r2237-r2291
Commissions entre r2237 et r2291
------------------------------------------------------------------------
r2291 | emillour | 2015-06-03 12:59:18 +0200 (Wed, 03 Jun 2015) | 3 lines
Added utility routines cbrt to compute cubic roots (used by some planetary physics packages) and getin_p (to read from def files in parallel).
EM
------------------------------------------------------------------------
r2290 | musat | 2015-06-03 12:45:50 +0200 (Wed, 03 Jun 2015) | 3 lines
One more bug correction to frequency of pressure level outputs' by XIOS
AC/IM
------------------------------------------------------------------------
r2288 | musat | 2015-05-27 15:09:31 +0200 (Wed, 27 May 2015) | 2 lines
Oubli attribut detect_missing_value
------------------------------------------------------------------------
r2287 | crio | 2015-05-22 17:52:37 +0200 (Fri, 22 May 2015) | 1 line
------------------------------------------------------------------------
r2286 | emillour | 2015-05-20 15:27:07 +0200 (Wed, 20 May 2015) | 5 lines
- fix in dyn3d the array out of bound issue (in qminimum) that was only corrected in dyn3dmem in rev 2285.
- comment out many invasive debug writes.
EM
------------------------------------------------------------------------
r2285 | fhourdin | 2015-05-20 11:03:24 +0200 (Wed, 20 May 2015) | 4 lines
Correction de depassement de tableau
Bug fixing
------------------------------------------------------------------------
r2284 | jyg | 2015-05-19 18:33:30 +0200 (Tue, 19 May 2015) | 9 lines
Bug fixing concerning aerosol scavenging (bugs in cvltr_scav and in lsc_scav) and aerosol concentration output (the values at beginning, instead of end, of time step were output).
------------------------------------------------------------------------
r2282 | musat | 2015-05-18 14:27:29 +0200 (Mon, 18 May 2015) | 2 lines
Correction sorties pression par XIOS dans le cas ok_all_xml
------------------------------------------------------------------------
r2281 | crisi | 2015-05-18 11:09:37 +0200 (Mon, 18 May 2015) | 3 lines
Camille Risi: corrections of bugs for the isotopic part
------------------------------------------------------------------------
r2280 | musat | 2015-05-15 18:57:58 +0200 (Fri, 15 May 2015) | 4 lines
Oooops
Put operation="average" instead of instant
------------------------------------------------------------------------
r2279 | musat | 2015-05-15 18:48:41 +0200 (Fri, 15 May 2015) | 4 lines
Add detect_missing_value for outputs by XIOS of
the variables at standard level pressures (850hPa,
700hPa, etc)
------------------------------------------------------------------------
r2278 | fhourdin | 2015-05-15 18:06:07 +0200 (Fri, 15 May 2015) | 3 lines
Modification sur la prise en compte de la gustinness
------------------------------------------------------------------------
r2277 | fhourdin | 2015-05-15 16:24:34 +0200 (Fri, 15 May 2015) | 4 lines
Ajout d'une instruction RECURSIVE pour les routines d'advection
qui s'appellent elles-même.
------------------------------------------------------------------------
r2273 | fairhead | 2015-05-13 12:35:29 +0200 (Wed, 13 May 2015) | 10 lines
Ooooops
Correction bug introduit dans r2225 qui menait au plantage du routage dans ORCHIDEE
----------------------------------------------------------------
Correcting bug introduced in r2225 that lead to ORCHIDEE crashing in initialisation
of the routing
------------------------------------------------------------------------
r2271 | musat | 2015-05-12 18:27:47 +0200 (Tue, 12 May 2015) | 5 lines
Corrections for standard level pressure outputs hist*NMC.nc by XIOS and IOIPSL. For XIOS one need to specify a default_value="1.e+20" in the field_def_lmdz.xml and a detect_missing_value=".true." in the file_def_histins_lmdz.xml, file_def_histday_lmdz.xml, etc files.
------------------------------------------------------------------------
r2270 | crisi | 2015-05-07 17:45:04 +0200 (Thu, 07 May 2015) | 4 lines
Adding isotopes in the dynamics and more generally tracers of tracers.
CRisi
------------------------------------------------------------------------
r2269 | emillour | 2015-04-30 16:25:35 +0200 (Thu, 30 Apr 2015) | 3 lines
Typos in variable names which lead to the makefile being systematically recreated when compiling.
EM
------------------------------------------------------------------------
r2268 | idelkadi | 2015-04-30 11:12:46 +0200 (Thu, 30 Apr 2015) | 2 lines
Mise a jour et corrections pour les sorties Cosp via Xios.
------------------------------------------------------------------------
r2267 | fhourdin | 2015-04-23 02:07:44 +0200 (Thu, 23 Apr 2015) | 15 lines
Modification du modèle du thermique pour diminuer entre autres la sensibilité au pas de temps. Convergence numérique si iflag_thermals_ed=7.
