WRF - Historial de revisiones

Lluis.fita: /* netcdf2wps */

2026-09-07T12:54:44Z

netcdf2wps

← Revisión anterior		Revisión del 12:54 7 sep 2026
Línea 353:		Línea 353:
	'''How to proceed:'''		'''How to proceed:'''
	As the tool substitutes <CODE>ungrib</CODE>, make sense to keep using the standard folder name.		As the tool substitutes <CODE>ungrib</CODE>, make sense to keep using the standard folder name.
	1. Where one wants to simulate (remember, always in <CODE>salidas/</CODE>), create a folder <CODE>ungrib</CODE>		1. Where one wants to simulate (remember, always in <CODE>salidas/</CODE>), create a folder <CODE>ungrib</CODE>. For example:
	<PRE style="shell">		<PRE style="shell">
	mkdir -p [ExpName]/ungrib		mkdir -p /home/[NameOfUser]/salidas/RUN/[ExpName]/ungrib
	cd [ExpName]/ungrib		cd /home/[NameOfUser]/salidas/RUN/[ExpName]/ungrib
	</PRE>		</PRE>

	1. Copy the namelist, the job and link the tool:		2. Copy the namelist, the job and link the tool:
	<PRE style="shell">		<PRE style="shell">
	cp /share/tools/RegIPSL/tools/nc2wps/namelist.nc2wps.ERA5-CIMA ./namelist.nc2wps		cp /share/tools/RegIPSL/tools/nc2wps/namelist.nc2wps.ERA5-CIMA ./namelist.nc2wps
Línea 366:		Línea 366:
	</PRE>		</PRE>

	1. Modify dates in the namelist. Take for example a simulation that covers the period: 20260628 - 20260703.		3. Modify dates in the namelist. Take for example a simulation that covers the period: 20260628 - 20260703.
	<PRE style="shell">		<PRE style="shell">
	~~cp /share/tools/RegIPSL/tools/nc2wps/namelist.nc2wps.ERA5-CIMA ./namelist.nc2wps~~
	sed -i 's/begchk1yr/2026/g' namelist.nc2wps		sed -i 's/begchk1yr/2026/g' namelist.nc2wps
	sed -i 's/begchk1mn/06/g' namelist.nc2wps		sed -i 's/begchk1mn/06/g' namelist.nc2wps
Línea 378:		Línea 377:
	</PRE>		</PRE>

	1. Modify values inside job (verify use of cores and nodes). It is advise to use less cores than time-steps to process. Now it is provided by using 1 node of 40 cores, but only parallelized using 16 cores:		4. Modify values inside job (verify use of cores and nodes). It is advise to use less cores than time-steps to process. Now it is provided by using 1 node of 40 cores, but only parallelized using 16 cores:
	<PRE style="shell">		<PRE style="shell">
	sed -i 's/[user]/[NameOfUser]/g' launch_nc2wps_par_intel.pbs		sed -i 's/[user]/[NameOfUser]/g' launch_nc2wps_par_intel.pbs
	</PRE>		</PRE>

	It is also required to modify the folder where the output files will be kept. The job has been programed to move all the output to another folder than the one where the program run. Otherwise, all will be removed		It is also required to modify the folder where the output files will be kept. The job has been programed to move all the output to another folder than the one where the program run. Otherwise, all will be removed. For example:
	<PRE style="shell">		<PRE style="shell">
	- odir:		odir=/home/[NameOfUser]/salidas/EXPS/[ExpName]/ungrib
	</PRE>		</PRE>

	1. Launch the script as usual:		5. Launch the script as usual:
	<PRE style="shell">		<PRE style="shell">
	qsub launch_nc2wps_par_intel.pbs		qsub launch_nc2wps_par_intel.pbs
	</PRE>		</PRE>

	1. At the end of the exectuion the files <CODE>WRF4L_nc2wps_par.o[numJOBid], netcdf2wps_par_[nnnn]</CODE> will be created where information of the execution is kept. Revise them in case the files are not where they should be ...		6. At the end of the exectuion the files <CODE>WRF4L_nc2wps_par.o[numJOBid], netcdf2wps_par_[nnnn]</CODE> will be created where information of the execution is kept. Revise them in case the files are not where they should be ...

