|
385 | pdpsrf(ig)=(0.0000001*kp(ig)/p00-zplev(ig))/subtimestep |
---|
386 | pdicen2(ig)=-pdpsrf(ig)/g |
---|
387 | ENDIF |
---|
388 | |
---|
389 | ELSE ! no condsub |
---|
390 | pdpsrf(ig)=0. |
---|
391 | pdicen2(ig)=0. |
---|
392 | pdtsrfc(ig)=0. |
---|
393 | ENDIF |
---|
394 | ENDDO ! ig |
---|
395 | enddo ! subtimestep |
---|
396 | |
---|
397 | ! Updating pressure, temperature and ice reservoir |
---|
398 | DO ig=1,klon |
---|
399 | pdpsrf(ig)=(zplev(ig)+pdpsrf(ig)*subtimestep-pplev(ig,1))/ptimestep |
---|
400 | ! Two options here : 1 ok, 2 is wrong |
---|
401 | pdicen2(ig)=(zpicen2(ig)+pdicen2(ig)*subtimestep-picen2(ig))/ptimestep |
---|
402 | !pdicen2(ig)=-pdpsrf(ig)/g |
---|
403 | |
---|
404 | pdtsrfc(ig)=((ztsrf(ig)+pdtsrfc(ig)*subtimestep)-(ztsrfhist(ig)))/ptimestep |
---|
405 | |
---|
406 | ! security |
---|
407 | if (picen2(ig) + pdicen2(ig)*ptimestep.lt.0.) then |
---|
408 | write(*,*) 'WARNING in condense_n2:' |
---|
409 | write(*,*) picen2(ig),pdicen2(ig)*ptimestep |
---|
410 | pdicen2(ig)= -picen2(ig)/ptimestep |
---|
411 | pdpsrf(ig)=-pdicen2(ig)*g |
---|
412 | endif |
---|
413 | |
---|
414 | if(.not.picen2(ig).ge.0.) THEN |
---|
415 | ! if(picen2(ig) + pdicen2(ig)*ptimestep.le.-1.e-8) then |
---|
416 | print*, 'WARNING NEG RESERVOIR in condense_n2: picen2(',ig,')=', picen2(ig) + pdicen2(ig)*ptimestep |
---|
417 | ! pdicen2(ig)= -picen2(ig)/ptimestep |
---|
418 | ! else |
---|
419 | picen2(ig)=0.0 |
---|
420 | ! endif |
---|
421 | endif |
---|
422 | ENDDO |
---|
423 | |
---|
424 | ! *************************************************************** |
---|
425 | ! Correction to account for redistribution between sigma or hybrid |
---|
426 | ! layers when changing surface pressure (and warming/cooling |
---|
427 | ! of the n2 currently changing phase). |
---|
428 | ! ************************************************************* |
---|
429 | if (.not.fast) then |
---|
430 | DO ig=1,klon |
---|
431 | if (condsub(ig)) then |
---|
432 | |
---|
433 | ! Mass fluxes through the sigma levels (kg.m-2.s-1) (>0 when up) |
---|
434 | ! """""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" |
---|
435 | zmflux(1) = -zcondices(ig) |
---|
436 | DO l=1,klev |
---|
437 | zmflux(l+1) = zmflux(l) -zcondicea(ig,l) & |
---|
438 | + (bp(l)-bp(l+1))*(zfallice(ig,1)-zmflux(1)) |
---|
439 | ! zmflux set to 0 if very low to avoid: top layer is disappearing in v1ld |
---|
440 | if (abs(zmflux(l+1)).lt.1E-13.OR.bp(l+1).eq.0.) zmflux(l+1)=0. |
---|
441 | END DO |
---|
442 | |
---|
443 | ! Mass of each layer |
---|
444 | ! ------------------ |
---|
445 | DO l=1,klev |
---|
446 | masse(l)=(pplev(ig,l) - pplev(ig,l+1))/g |
---|
447 | END DO |
---|
448 | |
---|
449 | |
---|
450 | ! Corresponding fluxes for T,U,V,Q |
---|
451 | ! """""""""""""""""""""""""""""""" |
---|
452 | ! averaging operator for TRANSPORT |
---|