Ne concerne que la version iflag_thermals_ed=8
Premier essaie de prise en compte d'une variabilité verticale sous maille de l'eau pour accroître la converture nuageuse.
Sous la clé iflag_coudth_vert=1 (1 par défaut)
Modification of the thermal plume model in case iflag_thermals_ed=8
Modification of the bigaussian scheme for clouds associated with
thermals. Activated if iflag_cloudth_vert=1
------------------------------------------------------------------------
r2266 | lguez | 2015-04-22 15:30:47 +0200 (Wed, 22 Apr 2015) | 3 lines
Avoid cryptic compilation errors when trying to create ce0l in parallel: give a clear message in compilation script instead.
------------------------------------------------------------------------
r2265 | jyg | 2015-04-18 17:06:36 +0200 (Sat, 18 Apr 2015) | 13 lines
Bug fixes concerning the number "nqo" of water phases in "q" array: part II (sequel of fixes of revision 2262).
Modified files:
carbon_cycle_mod.F90
ini_histrac.h
phyredem.F90
phyetat0.F90
traclmdz_mod.F90
write_histrac.h
------------------------------------------------------------------------
r2264 | fhourdin | 2015-04-17 22:25:14 +0200 (Fri, 17 Apr 2015) | 3 lines
Bug fixing
------------------------------------------------------------------------
r2263 | lguez | 2015-04-17 20:14:36 +0200 (Fri, 17 Apr 2015) | 12 lines
In the sequential version, ini_getparam and fin_getparam were not called so guide_init created a file "fort.99". Added call to ini_getparam and fin_getparam in guide_init. The created file is now "nudging_parameters_out.txt".
In the parallel version, ini_getparam was called from gcm, with the argument "out.def". So "out.def" was created even without nudging, it remained empty. Moved the call to ini_getparam from gcm to guide_init. Also changed the name of the created file to "nudging_parameters_out.txt", as in the sequential version. "out.def" was too vague and confusing a name.
------------------------------------------------------------------------
r2262 | jyg | 2015-04-17 16:34:43 +0200 (Fri, 17 Apr 2015) | 11 lines
Correction of a bug concerning the number 'nqo' of water phases transported by the dynamic : the default value (= 2, corresponding to vapour and liquid phases) was still explicitely present in various places.
Modified files:
infotrac.F90,
physiq.F90
------------------------------------------------------------------------
r2261 | fhourdin | 2015-04-17 16:23:39 +0200 (Fri, 17 Apr 2015) | 13 lines
Nouvelle option pour les drag océaniques séparant z0m, z0h, et z0q pour la quantité de mouvement, la température et l'humidité.
La séparation z0m/z0h est activée par la clée iflag_z0_oce==1.
Par ailleurs, on impose le rapport z0q/z0h = f_z0qh_oce Typiquement, f_z0qh_oce=1.55
New option to separate z0m, z0h and z0q z0m and z0h differ if iflag_z0_oce==1.
f_z0qh_oce = z0q/z0h is a constant factor
Alina and Frederic
------------------------------------------------------------------------
r2260 | fhourdin | 2015-04-16 16:22:45 +0200 (Thu, 16 Apr 2015) | 4 lines
Correction.
Bug fixing
------------------------------------------------------------------------
r2259 | jyg | 2015-04-15 12:51:04 +0200 (Wed, 15 Apr 2015) | 12 lines
Changes in Emanuel's deep convection scheme: the upper bound of deep convection loops is set at the first level above 22 km.
Modified files:
physiq.F90,
concvl.F90,
cva_driver.F90,
cv3a_compress.F90,
cv3a_uncompress.F90,
cv3_routines.F90
------------------------------------------------------------------------
r2257 | jyg | 2015-04-07 11:47:46 +0200 (Tue, 07 Apr 2015) | 1 line
Improvement of three loops in cv3_undilute2 in cv3_routines.F90
------------------------------------------------------------------------
r2256 | fairhead | 2015-04-03 16:45:26 +0200 (Fri, 03 Apr 2015) | 3 lines
New versions of xml XIOS files for variables that have been added lately to phys_output_ctrlout_mod.F90
------------------------------------------------------------------------
r2255 | jyg | 2015-04-03 15:51:31 +0200 (Fri, 03 Apr 2015) | 11 lines
Changes to pbl_surface and other routines concerning split/no-split.
+ pbl_surface_mod.F90: call cdrag for (w) region.
+ phyredem.F90: write wake_delta_pbl_TKE.
+ phys_output_write_mod.F90: control output of wake_delta_pbl_TKE by
IF(iflag_pbl_split>=1).
+ lmdz1d.F90: initialize wake_delta_pbl_TKE=0.
