XSUMO - panel help

Remote machine: This is the remote host on which SUMO will execute, eg. met.rdg.ac.uk, seymour.rdg.ac.uk, fuji.cfs.ac.uk, fermat.cfs.ac.uk etc.
Maximum memory (kwords): To run SUMO in batch mode on certain computers, you will need to specify the expected maximum memory usage in kilowords.
Maximum cputime (secs): To run SUMO in batch mode on certain computers, you will need to specify the expected maximum cpu time required by the job in seconds.
Remote userid: In order to use the XSUMO Submit button you will need to specify the userid on the remote host where the job is to run. Note that for most machines this use of Submit is disabled and hence you will not need to set this.
Local machine: This should be set to the address of the local host (for example, the computer you are currently using or, perhaps, another one on your network) where you will run the PRESUMO executable in order to create your final SUMO jobdeck/script.
SUMO home directory: This should specify the full pathname of the directory containing the SUMO nupdate program library on the remote host.
Orography ancillary data file directory: This is the full pathname of the ancillary data file containing the model orography - this must be specified if you wish to include orography in your analysis. It is typically required when model input data is available on model (eta) surfaces.
Directory containing data to be analysed: This is the directory on the remote host where the data (eg. UM output files) to be analysed by SUMO are stored.
Data type: Select the type corresponding to the type of data to be analysed by SUMO.
Number of longitude points: This corresponds to the number specified in the originating UM run (eg. 96 for a climate run in 1998) or in the dataset.
Number of latitude points (pole to equator): This should be set to the number of velocity gridpoints in a single hemisphere (eg. 36 in a climate run in 1998).
Number of hemispheres of data: Choose one value - note that SUMO will not work unless global data (2 hemispheres) are provided.
Number of input data levels: This is the number of full eta levels in the UM (if input data is on eta levels), or the number of pressure levels in the input dataset if the data is on pressure surfaces.
Number of output levels: This is the number of levels to which data is interpolated prior to output. It must be less than or equal to the number of input data levels.
Model contains orography: Select (red = yes) depending upon whether the input dataset was derived using surface orography (eg. a real earth). This can be turned off when input data is on pressure surfaces.
Pathname of RECONA file: This should specify the full pathname of the file RECONA that is appropriate for the originating UM job. It is required when the input fields are in PP format on model levels. However, if one is not specified then SUMO will compute model level information as best as it can.
Type of calendar in use: Select appropriate value.

Script

Script : Compilation/output panel

Compile source codes: Setting this to yes (red) will force compilation of SUMO each time it runs on the remote host. Otherwise, compilation will only occur is a pre-existing executable cannot be found. The latter approach will reduce the run time; however, you need to make sure that; SUMO is compiled if you are continouusly editing the source code.
Output type: Select one type of graphical output format. "no plot" will suppress any graphical output.
SUMO experiment identifier: Up to 8 characters may be used to specify a name for the current SUMO job. It does not have to be identical to the UM experiment id.

Script

Script : Directories panel

UTF directory: This is the full pathname of the directory on the remote host in which the UTF file will be saved after SUMO finishes executing.
UTF filename: This is the name to be given to the UTF file if produced. Do not include the ".utf" extension in the filename.
DRS directory: This is the full pathname of the directory on the remote host in which any DRS files will be saved after SUMO finishes executing.
netCDF directory: This is the full pathname of the directory on the remote host in which any netCDF files will be saved after SUMO finishes executing.
Work directory: This is the full pathname of the directory on the remote host in which SUMO will execute.