453 | ! """""""""""""""""""""""""""""""" |
---|
454 | |
---|
455 | ! Subtimestep loop to perform the redistribution gently and simultaneously with |
---|
456 | ! the other tendencies |
---|
457 | ! Estimation of subtimestep (using only the first layer, the most critical) |
---|
458 | dtmax=ptimestep |
---|
459 | if (zmflux(1).gt.1.e-20) then |
---|
460 | dtmax=min(dtmax,masse(1)*zqn2(ig,1)/abs(zmflux(1))) |
---|
461 | endif |
---|
462 | nsubtimestep= max(nint(ptimestep/dtmax),nint(2.)) |
---|
463 | subtimestep=ptimestep/float(nsubtimestep) |
---|
464 | |
---|
465 | ! New flux for each subtimestep |
---|
466 | do l=1,klev+1 |
---|
467 | w(l)=-zmflux(l)*subtimestep |
---|
468 | enddo |
---|
469 | ! initializing variables that will vary during subtimestep: |
---|
470 | do l=1,klev |
---|
471 | ztc(l) =pt(ig,l) |
---|
472 | zu(l) =pu(ig,l) |
---|
473 | zv(l) =pv(ig,l) |
---|
474 | do iq=1,nq |
---|
475 | zq(l,iq) = pq(ig,l,iq) |
---|
476 | enddo |
---|
477 | end do |
---|
478 | |
---|
479 | ! loop over nsubtimestep |
---|
480 | ! """""""""""""""""""""" |
---|
481 | do itsub=1,nsubtimestep |
---|
482 | ! Progressively adding tendancies from other processes. |
---|
483 | do l=1,klev |
---|
484 | ztc(l) =ztc(l) +(pdt(ig,l) + zdtlatent(ig,l))*subtimestep |
---|
485 | zu(l) =zu(l) +pdu( ig,l) * subtimestep |
---|
486 | zv(l) =zv(l) +pdv( ig,l) * subtimestep |
---|
487 | do iq=1,nq |
---|
488 | zq(l,iq) = zq(l,iq) + pdq(ig,l,iq)* subtimestep |
---|
489 | enddo |
---|
490 | end do |
---|
491 | |
---|
492 | ! Change of mass in each layer |
---|
493 | do l=1,klev |
---|
494 | masse(l)=masse(l)+pdpsrf(ig)*subtimestep*(pplev(ig,l) - pplev(ig,l+1))& |
---|
495 | /(g*pplev(ig,1)) |
---|
496 | end do |
---|
497 | |
---|
498 | ! Value transfert at the surface interface when condensation sublimation: |
---|
499 | |
---|
500 | if (zmflux(1).lt.0) then |
---|
501 | ! Surface condensation |
---|
502 | zum(1)= zu(1) |
---|
503 | zvm(1)= zv(1) |
---|
504 | ztm(1) = ztc(1) |
---|
505 | else |
---|
506 | ! Surface sublimation: |
---|
507 | ztm(1) = ztsrf(ig) + pdtsrfc(ig)*ptimestep |
---|
508 | zum(1) = 0 |
---|
509 | zvm(1) = 0 |
---|
510 | end if |
---|
511 | do iq=1,nq |
---|
512 | zqm(1,iq)=0. ! most tracer do not condense ! |
---|
513 | enddo |
---|
514 | ! Special case if the tracer is n2 gas |
---|
515 | if (igcm_n2.ne.0) zqm(1,igcm_n2)=1. |
---|
516 | |
---|
517 | ztm(2:klev+1)=0. |
---|
518 | zum(2:klev+1)=0. |
---|
519 | zvm(2:klev+1)=0. |
---|
520 | zqm1(1:klev+1)=0. |
---|
521 | |
---|
522 | ! Van Leer scheme: |
---|
523 | call vl1d(klev,ztc,2.,masse,w,ztm) |
---|
524 | call vl1d(klev,zu ,2.,masse,w,zum) |
---|
525 | call vl1d(klev,zv ,2.,masse,w,zvm) |
---|
526 | do iq=1,nq |
---|
527 | do l=1,klev |
---|
528 | zq1(l)=zq(l,iq) |
---|
529 | enddo |
---|
530 | zqm1(1)=zqm(1,iq) |
---|
531 | call vl1d(klev,zq1,2.,masse,w,zqm1) |