+ phys_output_ctrlout_mod.F90: suppression of accents in some variable
attributes.
+ cva_driver.F90: suppression of a print introduced in version 2253.
------------------------------------------------------------------------
r2254 | fhourdin | 2015-04-03 12:16:36 +0200 (Fri, 03 Apr 2015) | 14 lines
Modification du calcul des flux air/mer
1) Introduction d'un facteur f_qsat_oce=0.98 devant qsat dans le calcul de l'évaporation sur océan pour tenir compte de la moindre évaporation de l'eau salée.
2) Introduction d'une différentiation entre z0 pour le sensible, z0h, et le latent, z0q, imposé constant z0q=f_z0qh_oce*z0h
Modification of air/sea fluxes computation
1) Introduction of a correcting factor f_qsat_oce=0.98 on qsat to account for the weaker evaporation of salty water.
2) Introduction of z0q=f_z0qh_oce*z0h
------------------------------------------------------------------------
r2253 | jyg | 2015-03-30 11:08:45 +0200 (Mon, 30 Mar 2015) | 27 lines
1/ Introduction of two variables in the ".def" files:
(i) cvl_sig2feed is the top of the convective feeding layer in sigma coordinates (D=0.97);
(ii) cvl_comp_threshold is the threshold fraction of convective points below which compression occurs (D=1.).
2/ Corrections of various bugs revealed by the changes in compression:
- correct bugs in cv3a_uncompress.F90 for 3 fields used for convective scavenging.
- add a reset to zero of "sig" and "w0" for non-convective points (cva_driver.F90).
- in cv3_routines.F90, correct bounds of a few loops in cv3_undilute2, correct the reset of the no-convection counter in cv3_yield.
- in phys_output_write_mod.F90, correct output of wdtrainA and wdtrainM.
3/ Improve declarations in various subroutines.
Modified files:
conema3.h
cv3param.h
cv3p1_closure.F90
conf_phys_m.F90
cv3a_compress.F90
phys_output_write_mod.F90
cv3_routines.F90
concvl.F90
cva_driver.F90
cv3a_uncompress.F90
------------------------------------------------------------------------
r2252 | fhourdin | 2015-03-27 17:35:52 +0100 (Fri, 27 Mar 2015) | 4 lines
Poursuite du nettoyage de phyetat0, et retour à 1+1=2.
Clean and clean, there will always remain soemthing out of it.
------------------------------------------------------------------------
r2251 | fhourdin | 2015-03-26 18:28:25 +0100 (Thu, 26 Mar 2015) | 4 lines
Suite du nettoyage dans phyetat0
Keep on cleaning and cleaning ...
------------------------------------------------------------------------
r2250 | emillour | 2015-03-26 16:28:42 +0100 (Thu, 26 Mar 2015) | 4 lines
Improving the makelmdz* utilities:
Directory phy*** in dynlonlat_phylonlat is assumed to be the same "root" as provided by the name of the physics package (e.g. "phylmd" for "phylmd_bis" or "phylmdmine" or "phylmd.2", etc.) to ease things for those who work with multiple copies of a physics directory.
FH
------------------------------------------------------------------------
r2249 | lguez | 2015-03-25 19:17:55 +0100 (Wed, 25 Mar 2015) | 1 line
abort_message was not defined.
------------------------------------------------------------------------
r2248 | lguez | 2015-03-25 19:04:54 +0100 (Wed, 25 Mar 2015) | 4 lines
Correcting an imbroglio from revision 2247. Directory dynlmdz_phylmd does not exist any longer, and ce0l.F90 should have been modified in directory dynlonlat_phylonlat/phylmd.
------------------------------------------------------------------------
r2247 | lguez | 2015-03-25 18:24:15 +0100 (Wed, 25 Mar 2015) | 6 lines
We want to keep the same "*.def" files with programs ce0l and gcm. But there is only a sequential version of program ce0l, which has not the FFT filter. So we have to ignore the setting of use_filtre_fft in program ce0l. Moved the test on use_filtre_fft from procedure conf_gcm to main units ce0l and gcm.
------------------------------------------------------------------------
r2246 | fhourdin | 2015-03-25 12:09:58 +0100 (Wed, 25 Mar 2015) | 4 lines
Petite correction pour tourner avec contrôle de dépassement de tableau small bug fixing
------------------------------------------------------------------------
r2245 | fhourdin | 2015-03-24 19:30:09 +0100 (Tue, 24 Mar 2015) | 4 lines
Initialisation alp_bl_conv=0 pour resoudre un plantage en 1D bug fixing
------------------------------------------------------------------------
r2244 | fhourdin | 2015-03-24 15:17:55 +0100 (Tue, 24 Mar 2015) | 4 lines
Correction pour le 1D + toilettage
Bugfixing for 1D + cleaning
------------------------------------------------------------------------
r2243 | fhourdin | 2015-03-24 14:28:51 +0100 (Tue, 24 Mar 2015) | 34 lines
Revisite de la formule des flux de surface (en priorité sur l'océan) en tenant compte des bourrasques de vent et de la différence entre les hauteurs de rugosités pour la quantité de mouvement, l'enthalpie et éventuellement l'humidité.