Data

Data : COMMON datablock panel

Initial data directory: This should currently be set to the full pathname of the directory (not file) containing the model orography file. If SUMO will not be using an orography file, then it can point to the work directory.
User home directory on remote host: This should point to your home directory on the remote host.
Directory containing data to analyse: This is the directory on the remote host where the data (eg. UM output files) to be analysed by SUMO are stored.
Analysis start date: A real value denoting the start date of the analysis period. This may be a positive number, in which case it will be the starting day number (corresponding to an elapsed time of a model integration) with any decimal fractions corresponding to the hour time - eg. 15.25 means 15 days and 6 hours into the model integration. Alternatively, the user may specify a negative number of the form -yymmdd, where yy are the last two digits of the year, mm is a two digit month (eg. 07 = July) and dd is a two digit day number.
Analysis end date: A real value denoting the end date of the analysis period. This may be a positive number, in which case it will be the ending day number (corresponding to an elapsed time of a model integration) with any decimal fractions corresponding to the hour time - eg. 15.25 means 15 days and 6 hours into the model integration. Alternatively, the user may specify a negative number of the form -yymmdd, where yy are the last two digits of the year, mm is a two digit month (eg. 07 = July) and dd is a two digit day number.
Analysis start century: Century of the start time of the analysis, eg 19 for years 1900 to 1999.
Analysis end century: Century of the end time of the analysis, eg 19 for years 1900 to 1999.
Analysis start hour: Hour time of the start of the analysis, eg 12 for 1200 GMT.
Analysis end hour: Hour time of the end of the analysis, eg 12 for 1200 GMT.
Number of days used in running means: Set to, eg., 3.0 to compute 3-day runningmeans. If running means are not required at output, then a value of 1.0 should be appropriate. If the duration of the analysis is less than 1.0 days, then enter a time equivalent to the duration of the analysis in days.
Diagnostic types required: Select one or more options.
Output surface type: Select one option. Do not select eta (model levels) if input data is not on model levels.
Enable output of printed analyses: Select if porinted output is required. This will appear in the logfile (stdout file).
Enable output of UTF file: Select if graphical output required in UTF (UGAMP Transfer Format).
Enable output of DRS files: select if graphical output required in DRS format.
Number of sig. figs. in printed output: Specify the number of significant digits to be used when printing output arrays. Usually a value of 3 will suffice.
Number of chars. in UTF packing: Specify the number of ASCII characters to be used to encode each data value in the UTF file. Usually a value of 2 will suffice. Specifying a negative value will result in the logarithm of the number being taken prior to packing.
Sampling frequency: This is the frequency in timesteps at which the input data will be sampled. For non-model data, assume that there are 24 timesteps per day and specify the sampling frequency in hours.
Frequency of printed output: This is the frequency in timesteps at which the printed output will be produced. For non-model data, assume that there are 24 timesteps per day and specify the output frequency in hours.
Frequency of plotable output: This is the frequency in timesteps at which the graphical output will be produced. For non-model data, assume that there are 24 timesteps per day and specify the output frequency in hours.
SUMO job name: This character string will be appended to various output files and datasets and can be used to indicate the source of the data, or otherwise identify the data in the output file.
SUMO job number: Specify a real number that will be used in UTF files to indicate the job number creating the UTF file.
Enable mass-weighted means: This is a switch to enable gridpoint fields to be mass-weighted (i.e. multiplied by isentropic sigma) when interpolated to isentropic surfaces.
Enable mask-weighted means: This is a switch to enable land/air mask weighting when computing spatial and temporal averages.
Enable isentropic-weighted means: This is a switch to enable isentropic sigma mask weighting when computing spatial and temporal averages. Currently only applies to output on isentropic surfaces.

Data

Data : MODELCONSTANTS datablock panel

Number of timesteps per day: The number of model timesteps per day. Set to 24 for real data.
Start century of run or of diagnostics: Two digit integer value representing the starting century of the analysis data/model run, not that of the restart job. If the diagnostic start and end times are specified using dates, then the user should enter the value used for the analysis starting century in the COMMON datablock.
Start year of run or of diagnostics: Two digit integer value representing the starting year of the analysis data/model run, not that of the restart job. If the diagnostic start and end times are specified using dates, then the user should enter the value used for the analysis starting year in the COMMON datablock.
Start month of run or of diagnostics: Two digit integer value representing the starting month of the analysis data/model run, not that of the restart job. If the diagnostic start and end times are specified using dates, then the user should enter the value used for the analysis starting month in the COMMON datablock.
Start day of run or of diagnostics: Two digit integer value representing the starting day of the analysis data/model run, not that of the restart job. If the diagnostic start and end times are specified using dates, then the user should enter the value used for the analysis starting day in the COMMON datablock.
Start hour of run or of diagnostics: Two digit integer value representing the starting hour of the analysis data/model run, not that of the restart job. If the diagnostic start and end times are specified using dates, then the user should enter the value used for the analysis starting hour in the COMMON datablock.
Start minute of run or of diagnostics: Two digit integer value representing the starting minute of the analysis data/model run, not that of the restart job. If the diagnostic start and end times are specified using dates, then the user should enter the value used for the analysis starting minute in the COMMON datablock.