---|
532 | do l=2,klev |
---|
533 | zqm(l,iq)=zqm1(l) |
---|
534 | enddo |
---|
535 | enddo |
---|
536 | |
---|
537 | ! Correction to prevent negative value for qn2 |
---|
538 | if (igcm_n2.ne.0) then |
---|
539 | zqm(1,igcm_n2)=1. |
---|
540 | do l=1,klev-1 |
---|
541 | if (w(l)*zqm(l,igcm_n2).gt.zq(l,igcm_n2)*masse(l)) then |
---|
542 | zqm(l+1,igcm_n2)=max(zqm(l+1,igcm_n2), & |
---|
543 | (zqm(l,igcm_n2)*w(l) -zq(l,igcm_n2)*masse(l))/w(l+1) ) |
---|
544 | else |
---|
545 | exit |
---|
546 | endif |
---|
547 | end do |
---|
548 | end if |
---|
549 | |
---|
550 | ! Value transfert at the surface interface when condensation sublimation: |
---|
551 | if (zmflux(1).lt.0) then |
---|
552 | ! Surface condensation |
---|
553 | zum(1)= zu(1) |
---|
554 | zvm(1)= zv(1) |
---|
555 | ztm(1) = ztc(1) |
---|
556 | else |
---|
557 | ! Surface sublimation: |
---|
558 | ztm(1) = ztsrf(ig) + pdtsrfc(ig)*ptimestep |
---|
559 | zum(1) = 0 |
---|
560 | zvm(1) = 0 |
---|
561 | end if |
---|
562 | do iq=1,nq |
---|
563 | zqm(1,iq)=0. ! most tracer do not condense ! |
---|
564 | enddo |
---|
565 | ! Special case if the tracer is n2 gas |
---|
566 | if (igcm_n2.ne.0) zqm(1,igcm_n2)=1. |
---|
567 | |
---|
568 | !!! Source haze: 0.02 pourcent when n2 sublimes |
---|
569 | IF (source_haze) THEN |
---|
570 | IF (pdicen2(ig).lt.0) THEN |
---|
571 | DO iq=1,nq |
---|
572 | tname=noms(iq) |
---|
573 | if (tname(1:4).eq."haze") then |
---|
574 | !zqm(1,iq)=0.02 |
---|
575 | !zqm(1,iq)=-pdicen2(ig)*0.02 |
---|
576 | zqm(1,iq)=-pdicen2(ig)*ptimestep*0.02 |
---|
577 | !zqm(10,iq)=-pdicen2(ig)*ptimestep*100. |
---|
578 | !zqm(1,iq)=-pdicen2(ig)*1000000. |
---|
579 | |
---|
580 | endif |
---|
581 | ENDDO |
---|
582 | ENDIF |
---|
583 | ENDIF |
---|
584 | ztm(klev+1)= ztc(klev) ! should not be used, but... |
---|
585 | zum(klev+1)= zu(klev) ! should not be used, but... |
---|
586 | zvm(klev+1)= zv(klev) ! should not be used, but... |
---|
587 | do iq=1,nq |
---|
588 | zqm(klev+1,iq)= zq(klev,iq) |
---|
589 | enddo |
---|
590 | |
---|
591 | ! Tendencies on T, U, V, Q |
---|
592 | ! """"""""""""""""""""""" |
---|
593 | DO l=1,klev |
---|
594 | |
---|
595 | ! Tendencies on T |
---|
596 | zdtsig(ig,l) = (1/masse(l)) * & |
---|
597 | ( zmflux(l)*(ztm(l) - ztc(l)) & |
---|
598 | - zmflux(l+1)*(ztm(l+1) - ztc(l)) & |
---|
599 | + zcondicea(ig,l)*(ztcond(ig,l)-ztc(l)) ) |
---|
600 | |
---|
601 | ! Tendencies on U |
---|
602 | pduc(ig,l) = (1/masse(l)) * & |
---|
603 | ( zmflux(l)*(zum(l) - zu(l))& |
---|
604 | - zmflux(l+1)*(zum(l+1) - zu(l)) ) |
---|
605 | |
---|
606 | ! Tendencies on V |
---|
607 | pdvc(ig,l) = (1/masse(l)) * & |
---|
608 | ( zmflux(l)*(zvm(l) - zv(l)) & |
---|
609 | - zmflux(l+1)*(zvm(l+1) - zv(l)) ) |
---|
610 | |
---|
611 | END DO |
---|
612 | |
---|
613 | ! Tendencies on Q |
---|
614 | do iq=1,nq |
---|
