[RegCNET] BATS soil moisture subroutines and Postproc?

Fri Aug 16 16:22:32 CEST 2013

Hi Marshall,

There is one step to do before using grads to visialize your data. For instance your output is SRFxxxx.nc then you have to do: ./bin/GradsNCPrepare SRFxxxx.nc, then you gonna have a new file like SRFxxxx.nc.ctl
Now to plot with grads you can do:
open SRFxxxx.nc.ctl

Important: To plot your data with grads you always need to create a control file either by using GrADSNcPrepare or GrADSNcPlot

To get daily and/or monthly average you need to write a shell script. The best way to do that is to use NCO or CDO command.
Example: cdo timmean  SRFxxxx.nc SRFxxxx.avg.nc
./bin/GrADSNcPrepare  SRFxxxx.avg.nc
open SRFxxxx.avg.nc.ctl

Best,

Ismaila 

________________________________
 From: Marshall L. Mdoka <marshall.mdoka at gmail.com>
To: ggiulian at ictp.it 
Cc: regcnet at lists.ictp.it 
Sent: Friday, August 16, 2013 4:00 PM
Subject: Re: [RegCNET] BATS soil moisture subroutines and Postproc?

Hi All,

I have read most of the reference on Post processing and how it can help with analysis. I have installed most of them but still have problems addressing my previous problems such as viewing precip, air temp, winds in SRFxxxxx.nc and vertical variables in ATMxxxxx.nc. Could someone please kindly guide me with the process say from completing the simulation and I have my SRFxxxxx.nc in output directory. I would be very grateful for a step-by-step process to prepare final files that I can analyse, differentiate, calculate biases etc using Grads thereafter. Maybe if I could get an idea of how I can use the scripts in RegCM/bin for postprocessing to get dailies /monthly files? 

Trying to use ncdump shows me the variables I am seeking but how can I 
then view them since there are not showing in the list after querying 
the file using Grads?

From what I have been trying using the output SRFxxxxx.nc. Here are some errors I got:

1. 
ga-> sdfopen SA_SRF.1997010100.nc 
Scanning self-describing file:  SA_SRF.1997010100.nc
SDF file SA_SRF.1997010100.nc is open as file 1
LON set to -2.6e+06 2.6e+06 
LAT set to -2.45e+06 2.45e+06 
LEV set to -2.6e+06 -2.6e+06 
Time values set: 1997:1:1:3 1997:1:1:3 
E set to 1 1 

ga-> d ts
Warning: Could not retrieve "_FillValue" -- using 1e+20 instead
Out of buffer space
Contouring: 284 to 302 interval 2 

2. 

Using cdo monmean SRFxxxxxxxxxx.nc SRFavg.nc gives this:

Warning (cdfSetVar) : Inconsistent variable definition for xlat!
Warning (cdfSetVar) : Inconsistent variable definition for xlon!
cdo monmean: Processed 64469790 values from 25 variables over 248 timesteps. ( 2.46s )

I am at the moment not exactly sure of the sequential steps to follow at the Post-processing stage. There is a lot of given detailed guidelines in RegCM-4.3.5.6 up until the simulation step. I would be grateful for any assistance for this last processing step.

Kind regards,

Marshall

On Tue, Aug 6, 2013 at 5:34 PM, Marshall L. Mdoka <marshall.mdoka at gmail.com> wrote:

Dear Graziano,
>
>Thank you very much for immediate response. I will check on the subroutines. 
>
>Postproc.in, I used it to as Post-processing step to create the final merged output file that I will then analyse for some of those variables  (rt, t2m, ua, va etc) I have mentioned above. So it means I have to do the interpolation and time avergind steps you mention above before I can start my analyses. So I thought there was a mistake since I could not find postproc.in to include in the postprocessing of the simulations.
>
>Using ncdump -h to view a SRFxxxxxx.nc or ATMxxxxxx.nc shows more variables than can be viewed t if I open in Grads. That's why I raised it as I cannot view them. 
>
>Thank you.
>
>Marshall
>
>
>
>
>
>
>
>On Mon, Aug 5, 2013 at 4:55 PM, Graziano Giuliani <ggiulian at ictp.it> wrote:
>
>On 08/02/2013 03:10 PM, Marshall L. Mdoka wrote:
>>
>>Hi RegCNETers,
>>>
>>>I wanted to find out if I want to do some soil moisture perturbations in
>>>BATs where do I find these programmes to change in RegCM-4.3.5.6? In
>>>RegCM3, there are interf.F and initb.F.
>>>
[...]
>>
>>Those subroutines are now in the module mod_bats_mtrxbats in the batslib directory.
>>
>>[...]
>>
>>
>>As for postproc, I cant find postproc.in in any of the folders, could you
>>>please kindly assist me on how I should go about creating my final outputs.
>>>
Unclear what you mean by "final" output. What you were using the postproc for?
>>
>>-) Interpolate to another grid -> Use cdo
>>-) Perform time mean -> Use cdo or nco
>>-) Interpolate to vertical pressure levels -> Use sigma2p
>>
>>[...]
>>
>>
>>I also noticed that using the test.001.in I get a few variables in
>>>SRFxxxxxx.nc ATMxxxxxx.nc etc. How can I increase the variables to include
>>>the uwind, vwind, T2m, rt which are missing but vital to what I want to do?
>>>
[...]
>>
>>? Just ncdump -h the output file. The variables should be there.
>>
>>Graziano.
>>
>>-- 
>>Graziano Giuliani - Earth System Physics Section
>>The Abdus Salam International Centre for Theoretical Physics
>>Strada Costiera, 11 - I - 34151 Trieste Italy
>>_______________________________________________
>>RegCNET mailing list
>>RegCNET at lists.ictp.it
>>https://lists.ictp.it/mailman/listinfo.cgi/regcnet
>>
>
>
>
>-- 
>Marshall L. Mdoka
>Head of Section 
>Geography and Environmental Science,
>Monash South Africa,
>Private Bag X60,
>Roodepoort, 1725,
>Johannesburg, South Africa
>Tel: +27 (0)11 950 4105
>Fax: +27 (0)11 950 4166
>Cell: +27 (0)83 528 8553
>Email: marshall.mdoka at monash.edu
>Web: www.monash.ac.za 
>Wholly owned by Monash University Australia
>
>

-- 
Marshall L. Mdoka
Head of Section 
Geography and Environmental Science,
Monash South Africa,
Private Bag X60,
Roodepoort, 1725,
Johannesburg, South Africa
Tel: +27 (0)11 950 4105
Fax: +27 (0)11 950 4166
Cell: +27 (0)83 528 8553
Email: marshall.mdoka at monash.edu
Web: www.monash.ac.za 
Wholly owned by Monash University Australia

_______________________________________________
RegCNET mailing list
RegCNET at lists.ictp.it
https://lists.ictp.it/mailman/listinfo.cgi/regcnet
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ictp.it/pipermail/regcnet/attachments/20130816/160f2977/attachment-0001.htm>