Data

Data : UMFILENAME datablock panel

UM experiment id
UM job id
UM submodel id
UM timestandard indicator
UM filetype indicator
UM filestream indicator: The above are all required when the input data is in the form of a UM-generated post-processed file, on either model or pressure surfaces.; Filenames must be of the form ccfmaa.pjj4110 where; ccfm is the experiment id,; a is job id,; a signifies submodel producing the data,; . is the time standard indicator,; p is the file type indicator,; j indicates the filestream,; and j4110 defines the date.

Data

Data : LV datablock panel

Output surface values: The user should select as many of the output levels are are required; data will be interpolated to these surfaces. Note that if the number of output levels and the number of input levels are different, then the user must supply one value per output level. If the number of output levels and the number of input levels are identical, then any undefined surfaces will be determined by using a set of default values.; Values should be in hPa, Kelvin, or pv units depending upon the output surface type.

Data

Data : OR datablock panel

Printed orography to be output: Select if printout of orography required in stdout file.
Orography to be output for plotting: Select if orography required in plotable format.
Orography horizontal UTF mask to be output: Selecting this (along with the UTF output option) will output the horizontal orography UTF mask.
Orography north-south UTF mask to be output: Selecting this (along with the UTF output option) will output the north-south orography UTF mask.
Orography west-east UTF mask to be output: Selecting this (along with the UTF output option) will output the west-east orography UTF mask.
Orography UTF contour interval: This will be the orography contour interval specified in the UTF. The actual contour interval used by the UTF plotting package can subsequently be altered, so this value is not too important.

Data

Data : SG datablock panel

Diagnostic types required: Select one or more of GH (horizontal SG), GZ (zonal SG), GM (meridional SG), GP (profile SG) and G3 (3D output of SG) diagnostics.
Output time option required: Select one or more of instantaneous, time-mean, time series and running mean.
Fortran channel for printed output: If printed output has been selected, then any prinout of SG diagnostics requested will be output on this channel. NOTE: If channel 6 is not used for this, then the user should modify the jobscript so that this file is copied to a permanent location before the script finishes executing.
Output surface values: The user should delete the values not needed or modify the LV datablock if additional values are required. Note that the values in this section must (with the exception of null values and the value 0.0) agree with those specified in the LV datablock.

Data

Data : SG datablock panel

Data : SG : GH datablock panel

Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected one of UTF, DRS and netCDF output in the COMMON datablock.
UTF wind vectors: If you have selected UTF output then you can obtain wind vector information in UTF format to overlay on your final plots.
UTF contour interval: If you have selected UTF output then you can select a default contour interval for each field. Note that in practice the actual contour interval can be interactively altered at plotting time.
Enable output of latitude-longitude slices: Select this to output data on `horizontal' slices. This is the usual option selected.
Enable output of column totals: This option is available with UTF output on model levels.
Enable output of field minus zonal mean: This option is occasionally used with UTF output.
Southern and northern limits to Hovmoeller plots: This option is only available with UTF output.

Data

Data : SG datablock panel

Data : SG : GZ datablock panel

Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected one of UTF, DRS and netCDF output in the COMMON datablock.
UTF contour interval: If you have selected UTF output then you can select a default contour interval for each field. Note that in practice the actual contour interval can be interactively altered at plotting time.
Enable zonal mean output: This is the usual spatial option to select in this datablock.
Enable time-series output: This option can only be used with UTF output; other data formats should allow this sub-selecting to be done using the whole data file to be output.
Longitude along which north-south slices of variables will be output: This option for selecting longitudinal slices is only available with UTF output.
West and east longitudes between which averages of variables will be output: This is equivalent to Hovmoeller plots over a longitude (rather than latitude) band. This option for selecting longitudinal slices is only available with UTF output.

Data

Data : SG datablock panel

Data : SG : GM datablock panel

The options in this datablock are primarily intended for UTF output.
Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected UTF output in the COMMON datablock.
UTF contour interval: If you have selected UTF output then you can select a default contour interval for each field. Note that in practice the actual contour interval can be interactively altered at plotting time.
Enable meridional mean output: This is the usual spatial option to select in this datablock.
Enable time-series output: This option can only be used with UTF output; other data formats should allow this sub-selecting to be done using the whole data file to be output.
Latitude along which west-east slices of variables will be output: This option for selecting longitudinal slices is only available with UTF output.
South and north latitudes between which averages of variables will be output: This creates a Hovmoeller plot. This option for selecting longitudinal slices is only available with UTF output.