615 | if (iq.eq.igcm_n2) then |
---|
616 | ! SPECIAL Case when the tracer IS N2 : |
---|
617 | DO l=1,klev |
---|
618 | pdqc(ig,l,iq)= (1/masse(l)) * & |
---|
619 | ( zmflux(l)*(zqm(l,iq) - zq(l,iq)) & |
---|
620 | - zmflux(l+1)*(zqm(l+1,iq) - zq(l,iq))& |
---|
621 | + zcondicea(ig,l)*(zq(l,iq)-1.) ) |
---|
622 | END DO |
---|
623 | else |
---|
624 | DO l=1,klev |
---|
625 | pdqc(ig,l,iq)= (1/masse(l)) * & |
---|
626 | ( zmflux(l)*(zqm(l,iq) - zq(l,iq)) & |
---|
627 | - zmflux(l+1)*(zqm(l+1,iq) - zq(l,iq)) & |
---|
628 | + zcondicea(ig,l)*zq(l,iq) ) |
---|
629 | END DO |
---|
630 | end if |
---|
631 | enddo |
---|
632 | ! Update variables at the end of each subtimestep. |
---|
633 | do l=1,klev |
---|
634 | ztc(l) =ztc(l) + zdtsig(ig,l) *subtimestep |
---|
635 | zu(l) =zu(l) + pduc(ig,l) *subtimestep |
---|
636 | zv(l) =zv(l) + pdvc(ig,l) *subtimestep |
---|
637 | do iq=1,nq |
---|
638 | zq(l,iq) = zq(l,iq) + pdqc(ig,l,iq)*subtimestep |
---|
639 | enddo |
---|
640 | end do |
---|
641 | enddo ! loop on nsubtimestep |
---|
642 | ! Recomputing Total tendencies |
---|
643 | do l=1,klev |
---|
644 | pdtc(ig,l) = (ztc(l) - zt(ig,l) )/ptimestep |
---|
645 | pduc(ig,l) = (zu(l) - (pu(ig,l) + pdu(ig,l)*ptimestep))/ptimestep |
---|
646 | pdvc(ig,l) = (zv(l) - (pv(ig,l) + pdv(ig,l)*ptimestep))/ptimestep |
---|
647 | do iq=1,nq |
---|
648 | pdqc(ig,l,iq) = (zq(l,iq) - (pq(ig,l,iq) + pdq(ig,l,iq)*ptimestep))/ptimestep |
---|
649 | |
---|
650 | |
---|
651 | ! Correction temporaire |
---|
652 | if (iq.eq.igcm_n2) then |
---|
653 | if((pq(ig,l,iq) +(pdqc(ig,l,iq)+ pdq(ig,l,iq))*ptimestep) & |
---|
654 | .lt.0.01) then ! if n2 < 1 % ! |
---|
655 | write(*,*) 'Warning: n2 < 1%' |
---|
656 | pdqc(ig,l,iq)=(0.01-pq(ig,l,iq))/ptimestep-pdq(ig,l,iq) |
---|
657 | end if |
---|
658 | end if |
---|
659 | |
---|
660 | enddo |
---|
661 | end do |
---|
662 | ! *******************************TEMPORAIRE ****************** |
---|
663 | if (klon.eq.1) then |
---|
664 | write(*,*) 'nsubtimestep=' ,nsubtimestep |
---|
665 | write(*,*) 'masse avant' , (pplev(ig,1) - pplev(ig,2))/g |
---|
666 | write(*,*) 'masse apres' , masse(1) |
---|
667 | write(*,*) 'zmflux*DT, l=1' , zmflux(1)*ptimestep |
---|
668 | write(*,*) 'zmflux*DT, l=2' , zmflux(2)*ptimestep |
---|
669 | write(*,*) 'pq, l=1,2,3' , pq(1,1,1), pq(1,2,1),pq(1,3,1) |
---|
670 | write(*,*) 'zq, l=1,2,3' , zq(1,1), zq(2,1),zq(3,1) |
---|
671 | write(*,*) 'dq*Dt l=1' , pdq(1,1,1)*ptimestep |
---|
672 | write(*,*) 'dqcond*Dt l=1' , pdqc(1,1,1)*ptimestep |
---|
673 | end if |
---|
674 | |
---|
675 | ! *********************************************************** |
---|
676 | end if ! if (condsub) |
---|
677 | END DO ! loop on ig |
---|
678 | endif ! not fast |
---|
679 | |
---|
680 | ! ************ Option Olkin to prevent N2 effect in the south******** |
---|
681 | 112 continue |
---|