	'''NOTE:''' when using <CODE>nc2wps</CODE>, the names of the output files will be FILE3H1, one needs to modify this value in the <CODE>'namelist.wps'</CODE>, otherwise <CODE>metgrid</CODE> will not get them.		'''NOTE:''' when using <CODE>nc2wps</CODE>, the names of the output files will be FILE3H1, one needs to modify this value in the <CODE>'namelist.wps'</CODE>, otherwise <CODE>metgrid</CODE> will not get them.

Lluis.fita: /* netcdf2wps */

2026-09-07T12:51:05Z

netcdf2wps

← Revisión anterior		Revisión del 12:51 7 sep 2026
Línea 328:		Línea 328:
	==== netcdf2wps ====		==== netcdf2wps ====
	This is a tool to directly use netCDF files to create the outputs from <CODE>ungrib</CODE>. Is a tool written in Fortran parallelized along the time axis.		This is a tool to directly use netCDF files to create the outputs from <CODE>ungrib</CODE>. Is a tool written in Fortran parallelized along the time axis.

	[https://gitlab.in2p3.fr/ipsl/lmd/intro/regipsl/regipsl/-/wikis/Tools/nc2ps https://gitlab.in2p3.fr/ipsl/lmd/intro/regipsl/regipsl/-/wikis/Tools/nc2ps]		[https://gitlab.in2p3.fr/ipsl/lmd/intro/regipsl/regipsl/-/wikis/Tools/nc2ps https://gitlab.in2p3.fr/ipsl/lmd/intro/regipsl/regipsl/-/wikis/Tools/nc2ps]

Línea 335:		Línea 336:
	</PRE>		</PRE>

	<CODE>nc2wps </CODE> uses a ~~utiliza una~~ namelist ~~para configurarse. Ya se encuentra compilada con intel en hydra acá~~:		<CODE>nc2wps</CODE> it is already compiled for intel in <CODE>hydra</CODE> and uses a namelist as for the configuration:
			<PRE style="shell">
	/share/tools/RegIPSL/tools/nc2wps/netcdf2wps_par		/share/tools/RegIPSL/tools/nc2wps/netcdf2wps_par
			</PRE>

	~~Hay un ejemplo de~~ namelist ~~preparada para los~~ ERA5 ~~del CIMA disponible acá~~:		There is an example of namelist for ERA5 as they are in hydra:
			<PRE style="shell">
	/share/tools/RegIPSL/tools/nc2wps/namelist.nc2wps.ERA5-CIMA		/share/tools/RegIPSL/tools/nc2wps/namelist.nc2wps.ERA5-CIMA
			</PRE>

	~~Para correr la herramienta~~, ~~cómo está paralelizada~~ (~~en el eje de tiempo~~) ~~se necesita un~~ job~~, que ya tenés uno preparado acá~~:		In order to run the tool, as it is parallelized (along the time axis) a PBS job is required. There is already a prepared script for hydra available from here:
			<PRE style="shell">
	/share/tools/workflows/direct/WRF/hydra/launch_nc2wps_par_intel.pbs		/share/tools/workflows/direct/WRF/hydra/launch_nc2wps_par_intel.pbs
			</PRE>

	~~Cómo proceder~~: ~~Cómo la herramienta simplemente sustituye a~~ ungrib, ~~yo recomiendo usarla de la siguiente manera:~~		'''How to proceed:'''
	1.~~- En donde se quiera simular~~ (~~recordá~~, ~~siempre en~~ salidas/), ~~crear un directorio '~~ungrib'		As the tool substitutes <CODE>ungrib</CODE>, make sense to keep using the standard folder name.
	mkdir ungrib		1. Where one wants to simulate (remember, always in <CODE>salidas/</CODE>), create a folder <CODE>ungrib</CODE>
	cd ungrib		<PRE style="shell">
			mkdir -p [ExpName]/ungrib
			cd [ExpName]/ungrib
			</PRE>