Data

Data : SG datablock panel

Data : SG : GP datablock panel

The options in this datablock are primarily intended for UTF output.
Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected UTF output in the COMMON datablock.
Coordinate latitude and longitude: Specify the latitude and longitude (the latter in the range 0 <= longitude < 360 degrees) of up to 10 locations where vertical profiles will be generated.
Enable column profile output: Select if required. Graphical output in a UTF file and printed output are available.
Enable column total output: Select if required. Graphical output in a UTF file and printed output are available.

Data

Data : SG datablock panel

Data : SG : G3 datablock panel

This datablock is designed to enable the user to output 3-dimensional SG fields, or spatially-defined subsets thereof, in a format that can be controlled by the user. By default, the output will be written in a format specified by the SUMO subroutine OPSG3D. To access this facility, you must select the UTF output option in the COMMON datablock - this facility is regarded as an extension of the graphical output option.

It is not necessary to use the format in OPSG3D - you can replace this routine with your own code as follows:

1. Create an nupdate modification file containing the following

*IDENT MYSGCODE
*D SGDIAGS.1348,SGDIAGS.1467
      SUBROUTINE OPSG3D(LPNOW)
Add you own code in here, writing any data to channel a valid
channel number - see below. Code should be written using fortran 90.
Use the existing OPSG3D subroutine as a template to ensure you
access all the necessary variables and data.
      RETURN
      END

2. Save this file in a location on the remote host.

3. Run XSUMO taking note of the following points:

(a) Select the UTF output option.

(b) Select SG and G3 diagnostics.

SUMO code panel, by adding an entry along the lines of

*READ /my_directory_pathname/name_of_file_with_code_changes

(d) Select required fields in the G3 datablock panel.

(e) Ensure that you compile the fortran code, at least for the first successful run, using the Compile source code option.

(f) Do not use the XSUMO "Submit" button.

(g) Quit from XSUMO after "Process" has completed successfully. Then modify the script to ensure that your chosen file, corresponding to the channel number, is saved to a permanent location.

Output

Select the required fields.

Fortran channel for output

Select a valid channel number. At version 1.5 of SUMO the following channels are available: 1, 17, 18, 19, 20, 21, 22, 23, 24. See the complete user guide for available channels with other versions. If you do not select channel 6, you will need to quit from XSUMO after "Process" has completed successfully and then modify the script to ensure that your chosen file, corresponding to the channel number, is saved to a permanent location.

Data

Data : TF datablock panel

Diagnostic types required

Select one or more of FH (horizontal TF), FZ (zonal TF), FM (meridional TF), FP (profile TF) and F3 (3D output of TF) diagnostics.

Fortran channel for printed output

If printed output has been selected, then any prinout of TF diagnostics requested will be output on this channel. NOTE: If channel 6 is not used for this, then the user should modify the jobscript so that this file is copied to a permanent location before the script finishes executing.

Output surface values

The user should delete the values not needed or modify the LV datablock if additional values are required. Note that the values in this section must (with the exception of null values and the value 0.0) agree with those specified in the LV datablock.

Flux types required

Select those transient flux components required.

High-low pass cutoff

Typically this is chosen to be 3.0 days. However, the only restriction on the value chosen is that the total duration of the analysis (DUR, days), the sampling frequency (SF, days) and the cutoff (CUT, days) must obey the following relationship

DUR/SF + 1 = CUT/SF times N

where N is an integer. So if the analysis is from days 0.0 to 29.75 ( a duration of 29.75 days), the sampling frequency is 0.25 days (i.e there are 120 samples) and the cutoff is 3.0 days, then n is 10.

Specify transient flux formulae

Modify as many entries here as necessary to specify exactly the transient fluxes required. The mask determines how any summations or averaging are computed (see the Maskhelp button), the factor is a multiplicative factor, and the field is a code number denoting the field (see the Fieldhelp button). Thus the entry

0 1.0 5 5 0.0 0 0

will compute the covariance of the zonal wind, while

B 0.5 5 5 0.5 6 6

will compute the eddy kinetic energy with only real (i.e. above ground - since the land/air mask is zero underground) data being used in the summations and averages.

Note that the "number" in the "Variable" column should be used in the FH, FZ, FM, FP and F3 datablock panels to reference the individual transient fluxes.