682 | if (olkin) then |
---|
683 | DO ig=1,klon |
---|
684 | if (latitude(ig).lt.0) then |
---|
685 | pdtsrfc(ig) = max(0.,pdtsrfc(ig)) |
---|
686 | pdpsrf(ig) = 0. |
---|
687 | pdicen2(ig) = 0. |
---|
688 | do l=1,klev |
---|
689 | pdtc(ig,l) = max(0.,zdtlatent(ig,l)) |
---|
690 | pduc(ig,l) = 0. |
---|
691 | pdvc(ig,l) = 0. |
---|
692 | do iq=1,nq |
---|
693 | pdqc(ig,l,iq) = 0 |
---|
694 | enddo |
---|
695 | end do |
---|
696 | end if |
---|
697 | END DO |
---|
698 | end if |
---|
699 | ! ******************************************************************* |
---|
700 | |
---|
701 | ! *************************************************************** |
---|
702 | ! Ecriture des diagnostiques |
---|
703 | ! *************************************************************** |
---|
704 | |
---|
705 | ! DO l=1,klev |
---|
706 | ! DO ig=1,klon |
---|
707 | ! Taux de cond en kg.m-2.pa-1.s-1 |
---|
708 | ! tconda1(ig,l)=zcondicea(ig,l)/(pplev(ig,l)-pplev(ig,l+1)) |
---|
709 | ! Taux de cond en kg.m-3.s-1 |
---|
710 | ! tconda2(ig,l)=tconda1(ig,l)*pplay(ig,l)*g/(r*pt(ig,l)) |
---|
711 | ! END DO |
---|
712 | ! END DO |
---|
713 | ! call WRITEDIAGFI(klon,'tconda1', & |
---|
714 | ! 'Taux de condensation N2 atmospherique /Pa', & |
---|
715 | ! 'kg.m-2.Pa-1.s-1',3,tconda1) |
---|
716 | ! call WRITEDIAGFI(klon,'tconda2', & |
---|
717 | ! 'Taux de condensation N2 atmospherique /m', & |
---|
718 | ! 'kg.m-3.s-1',3,tconda2) |
---|
719 | |
---|
720 | |
---|
721 | return |
---|
722 | end subroutine condense_n2 |
---|
723 | |
---|
724 | !------------------------------------------------------------------------- |
---|
725 | |
---|
726 | real function tcond_n2(p,vmr) |
---|
727 | ! Calculates the condensation temperature for N2 at pressure P and vmr |
---|
728 | implicit none |
---|
729 | real, intent(in):: p,vmr |
---|
730 | |
---|
731 | ! tcond_n2 = (1.)/(0.026315-0.0011877*log(.7143*p*vmr)) |
---|
732 | IF (p.ge.0.529995) then |
---|
733 | ! tcond Fray and Schmitt for N2 phase beta (T>35.6 K) FIT TB |
---|
734 | ! tcond_n2 = (1.)/(1./63.1470-296.925/(2.5e5*0.98)*log(1./(0.125570*1.e5)*p*vmr)) |
---|
735 | tcond_n2 = (1.)/(0.01583606505-1.211938776e-3*log(7.963685594e-5*p*vmr)) |
---|
736 | ELSE |
---|
737 | ! tcond Fray and Schmitt for N2 phase alpha(T<35.6 K) FIT BT |
---|
738 | ! tcond_n2 = (1.)/(1./35.6-296.925/(2.5e5*1.09)*log(1./(0.508059)*p*vmr)) |
---|
739 | tcond_n2 = (1.)/(1./35.6-1.089633028e-3*log(1.968275338*p*vmr)) |
---|
740 | ENDIF |
---|
741 | return |
---|
742 | end function tcond_n2 |
---|
743 | |
---|
744 | !------------------------------------------------------------------------- |
---|
745 | |
---|
746 | real function pcond_n2(t,vmr) |
---|
747 | ! Calculates the condensation pressure for N2 at temperature T and vmr |
---|
748 | implicit none |
---|
749 | real, intent(in):: t,vmr |
---|
750 | |
---|