	2.~~- Copiar el ejemplo de~~ namelist, el job ~~y linkar el programa compilado~~:		1. Copy the namelist, the job and link the tool:
			<PRE style="shell">
	cp /share/tools/RegIPSL/tools/nc2wps/namelist.nc2wps.ERA5-CIMA ./namelist.nc2wps		cp /share/tools/RegIPSL/tools/nc2wps/namelist.nc2wps.ERA5-CIMA ./namelist.nc2wps
	cp /share/tools/workflows/direct/WRF/hydra/launch_nc2wps_par_intel.pbs ./		cp /share/tools/workflows/direct/WRF/hydra/launch_nc2wps_par_intel.pbs ./
	ln -s /share/tools/RegIPSL/tools/nc2wps/netcdf2wps_par		ln -s /share/tools/RegIPSL/tools/nc2wps/netcdf2wps_par
			</PRE>

	3.~~- Modificar las fechas dentro de la~~ namelist. ~~Pongamos por ejemplo que tienes una simlación que cubre 2 meses~~: de 20260628 a 20260703.		1. Modify dates in the namelist. Take for example a simulation that covers the period: 20260628 - 20260703.
			<PRE style="shell">
	cp /share/tools/RegIPSL/tools/nc2wps/namelist.nc2wps.ERA5-CIMA ./namelist.nc2wps		cp /share/tools/RegIPSL/tools/nc2wps/namelist.nc2wps.ERA5-CIMA ./namelist.nc2wps
	sed -i 's/begchk1yr/2026/g' namelist.nc2wps		sed -i 's/begchk1yr/2026/g' namelist.nc2wps
Línea 363:		Línea 376:
	sed -i 's/endmetyr-endmetmn-endmetda_endmetho:endmetmi:endmetss/2026-07-03_00:00:00/g' namelist.nc2wps		sed -i 's/endmetyr-endmetmn-endmetda_endmetho:endmetmi:endmetss/2026-07-03_00:00:00/g' namelist.nc2wps
	sed -i 's/INfreqdata/10800/g' namelist.nc2wps		sed -i 's/INfreqdata/10800/g' namelist.nc2wps
			</PRE>

	4.~~- Modificar los valores dentro del~~ job ~~para lanzar el nc2wps~~ (~~verificando el uso de~~ cores ~~y nodos~~). ~~Se recomienda no utilizar más~~ cores ~~que pasos de tiempo a procesar~~. ~~Ahora está puesto que tome todo un nodo de~~ 40 ~~procesadores y sólo paralelice con~~ 16 cores:		1. Modify values inside job (verify use of cores and nodes). It is advise to use less cores than time-steps to process. Now it is provided by using 1 node of 40 cores, but only parallelized using 16 cores:
	sed -i 's/[user]/~~blazquez~~/g' launch_nc2wps_par_intel.pbs		<PRE style="shell">
			sed -i 's/[user]/[NameOfUser]/g' launch_nc2wps_par_intel.pbs
			</PRE>

	~~También hace falta modificar el directorio de salida (donde se guardarán los archivos FILE)~~. ~~La script está programada para que no se guarden los archivos ahí donde corre el programa~~. ~~Si no~~, ~~se va a borrar todo~~		It is also required to modify the folder where the output files will be kept. The job has been programed to move all the output to another folder than the one where the program run. Otherwise, all will be removed
			<PRE style="shell">
	- odir:		- odir:
			</PRE>

	5.~~- Lanzar la~~ script ~~como de costumbre~~		1. Launch the script as usual:
			<PRE style="shell">
	qsub launch_nc2wps_par_intel.pbs		qsub launch_nc2wps_par_intel.pbs
			</PRE>

	6.~~- Se escribirá un archivo al final~~ WRF4L_nc2wps_par.o[numJOBid] ~~en donde aparecerá información de la ejecución~~. ~~Revisarlo por si no están los archivos donde tocaría~~...		1. At the end of the exectuion the files <CODE>WRF4L_nc2wps_par.o[numJOBid], netcdf2wps_par_[nnnn]</CODE> will be created where information of the execution is kept. Revise them in case the files are not where they should be ...

