[gradsusr] Regarding Temporal Correlation

Girija Kalyani kalyani at ncaor.gov.in
Wed Mar 30 08:29:31 EDT 2016


Thanks Mr.Kishore.
This was mentioned by Mr.Stephen earlier. It is not working. I get the same error
let me put down everything.
this is my file's description

ga-> q file 1
File 1 :
  Descriptor: C:\SeaIceStudies-Girija\IMP\Datsets\All_ECMWF\ALL.nc
  Binary: C:\SeaIceStudies-Girija\IMP\Datsets\All_ECMWF\ALL.nc
  Type = Gridded
  Xsize = 360  Ysize = 181  Zsize = 1  Tsize = 444  Esize = 1
  Number of Variables = 10
     ci  0  t,y,x  Sea-ice cover
     sst  0  t,y,x  Sea surface temperature
     p58.162  0  t,y,x  Vertical integral of ozone
     msl  0  t,y,x  Mean sea level pressure
     u10  0  t,y,x  10 metre U wind component
     v10  0  t,y,x  10 metre V wind component
     t2m  0  t,y,x  2 metre temperature
     al  0  t,y,x  Albedo
     si10  0  t,y,x  10 metre wind speed
     ishf  0  t,y,x  Instantaneous surface sensible heat flux

After you mentioned I kept trying, Now I get a new error;

ga-> d tcorr(tsvar,ci,t=1,t=444)
TCORR:  dim = 3, start = 1, end = 444
Data Request Warning:  Request is completely outside file limits
  Entire grid contents are set to missing data
  Grid limits of file:     X = 1 360  Y = 1 181  Z = 1 1  T = 1 444  E = 1 1
  Grid limits of request:  X = 0 360  Y = -90 -60  Z = 1 1  T = 1 1  E = 1 1
  Warning issued for variable = ci
Cannot contour grid - all undefined values

Is something wrong with my dimensions?


Girija Kalyani
Junior Research Fellow
Division of Polar Remote Sensing
NCAOR

________________________________
From: gradsusr-bounces at gradsusr.org on behalf of Kishore Ragi
Sent: Wednesday, March 30, 2016 4:18 PM
To: GrADS Users Forum
Subject: Re: [gradsusr] Regarding Temporal Correlation

the method you mentioned is working quite well with the data I have.

You just try the following

define tsvar=z          instead of  ts_var=z   if you do not run the .gs file.

Hope this helps you.

Regards,
Kishore

PhD student in climate science,
National Institute of Technology, Rourkela,
India- 769008
Phone: +91-8908671881

http://www.researchgate.net/profile/Kishore_Ragi

https://in.linkedin.com/in/kishoreragi

http://kishoreragi.webnode.com/

On Wed, Mar 30, 2016 at 8:18 PM, Girija Kalyani <kalyani at ncaor.gov.in<mailto:kalyani at ncaor.gov.in>> wrote:
Thanks for the response Mr.Kishore. I'm initially trying out to do what they did with the sample file.
They have 'z' variable representing geopotential height.
But the script isn't working with their data as well.
Once it works well with sample I wanted to shift to my data.
I have sea-ice, wind, MSL, SLP, Albedo data for past 36 years.
All of same spatial and temporal extent.
I wanted to perform spatio-temporal correlational analysis with atleast 2 variables (of those that are mentioned above).
I hope I was clear enough.
Many Thanks




________________________________
From: gradsusr-bounces at gradsusr.org<mailto:gradsusr-bounces at gradsusr.org> on behalf of Kishore Ragi
Sent: Wednesday, March 30, 2016 12:57 PM

To: GrADS Users Forum
Subject: Re: [gradsusr] Regarding Temporal Correlation

Hi,

z is their variable (geopotential height). They have defined ts_var for a specific location for the period of time. Then they tried to correlate that point with
'set lat 0 90'
    'set lon 180 360'   region. You see what you are trying to do. If it is same you are trying to, its fine (but choose the variable that you have in the file). If not, lets us know what you want so that one of the group members may help you out.