Data

Data : TF datablock panel

Data : TF : FH datablock panel

Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected one of UTF, DRS and netCDF output in the COMMON datablock.
UTF contour interval: If you have selected UTF output then you can select a default contour interval for each field. Note that in practice the actual contour interval can be interactively altered at plotting time.
Enable output of latitude-longitude slices: Select this to output data on `horizontal' slices.

Data

Data : TF datablock panel

Data : TF : FZ datablock panel

Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected one of UTF, DRS and netCDF output in the COMMON datablock.
UTF contour interval: If you have selected UTF output then you can select a default contour interval for each field. Note that in practice the actual contour interval can be interactively altered at plotting time.
Enable zonal mean output: This is the usual spatial option to select in this datablock.
Longitude along which north-south slices of variables will be output: This option for selecting longitudinal slices is only available with UTF output.
West and east longitudes between which averages of variables will be output: This is equivalent to Hovmoeller plots over a longitude (rather than latitude) band. This option for selecting longitudinal slices is only available with UTF output.

Data

Data : TF datablock panel

Data : TF : FM datablock panel

The options in this datablock are primarily intended for UTF output.
Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected UTF output in the COMMON datablock.
UTF contour interval: If you have selected UTF output then you can select a default contour interval for each field. Note that in practice the actual contour interval can be interactively altered at plotting time.
Enable meridional mean output: This is the usual spatial option to select in this datablock.
Enable time-series output: This option can only be used with UTF output; other data formats should allow this sub-selecting to be done using the whole data file to be output.
Latitude along which west-east slices of variables will be output: This option for selecting longitudinal slices is only available with UTF output.
South and north latitudes between which averages of variables will be output: This creates a Hovmoeller plot. This option for selecting longitudinal slices is only available with UTF output.

Data

Data : TF datablock panel

Data : TF : FP datablock panel

The options in this datablock are primarily intended for UTF output.
Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected UTF output in the COMMON datablock.
Coordinate latitude and longitude: Specify the latitude and longitude (the latter in the range 0 <= longitude < 360 degrees) of up to 10 locations where vertical profiles will be generated.
Enable column profile output: Select if required. Graphical output in a UTF file and printed output are available.

Data

Data : TF datablock panel

Data : TF : F3 datablock panel

This datablock is designed to enable the user to output 3-dimensional TF fields, or spatially-defined subsets thereof, in a format that can be controlled by the user. By default, the output will be written in a format specified by the SUMO subroutine F3OUT. To access this facility, you must select the UTF output option in the COMMON datablock - this facility is regarded as an extension of the graphical output option.

It is not necessary to use the format in F3OUT - you can replace this routine with your own code as follows:

1. Create an nupdate modification file containing the following

*IDENT MYTFCODE
*D TRNFLX.116,TRNFLX.206
      SUBROUTINE F3OUT(KF,KPASS,PX,KLEV)
Add you own code in here, writing any data to channel a valid
channel number - see below. Code should be written using fortran 90.
Use the existing TRNFLX subroutine as a template to ensure you
access all the necessary variables and data.
      RETURN
      END

2. Save this file in a location on the remote host.

3. Run XSUMO taking note of the following points:

(a) Select the UTF output option.

(b) Select TF and F3 diagnostics.

SUMO code panel, by adding an entry along the lines of

*READ /my_directory_pathname/name_of_file_with_code_changes

(d) Select required fields in the F3 datablock panel.

(e) Ensure that you compile the fortran code, at least for the first successful run, using the Compile source code option.

(f) Do not use the XSUMO "Submit" button.

(g) Quit from XSUMO after "Process" has completed successfully. Then modify the script to ensure that your chosen file, corresponding to the channel number, is saved to a permanent location.

Output

Select the required fields.

Fortran channel for output

Data

Data : OD datablock panel

Diagnostic types required: Select one or more of OH (horizontal OD), OZ (zonal OD), OM (meridional OD), OP (profile OD) and O3 (3D output of OD) diagnostics.
Output time option required: Select one or more of instantaneous, time-mean and time series.
Fortran channel for printed output: If printed output has been selected, then any prinout of SG diagnostics requested will be output on this channel. NOTE: If channel 6 is not used for this, then the user should modify the jobscript so that this file is copied to a permanent location before the script finishes executing.
Output surface values: The user should delete the values not needed or modify the LV datablock if additional values are required. Note that the values in this section must (with the exception of null values and the value 0.0) agree with those specified in the LV datablock.