	~~NOTA~~: ~~Al usar el~~ nc2wps, ~~los FILE se van a llamar~~ FILE3H1, ~~se tiene que modificar la~~ 'namelist.wps' ~~con ese nuevo nombre de archivo~~, ~~si no no los va a encontrar~~.		'''NOTE:''' when using <CODE>nc2wps</CODE>, the names of the output files will be FILE3H1, one needs to modify this value in the <CODE>'namelist.wps'</CODE>, otherwise <CODE>metgrid</CODE> will not get them.

	=== metgrid ===		=== metgrid ===

Lluis.fita: /* ungrib */

2026-09-07T12:36:27Z

ungrib

@@ Línea 251: / Línea 251: @@
 === ungrib ===
 Prepare and unpack grib files from the GCM forcing
 <B>NOTE:</B> In case atmospheric forcing data is not in GRIB format, and it is in standard netCDF format, user will need to use the [https://gitlab.in2p3.fr/ipsl/lmd/intro/regipsl/regipsl/-/wikis/Tools/nc2ps nc2wps] tool. Which is already compiled in: <code>/share/tools/RegIPSL/tools/nc2wps</code>.
@@ Línea 323: / Línea 325: @@
 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
 </pre>
 === metgrid ===

Lluis.fita: /* WRF: known errors */

2026-02-23T19:58:16Z

WRF: known errors

← Revisión anterior		Revisión del 19:58 23 feb 2026
Línea 532:		Línea 532:

	This error indicates that condition not all of the necessary variables where found in the files generated by ungrib. A possible cause for this is that when generating the links of the ungrib input, not all the files (and so, variables) where selected. Be aware that running ungrib will still figure as "successful", but only a few variables will have been processed. It is suggested taking a look in the ungrib.exe log where a table showing variables found are marked with an "X".		This error indicates that condition not all of the necessary variables where found in the files generated by ungrib. A possible cause for this is that when generating the links of the ungrib input, not all the files (and so, variables) where selected. Be aware that running ungrib will still figure as "successful", but only a few variables will have been processed. It is suggested taking a look in the ungrib.exe log where a table showing variables found are marked with an "X".

			== not enough vertical levels ==
			Real can provide error if the vertical configuration is not properly set-up

			For example in namelist.input:
			<PRE>
			e_vert = 60,
			dzbot = 5.
			dzstretch_s = 1.11
			dzstretch_u = 1.10
			p_top_requested = 1000,
			</PRE>

			<PRE>
			using new automatic levels program
			1 5.000000 5.000000 0.9994049
			2 5.549500 10.54950 0.9987448 1.109900
			3 6.159329 16.70883 0.9980126 1.109889
			4 6.836097 23.54493 0.9972007 1.109877
			5 7.587133 31.13206 0.9963003 1.109863
			(...)
			58 1577.883 17065.10 0.1250521 1.100000
			59 1735.671 18800.77 0.1000477 1.100000
			d01 2018-11-10_00:00:00 You need one of four things:
			d01 2018-11-10_00:00:00 1) More eta levels: e_vert
			d01 2018-11-10_00:00:00 2) A lower p_top: p_top_requested
			d01 2018-11-10_00:00:00 3) Increase the lowest eta thickness: dzbot
			d01 2018-11-10_00:00:00 4) Increase the stretching factor: dzstretch_s or dzstretch_u
			d01 2018-11-10_00:00:00 All are namelist options
			-------------- FATAL CALLED ---------------
			FATAL CALLED FROM FILE: <stdin> LINE: 7709
			not enough eta levels to reach p_top
			-------------------------------------------
			</PRE>

			It works, just increasing <CODE>e_vert</CODE>
			<PRE>
			e_vert = 65,
			</PRE>

	= Additional information =		= Additional information =

Lluis.fita: /* WRF4L: Lluís' WRF work-flow management */

2025-07-31T11:11:30Z

WRF4L: Lluís' WRF work-flow management

← Revisión anterior		Revisión del 11:11 31 jul 2025
Línea 539:		Línea 539:
	For information about Lluí's WRF work-flow management visit here [[WRF4L]]		For information about Lluí's WRF work-flow management visit here [[WRF4L]]