Regards,
Kishore



PhD student in climate science,
National Institute of Technology, Rourkela,
India- 769008
Phone: +91-8908671881<tel:%2B91-8908671881>

http://www.researchgate.net/profile/Kishore_Ragi

https://in.linkedin.com/in/kishoreragi

http://kishoreragi.webnode.com/

On Wed, Mar 30, 2016 at 5:52 PM, Girija Kalyani <kalyani at ncaor.gov.in<mailto:kalyani at ncaor.gov.in>> wrote:
Oh Yes.
I check the variables the similar way. And Im aware that this variable is not in the file.
But it is given in this link (http://gradsaddict.blogspot.in/2013/05/tutorial-use-tcorr-function-to-create.html)
that time-series should be defined this way.


Girija Kalyani
Junior Research Fellow
Division of Polar Remote Sensing
NCAOR

________________________________
From: gradsusr-bounces at gradsusr.org<mailto:gradsusr-bounces at gradsusr.org> on behalf of Kishore Ragi
Sent: Wednesday, March 30, 2016 10:31 AM

To: GrADS Users Forum
Subject: Re: [gradsusr] Regarding Temporal Correlation

check the variables in the file using

q file

If you find z as one of the variable, you should not have any problem with your code.

Also, I am curious to know why you want to define ts_var=z

Regards,
Kishore

PhD student in climate science,
National Institute of Technology, Rourkela,
India- 769008
Phone: +91-8908671881<tel:%2B91-8908671881>

http://www.researchgate.net/profile/Kishore_Ragi

https://in.linkedin.com/in/kishoreragi

http://kishoreragi.webnode.com/

On Wed, Mar 30, 2016 at 2:46 PM, Girija Kalyani <kalyani at ncaor.gov.in<mailto:kalyani at ncaor.gov.in>> wrote:
Thanks for the response Mr.Stephen.
I tried removing the underscore as directed, but still the error did not change.
It produces
Syntax Error: Invalid Operand
'z' is not a function name
Define Error: Invalid Expression


Girija Kalyani
Junior Research Fellow
Division of Polar Remote Sensing
NCAOR

________________________________
From: gradsusr-bounces at gradsusr.org<mailto:gradsusr-bounces at gradsusr.org> on behalf of Stephen McMillan
Sent: Tuesday, March 29, 2016 5:54 PM
To: GrADS Users Forum
Subject: Re: [gradsusr] Regarding Temporal Correlation

Girija,

Remove the "_" from your variable name.  GrADS doesn't like that or other special characters (numerals are okay if not the first character).

In the future, when you have an error, please say what the error was.  This was an easy case, but sometimes one could only guess.

Good luck--
Stephen Mc

On Tue, Mar 29, 2016 at 7:41 AM, Girija Kalyani <kalyani at ncaor.gov.in<mailto:kalyani at ncaor.gov.in>> wrote:
Good Evening,

I'm using Grads 2.0.2 on Windows desktop.
I wanted to perform temporal correlation as given in the Gradsaholic website
http://gradsaddict.blogspot.in/2013/05/tutorial-use-tcorr-function-to-create.html,http://gradsaddict.blogspot.in/2013/05/tutorial-use-tcorr-function-to-create.html,
but I get an error when the time series variable is defined.
Given

set t 100 220
set lat -90 -50
set lon 0 360
ts_var=z

I have defined the settings as mentioned above, where the last line shows an error.
Im unable save the timeseries in the variable.

Any help is highly appreciated!

_______________________________________________
gradsusr mailing list
gradsusr at gradsusr.org<mailto:gradsusr at gradsusr.org>
http://gradsusr.org/mailman/listinfo/gradsusr



_______________________________________________
gradsusr mailing list
gradsusr at gradsusr.org<mailto:gradsusr at gradsusr.org>
http://gradsusr.org/mailman/listinfo/gradsusr



_______________________________________________
gradsusr mailing list
gradsusr at gradsusr.org<mailto:gradsusr at gradsusr.org>
http://gradsusr.org/mailman/listinfo/gradsusr



_______________________________________________
gradsusr mailing list
gradsusr at gradsusr.org<mailto:gradsusr at gradsusr.org>
http://gradsusr.org/mailman/listinfo/gradsusr


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://gradsusr.org/pipermail/gradsusr/attachments/20160330/6a245c24/attachment-0001.html 


More information about the gradsusr mailing list