751 | ! tcond_n2 = (1.)/(0.026315-0.0011877*log(.7143*p*vmr)) |
---|
752 | IF (t.ge.35.6) then |
---|
753 | ! tcond Fray and Schmitt for N2 phase beta (T>35.6 K) FIT TB |
---|
754 | ! pcond_n2 = 0.125570*1.e5/vmr*exp((2.5e5*0.98)/296.925*(1./63.1470-1./t)) |
---|
755 | pcond_n2 = 0.125570e5/vmr*exp(825.1241896*(1./63.147-1./t)) |
---|
756 | ELSE |
---|
757 | ! tcond Fray and Schmitt for N2 phase alpha(T<35.6 K) FIT TB |
---|
758 | ! pcond_n2 = 0.508059/vmr*exp((2.5e5*1.09)/296.925*(1./35.6-1./t)) |
---|
759 | pcond_n2 = 0.508059/vmr*exp(917.7401701*(1./35.6-1./t)) |
---|
760 | ENDIF |
---|
761 | return |
---|
762 | end function pcond_n2 |
---|
763 | |
---|
764 | !------------------------------------------------------------------------- |
---|
765 | |
---|
766 | real function glob_average2d(var) |
---|
767 | ! Calculates the global average of variable var |
---|
768 | use comgeomfi_h |
---|
769 | use dimphy, only: klon |
---|
770 | USE mod_grid_phy_lmdz, ONLY: nbp_lon, nbp_lat |
---|
771 | use geometry_mod, only: cell_area, latitude |
---|
772 | |
---|
773 | implicit none |
---|
774 | |
---|
775 | ! INTEGER klon |
---|
776 | REAL var(klon) |
---|
777 | INTEGER ig |
---|
778 | |
---|
779 | glob_average2d = 0. |
---|
780 | DO ig=2,klon-1 |
---|
781 | glob_average2d = glob_average2d + var(ig)*cell_area(ig) |
---|
782 | END DO |
---|
783 | glob_average2d = glob_average2d + var(1)*cell_area(1)*nbp_lon |
---|
784 | glob_average2d = glob_average2d + var(klon)*cell_area(klon)*nbp_lon |
---|
785 | glob_average2d = glob_average2d/(totarea+(cell_area(1)+cell_area(klon))*(nbp_lon-1)) |
---|
786 | |
---|
787 | end function glob_average2d |
---|
788 | |
---|
789 | ! ***************************************************************** |
---|
790 | |
---|
791 | subroutine vl1d(klev,q,pente_max,masse,w,qm) |
---|
792 | ! |
---|
793 | ! Operateur de moyenne inter-couche pour calcul de transport type |
---|
794 | ! Van-Leer " pseudo amont " dans la verticale |
---|
795 | ! q,w sont des arguments d'entree pour le s-pg .... |
---|
796 | ! masse : masse de la couche Dp/g |
---|
797 | ! w : masse d'atm ``transferee'' a chaque pas de temps (kg.m-2) |
---|
798 | ! pente_max = 2 conseillee |
---|
799 | ! -------------------------------------------------------------------- |
---|
800 | IMPLICIT NONE |
---|
801 | |
---|
802 | ! Arguments: |
---|
803 | ! ---------- |
---|
804 | integer klev |
---|
805 | real masse(klev),pente_max |
---|
806 | REAL q(klev),qm(klev+1) |
---|
807 | REAL w(klev+1) |
---|
808 | ! |
---|
809 | ! Local |
---|
810 | ! --------- |
---|
811 | ! |
---|
812 | INTEGER l |
---|
813 | ! |
---|
814 | real dzq(klev),dzqw(klev),adzqw(klev),dzqmax |
---|
815 | real sigw, Mtot, MQtot |
---|
816 | integer m |
---|
817 | |
---|
818 | |
---|
819 | ! On oriente tout dans le sens de la pression |
---|