Data

Data : OD datablock panel

Data : OD : OH datablock panel

Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected one of UTF, DRS and netCDF output in the COMMON datablock.
UTF wind vectors: If you have selected UTF output then you can obtain wind vector information in UTF format to overlay on your final plots.
UTF contour interval: If you have selected UTF output then you can select a default contour interval for each field. Note that in practice the actual contour interval can be interactively altered at plotting time.
Enable output of latitude-longitude slices: Select this to output data on `horizontal' slices. This is the usual option selected.
Enable output of column totals: This option is available with UTF output on model levels.
Enable output of field minus zonal mean: This option is occasionally used with UTF output.
Southern and northern limits to Hovmoeller plots: This option is only available with UTF output.

Data

Data : OD datablock panel

Data : OD : OZ datablock panel

Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected one of UTF, DRS and netCDF output in the COMMON datablock.
UTF contour interval: If you have selected UTF output then you can select a default contour interval for each field. Note that in practice the actual contour interval can be interactively altered at plotting time.
Enable zonal mean output: This is the usual spatial option to select in this datablock.
Enable time-series output: This option can only be used with UTF output; other data formats should allow this sub-selecting to be done using the whole data file to be output.
Longitude along which north-south slices of variables will be output: This option for selecting longitudinal slices is only available with UTF output.
West and east longitudes between which averages of variables will be output: This is equivalent to Hovmoeller plots over a longitude (rather than latitude) band. This option for selecting longitudinal slices is only available with UTF output.

Data

Data : OD datablock panel

Data : OD : OM datablock panel

The options in this datablock are primarily intended for UTF output.
Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected UTF output in the COMMON datablock.
UTF contour interval: If you have selected UTF output then you can select a default contour interval for each field. Note that in practice the actual contour interval can be interactively altered at plotting time.
Enable meridional mean output: This is the usual spatial option to select in this datablock.
Enable time-series output: This option can only be used with UTF output; other data formats should allow this sub-selecting to be done using the whole data file to be output.
Latitude along which west-east slices of variables will be output: This option for selecting longitudinal slices is only available with UTF output.
South and north latitudes between which averages of variables will be output: This creates a Hovmoeller plot. This option for selecting longitudinal slices is only available with UTF output.

Data

Data : OD datablock panel

Data : OD : OP datablock panel

The options in this datablock are primarily intended for UTF output.
Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected UTF output in the COMMON datablock.
Coordinate latitude and longitude: Specify the latitude and longitude (the latter in the range 0 <= longitude < 360 degrees) of up to 10 locations where vertical profiles will be generated.
Enable column profile output: Select if required. Graphical output in a UTF file and printed output are available.
Enable column total output: Select if required. Graphical output in a UTF file and printed output are available.

Data

Data : OD datablock panel

Data : OD : O3 datablock panel

This datablock is designed to enable the user to output 3-dimensional OD fields, or spatially-defined subsets thereof, in a format that can be controlled by the user. By default, the output will be written in a format specified by the SUMO subroutine OPOD3D. To access this facility, you must select the UTF output option in the COMMON datablock - this facility is regarded as an extension of the graphical output option.

It is not necessary to use the format in OPOD3D - you can replace this routine with your own code as follows:

1. Create an nupdate modification file containing the following

*IDENT MYODCODE
*D OPDYNMC.2399,OPDYNMC.2518
      SUBROUTINE OPOD3D(LPNOW)
Add you own code in here, writing any data to channel a valid
channel number - see below. Code should be written using fortran 90.
Use the existing OPOD3D subroutine as a template to ensure you
access all the necessary variables and data.
      RETURN
      END

2. Save this file in a location on the remote host.

3. Run XSUMO taking note of the following points:

(a) Select the UTF output option.

(b) Select OD and O3 diagnostics.

SUMO code panel, by adding an entry along the lines of

*READ /my_directory_pathname/name_of_file_with_code_changes

(d) Select required fields in the O3 datablock panel.

(e) Ensure that you compile the fortran code, at least for the first successful run, using the Compile source code option.

(f) Do not use the XSUMO "Submit" button.

(g) Quit from XSUMO after "Process" has completed successfully. Then modify the script to ensure that your chosen file, corresponding to the channel number, is saved to a permanent location.