Etape 2 :
- Séparation des z0 pour la quantité de mouvement et l'enthalpie. rugs (ou frugs, rugos, yrugos ...) disparait au profit de z0m, z0h. Les variables qui étaient à la fois dans pbl_surface_init et dans l'interface de pbl_surface sont suprimées de pbl_surface_init. On travaille directement pour ces variables (evap, z0, qsol, agesno) avec les versions de phys_state_var_mod (qui étaient précédemment dans phys_local_var_mod
- Nouveaux paramètres de contrôle :
* iflag_z0_oce (par défaut 0, et seule option active jusque là)
* z0m_seaice_omp, z0h_seaice_omp, comme leur nom l'indique utilisées dans surf_landice
* z0min appliqué sur z0m et z0h dans pbl_surface
- Introduction des fonction phyeta0_get et phyetat0_srf pour lire les conditions de initiales dans startphy.
Du coup une seule ligne suffit pour lire et contrôler d'éventuels problèmes.
- Pour la variable fxrugs, elle est remplacée par z0m(:,nbsrf+1)
Ce choix déjà utilisé pour d'autres variables pourrait être systématiser pour alléger l'interface de pbl_surface_mod.
- Dans les sorties, les variables rugs* ont été remplacées par des z0m* et z0h*
- Nettoyage des anciens alb1/alb2 dans les lectures/écritures des états de redémarrage (et dans pbl_surface_mod.F90).
------------------------------------------------------------------------
r2242 | emillour | 2015-03-24 09:08:43 +0100 (Tue, 24 Mar 2015) | 6 lines
Further reorganizing of the physics-dynamics interface.
Things related to dynamics grid and physics grid go in "dynlonlat_phylonlat".
Any physics phy*** related stuff (e.g. iniphysiq, ce0l) goes in a phy*** subdirectory of dynlonlat_phylonlat.
EM
------------------------------------------------------------------------
r2241 | fhourdin | 2015-03-23 22:02:32 +0100 (Mon, 23 Mar 2015) | 8 lines
Nettoyage des anciens albedo. Elimination de alb1 et alb2 dans pbl_surface (il s'agissait de commentaires) et dans le etats de démarrage.
Some cleaning of old albedo specification (alb1/alb2)
------------------------------------------------------------------------
r2240 | fhourdin | 2015-03-23 17:07:43 +0100 (Mon, 23 Mar 2015) | 21 lines
Revisite de la formule des flux de surface (en priorité sur l'océan) en tenant compte des bourrasques de vent et de la différence entre les hauteurs de rugosités pour la quantité de mouvement, l'enthalpie et éventuellement l'humidité.
Etape 1 :
Introduction d'un calcul de gustiness dans la physique gustiness(:)=f_gust_bl * ale_bl + f_gust_wk * ame_wk Cette variable est passée ensuite jusqu'au fin fond de la couche limite. L'étape 1 est prête à commettre, ne nécessite pas de nouvelles variables dans les startphy et assure la convergence numérique.
Introduction of gustiness in the surface flux computation.
Gustiness is computed from as gustiness(:)=f_gust_bl * ale_bl + f_gust_wk * ame_wk and pass through pbl_surface down to the routines that compute surface fluxes.
------------------------------------------------------------------------
r2239 | emillour | 2015-03-23 08:27:30 +0100 (Mon, 23 Mar 2015) | 8 lines
Reorganizing physics/dynamics interface:
- what is related to dynamics-physics interface is now in a seperate directory: dynlmdz_phy*** for physics in phy***
- 1d model and related dependencies (including a couple from "dynamics", set up as symbolic links) is now in subdirectory "dyn1d" of phy***.
- "bibio" directory is now "misc" and should only contain autonomous utilities.
- "cosp" is now a subdirectory of phylmd.
EM
------------------------------------------------------------------------
r2238 | emillour | 2015-03-20 11:54:59 +0100 (Fri, 20 Mar 2015) | 3 lines
Update of "makelmdz" and "create_make_gcm" scripts so that only required directories are scanned when building the makefile.
EM
------------------------------------------------------------------------
r2237 | fhourdin | 2015-03-20 07:53:17 +0100 (Fri, 20 Mar 2015) | 6 lines
Prise en compte des nouveaux alebedo dans les fichiers de redémarrage.
Retour à 1+1=2
Taking into account new albedos in restart files. 1+1=2