820 | ! W > 0 WHEN DOWN !!!!!!!!!!!!! |
---|
821 | |
---|
822 | do l=2,klev |
---|
823 | dzqw(l)=q(l-1)-q(l) |
---|
824 | adzqw(l)=abs(dzqw(l)) |
---|
825 | enddo |
---|
826 | |
---|
827 | do l=2,klev-1 |
---|
828 | if(dzqw(l)*dzqw(l+1).gt.0.) then |
---|
829 | dzq(l)=0.5*(dzqw(l)+dzqw(l+1)) |
---|
830 | else |
---|
831 | dzq(l)=0. |
---|
832 | endif |
---|
833 | dzqmax=pente_max*min(adzqw(l),adzqw(l+1)) |
---|
834 | dzq(l)=sign(min(abs(dzq(l)),dzqmax),dzq(l)) |
---|
835 | enddo |
---|
836 | |
---|
837 | dzq(1)=0. |
---|
838 | dzq(klev)=0. |
---|
839 | |
---|
840 | do l = 1,klev-1 |
---|
841 | |
---|
842 | ! Regular scheme (transfered mass < layer mass) |
---|
843 | ! ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
---|
844 | if(w(l+1).gt.0. .and. w(l+1).le.masse(l+1)) then |
---|
845 | sigw=w(l+1)/masse(l+1) |
---|
846 | qm(l+1)=(q(l+1)+0.5*(1.-sigw)*dzq(l+1)) |
---|
847 | else if(w(l+1).le.0. .and. -w(l+1).le.masse(l)) then |
---|
848 | sigw=w(l+1)/masse(l) |
---|
849 | qm(l+1)=(q(l)-0.5*(1.+sigw)*dzq(l)) |
---|
850 | |
---|
851 | ! Extended scheme (transfered mass > layer mass) |
---|
852 | ! ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
---|
853 | else if(w(l+1).gt.0.) then |
---|
854 | m=l+1 |
---|
855 | Mtot = masse(m) |
---|
856 | MQtot = masse(m)*q(m) |
---|
857 | if (m.lt.klev) then ! because some compilers will have problems |
---|
858 | ! evaluating masse(klev+1) |
---|
859 | do while ((m.lt.klev).and.(w(l+1).gt.(Mtot+masse(m+1)))) |
---|
860 | m=m+1 |
---|
861 | Mtot = Mtot + masse(m) |
---|
862 | MQtot = MQtot + masse(m)*q(m) |
---|
863 | if (m.eq.klev) exit |
---|
864 | end do |
---|
865 | endif |
---|
866 | if (m.lt.klev) then |
---|
867 | sigw=(w(l+1)-Mtot)/masse(m+1) |
---|
868 | qm(l+1)= (1/w(l+1))*(MQtot + (w(l+1)-Mtot)* & |
---|
869 | (q(m+1)+0.5*(1.-sigw)*dzq(m+1)) ) |
---|
870 | else |
---|
871 | w(l+1) = Mtot |
---|
872 | qm(l+1) = Mqtot / Mtot |
---|
873 | write(*,*) 'top layer is disapearing !' |
---|
874 | stop |
---|
875 | end if |
---|
876 | else ! if(w(l+1).lt.0) |
---|
877 | m = l-1 |
---|
878 | Mtot = masse(m+1) |
---|
879 | MQtot = masse(m+1)*q(m+1) |
---|
880 | if (m.gt.0) then ! because some compilers will have problems |
---|
881 | ! evaluating masse(0) |
---|
882 | do while ((m.gt.0).and.(-w(l+1).gt.(Mtot+masse(m)))) |
---|
883 | m=m-1 |
---|
884 | Mtot = Mtot + masse(m+1) |
---|
885 | MQtot = MQtot + masse(m+1)*q(m+1) |
---|
886 | if (m.eq.0) exit |
---|
887 | end do |
---|
888 | endif |
---|
889 | if (m.gt.0) then |
---|
890 | sigw=(w(l+1)+Mtot)/masse(m) |
---|
891 | qm(l+1)= (-1/w(l+1))*(MQtot + (-w(l+1)-Mtot)* & |
---|
892 | (q(m)-0.5*(1.+sigw)*dzq(m)) ) |
---|
893 | else |
---|
894 | qm(l+1)= (-1/w(l+1))*(MQtot + (-w(l+1)-Mtot)*qm(1)) |
---|
895 | end if |
---|
896 | end if |
---|
897 | enddo |
---|
898 | |
---|
899 | return |
---|
900 | end subroutine vl1d |
---|