	[[WRF4Lold]]		<!-- [[WRF4Lold]] -->

	= WRF URBAN =		= WRF URBAN =

Lluis.fita: /* WRF4L: Lluís' WRF work-flow management */

2025-07-31T08:51:57Z

WRF4L: Lluís' WRF work-flow management

← Revisión anterior		Revisión del 08:51 31 jul 2025
Línea 538:		Línea 538:
	= WRF4L: Lluís' WRF work-flow management =		= WRF4L: Lluís' WRF work-flow management =
	For information about Lluí's WRF work-flow management visit here [[WRF4L]]		For information about Lluí's WRF work-flow management visit here [[WRF4L]]

			[[WRF4Lold]]

	= WRF URBAN =		= WRF URBAN =

Lluis.fita: /* Move data to Papa */

2025-06-27T19:00:15Z

Move data to Papa

← Revisión anterior		Revisión del 19:00 27 jun 2025
Línea 482:		Línea 482:
	In order to move the outputs to Papa, it is necessary to provide access to Lluís (by now, the data manager of Papa).		In order to move the outputs to Papa, it is necessary to provide access to Lluís (by now, the data manager of Papa).

	To do that, let's say that the outputs are in <CODE>$odir</CODE> the user that ran the simulation should do:		To do that, let's say that the outputs are in <CODE>$odir</CODE> (the same as in <CODE>launch_intel_wrf.pbs</CODE>) the user that ran the simulation should do:
	<PRE STYLE="shell">		<PRE STYLE="shell">
	$ chmod a+x $odir		$ chmod a+x $odir
Línea 488:		Línea 488:
	</PRE>		</PRE>

	Notify Lluís about that vi an email.		Notify Lluís about that via an email, telling that in <CODE>$odir</CODE> there is a simulation to be copied to Papa.

	= WRF: known errors =		= WRF: known errors =

Lluis.fita: /* WRF */

2025-06-27T18:59:00Z

WRF

@@ Línea 478: / Línea 478: @@
 Timing for main: time [YYYY]-[MM]-[DD]_[HH]:[MI]:[SS] on domain  1:  3.86105 elapsed seconds
 </pre>
 = WRF: known errors =

Lluis.fita: /* WRF */

2025-06-24T18:24:12Z

WRF

← Revisión anterior		Revisión del 18:24 24 jun 2025
Línea 455:		Línea 455:
	</pre>		</pre>

	* If everything went fine one should have (the basic ones):		* If everything went fine one should have in <code>odir</code> (the basic ones):

	<pre>outwrf/wrfout_d[nn]_[YYYY]-[MM]-[DD]_[HH]:[MI]:[SS] outwrf/wrfrst_d[nn]_[YYYY]-[MM]-[DD]_[HH]:[MI]:[SS] ...		<pre>outwrf/wrfout_d[nn]_[YYYY]-[MM]-[DD]_[HH]:[MI]:[SS] outwrf/wrfrst_d[nn]_[YYYY]-[MM]-[DD]_[HH]:[MI]:[SS] ...

Lluis.fita: /* WRF */

2025-06-24T18:23:45Z

WRF

← Revisión anterior		Revisión del 18:23 24 jun 2025
Línea 450:		Línea 450:
	* Simulation (look on [http://www2.mmm.ucar.edu/wrf/users/docs/user_guide_V3/users_guide_chap5.htm\#_Description_of_Namelist description of namelist] for namelist configuration/specifications).		* Simulation (look on [http://www2.mmm.ucar.edu/wrf/users/docs/user_guide_V3/users_guide_chap5.htm\#_Description_of_Namelist description of namelist] for namelist configuration/specifications).

	* Getting the necessary PBS job (same folder for <code>real</code>)		* Getting the necessary PBS job (same folder for <code>real</code>), remember to change <CODE>idir</CODE>, <CODE>odir</CODE>, the mail as well as the <code>nodes, ppn, Nprocs</code>
	<pre style="shell">$ cp /share/WRF/launch_wrf_intel.pbs ./		<pre style="shell">$ cp /share/tools/workflows/direct/WRF/hydra/launch_wrf_intel.pbs ./
	qsub launch_wrf_intel.pbs		qsub launch_wrf_intel.pbs
	</pre>		</pre>