用4.5版本做区域GLDAS驱动数据问题求助

[12wlj]

你好，我在用hrldas-v4.5_master/hrldas/HRLDAS_forcing做GLDAS驱动数据时候，设置好namelist.input.GLDAS，运行create_forcing.exe，发生如图所示的报错情况。我检查了GLDAS_Qair_f_inst_2021010100是存在Qair_f_inst变量的，是Can't read variable: Q_f_inst的问题吗，

[cenlinhe]

did you follow up the instruction here:

hrldas/hrldas/docs/README.GLDAS

Lines 42 to 176 in 7d183b9

	3. Prepare GLDAS-based forcing data.
	In GLDASV2.1 are in netCDF format. The preprocesse codes are written in python and some modifications are added to create_forcing.F
	3.1 Get the GLDAS raw files. GLDAS version 2.1 uses netcdf format.
	They can be retrieved from here:
	https://disc.gsfc.nasa.gov/datasets/GLDAS_NOAH025_3H_2.1/summary
	I put these "raw" GLDAS files in one directory (e.g., /raw_GLDAS_data_path)
	--------------------------------------------------
	3.3 extract_gldas.py: extract the individual variables from the GLDAS netcdf files
	The current script uses nco(ncks) to extract individual fields from the GLDAS files. It will process entire years or sections of one year. You can change this by modifying the python script.
	Change these two paths for where raw GLDAS files are located and where separated(extracted) files will go.
	data_dir = "/raw_GLDAS_data_path";
	results_dir = "/extracted_GLDAS_data_path";
	For convenience, you may want to create directories in the results_dir for each variable and a few more for processed fields.
	`cd /extracted_GLDAS_data_path`
	`mkdir Rainf/ Snowf/ Wind/ Tair/ Qair/ Psurf/ SWdown/ LWdown/ SWdown24/ Precip/ U/ V/ INIT/`
	Run the python code:
	`python extract_glads.py`
	--------------------------------------------------
	3.4 extract_gldas_init.py: extract the initial-only fields (SWE_inst,Canopint_inst,AvgSurfT_inst,SoilMoi,SoilTMP) from the GLDAS NETCDF output files
	Change the date of initialization and run the script one time. The files should appear in the /extracted_GLDAS_data_path/INIT/ directory.
	Note: this step is not necessary unless you want to intialize using the GLDAS states
	Note: GLDAS soils are inverted (level 4 is actually level 1); this script takes care of that
	Run the python code:
	`python extract_glads_init.py`
	--------------------------------------------------
	3.5 intermediate clean-up (optional)
	If you are space-limited, at this point you can remove the raw GLDAS files. You have extracted all you need and they are no longer needed.
	--------------------------------------------------
	3.6 create_UV_netcdf.perl: split Wind to U, V; just set V=0 (GLDAS has wind speed, HRLDAS wants components)
	Again, you will need to change the paths in the Perl script along with the year/day ranges that are appropriate for your application.
	--------------------------------------------------
	3.7 combine_precips_netcdf.perl: combine precipitations to a total precipitation
	Again, you will need to change the paths in the Perl script along with the year/day ranges that are appropriate for your application.
	--------------------------------------------------
	# I didn't do this part, the forcing I used are 3-hourly
	#3.8 fill_SSRD.perl: do hourly interpolation of solar radiation via zenith angle
	#
	# - note original data are instantaneous 3-hourly
	#
	#Again, you will need to change the paths in the Perl script along with the year/day ranges that are appropriate for your application.
	#
	#There is an NCL script (check_SW.ncl) if you want to see what the interpolation is doing.
	#--------------------------------------------------
	3.6 Run the HRLDAS pre-processor code.
	At this point, you should have all the GLDAS output extracted into different directories at /extracted_GLDAS_data_path/. Now it's finally time to run the first step of the HRLDAS.
	cd /root_path/HRLDAS-v3.x/HRLDAS_COLLECT_DATA/run
	In the examples/GLDAS/ directory there is a namelist.input file that is consistent with the GLDAS procedure described above. You can modify this file for your case and place it in the pre-processor run directory:
	cp examples/GLDAS/namelist.input.GLDAS .
	--------------------------------------------------
	3.9.1 Modify the namelist
	Note: I have made changes to this namelist here. Please see an updated version in /HRLDAS_forcing/run/example/GLDAS/namelist.input.GLDAS
	Modifying the namelist.input.GLDAS to your case. The following is a short description of each line in the file. A longer description is available in README.namelist
	STARTDATE : start date of your simulation
	ENDDATE : end date of your simulation
	DATADIR : base directory for your GLDAS GRIB files (will be substituted in the paths below)
	OUTPUTDIR : where you want the forcing files to be placed
	FULL_IC_FRQ : set to 0 if you want initial conditions in the init file, -1 otherwise
	RAINFALL_INTERP : rainfall interpolation procedure ! 0=Nearest Neighbor; 1=more expensive, grid fill method
	RESCALE_SHORTWAVE : set to .FALSE. for GLDAS
	UPDATE_SNOW : do you want to read in snow during the simulation?
	FORCING_HEIGHT_2D : set to .FALSE. for GLDAS
	TRUNCATE_SW : set to .FALSE. for GLDAS
	EXPAND_LOOP : the number of loops to fill missing GLDAS data (e.g., soil over ocean). Set to at least 1 for GLDAS, may need to set higher if, for example, you have islands in your WRF grid that are not resolved by GLDAS.
	INIT_LAI : .true. to put LAI in the init file
	VARY_LAI : .true. to put time-varying LAI in the forcing files
	MASK_WATER : .true. to put missing values over water points (could reduce size if you compress later)
	geo_em_flnm = path and name of your geo_em file
	Zfile_template = template for your GLDAS terrain height file
	LANDSfile_template = template for your GLDAS land-sea file
	Tfile_template = template for your GLDAS temperature file
	Ufile_template = template for your GLDAS u-wind file
	Vfile_template = template for your GLDAS v-wind file
	Pfile_template = template for your GLDAS pressure file
	Qfile_template = template for your GLDAS q file
	LWfile_template = template for your GLDAS longwave file
	SWfile_primary = template for your GLDAS primary shortwave file
	SWfile_secondary = template for your GLDAS secondary shortwave file
	PCPfile_primary = template for your GLDAS primary precipitation file
	PCPfile_secondary = template for your GLDAS secondary precipitation file
	WEASDfile_template = template for your GLDAS SWE file (not necessary if you don't want initial conditions)
	CANWTfile_template = template for your GLDAS canopy water file (not necessary if you don't want initial conditions)
	SKINTfile_template = template for your GLDAS skin temperature file (not necessary if you don't want initial conditions)
	STfile_template = template for your GLDAS soil temperature file (not necessary if you don't want initial conditions)
	SMfile_template = template for your GLDAS soil moisture file (not necessary if you don't want initial conditions)
	Note: for GLDAS there will be no need to include a secondary SW and precipitation file, so just set them to be the same
	3.9.2 Run the pre-processor
	Note: please use create_forcing_netcdf.F for this purpose.
	`cp create_forcing_netcdf.F create_forcing.F`
	and compile again.
	Run the pre-processor with the GLDAS namelist
	./create_forcing.exe namelist.input.GLDAS

?

there are two additional discussions for references
#91
#103