Output

Select the required fields.

Fortran channel for output

Data

Data : ZF datablock panel

Diagnostic types required

Select zonal-mean and/or level-mean fields. Note that the latter is only available for UTF output; the level-mean data is included in the zonal-mean output in all cases.

Fortran channel for output

Output surface values

Flux types required

Select eddy flux and/or mean flux.

Output time option required

Select one or more options. Note that time-series are only available with UTF output - with other output types they are identical to instantaneous output but have more than two dimensions.

Specify zonal mean flux formulae

Modify as many entries here as necessary to specify exactly the transient fluxes required. The factor is a multiplicative factor, and the field is a code number denoting the field (see the Fieldhelp button). Thus the entry

1.0 5 5 0.0 0 0

will compute the product of the zonal wind with the zonal wind (zonal flux of zonal wind), while

1.0 5 18 1.0 6 18

will compute the sum of the zonal and meridional fluxes of temperature.

Print o/p

Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.

Graph o/p

Select those fields for which you require output in a graphical format. You need to have selected UTF output in the COMMON datablock.

UTF contour interval

If you have selected UTF output then you can select a default contour interval for each field. Note that in practice the actual contour interval can be interactively altered at plotting time.

Data

Data : XD datablock panel

Output time option required: Select one or both options.
Fortran channel for output: Select a valid channel number. At version 1.5 of SUMO the following channels are available: 1, 17, 18, 19, 20, 21, 22, 23, 24. See the complete user guide for available channels with other versions. If you do not select channel 6, you will need to quit from XSUMO after "Process" has completed successfully and then modify the script to ensure that your chosen file, corresponding to the channel number, is saved to a permanent location.
Fieldhelp: Use this button to determine which fields correspond to which field numbers.
Print o/p: Select those fields for which you required printout of analysis values. You need to ensure that printed output has been selected.
Graph o/p: Select those fields for which you require output in a graphical format. You need to have selected UTF output in the COMMON datablock.
UTF contour interval: If you have selected UTF output then you can select a default contour interval for each field. Note that in practice the actual contour interval can be interactively altered at plotting time.
Specify spatial option required.: Select one or both options. Note that level-means are only available with UTF output - they can be computed from the zonal-mean means available with DRS or netCDF output.

Code changes and definitions

Code : PRESUMO code panel

Note that PRESUMO will be compiled and run on the local host. All directories and files specified below must exist on that machine.

Use default PRESUMO

Selecting this option will mean that your job should use the standard PRESUMO source code and pre-compiled executable. In the vast majority of cases this is the option to select. However, if you really need to alter PRESUMO, perhaps in response to your own changes to code in SUMO, then do not use the default PRESUMO but create your own source file using the nupdate edits specified below.

nupdate code changes

When compiling PRESUMO from scratch you are advised to ensure that the following two lines of source code are included in your nupdate edit file:

*IDENT PRESUMO15
*READ XXXXX/presumo15bug.nupd

Here, the first line ensures that all new code is given an edit identifier (a requirement of nupdate) while the second line will ensure that any current bugfixes (or `official' extensions to the current version of PRESUMO) will all be included in your modified source. Make sure that

XXXXX

is altered accordingly. Finally, you will need to insert additional nupdate edit of the form

*READ /my_directory_pathname/name_of_file_with_code_changes

where the specified pathname and file MUST reside on the local host.

Code changes and definitions

Code : SUMO code panel

Note that SUMO will be compiled and run on the remote host. All directories and files specified below must exist on that machine.
The first two nupdate edits should be of the form *READ XXXXX/presumo15bug.nupd *IDENT SUMO15 Here, the first line will ensure that any current bugfixes (or `official' extensions to the current version of SUMO) will all be included in your modified source, and the second line ensures that all new code is given an edit identifier (a requirement of nupdate) while the second line . Make sure that XXXXX is altered accordingly.
To avoid compiling and running the whole of SUMO, which would be unnecessarily expensive, you should select those options which describe your job, input data format and output data format.
The reset button will reset all values - you should then check all entries and alter those that are appropriate.
Finally, you may incorporateyour own code changes via additional nupdate edits of the form
*READ /my_directory_pathname/name_of_file_with_code_changes
where the specified pathname and file MUST reside on the remote host.

Last modified / 11.06.99 / This page is maintained by Roger Brugge ( brugge@met.rdg.ac.uk ).