Re: Dimension problem

From: Dennis Shea <shea_at_nyahnyahspammersnyahnyah>
Date: Fri, 13 Oct 2006 10:36:56 -0600 (MDT)

I responded to the sender but neglected to cc ncl-talk.

From: Dennis Shea <shea_at_mailhub.cgd.ucar.edu>
Subject: Re: Dimension problem
To: araligin_at_onid.orst.edu

>
> I get the following error when I am computing the correlation coefficient.
>
>warning:esccr: Non-fatal conditions encountered: all missing or constant values

The above *warning* is likely due to one or more time
series having all the same value. EG: a value of 0.0
over desert areas. The correlation coef is derived
by dividing be the standard deviation, If all values
are the same, then the std. dev. is also 0.0.

>fatal:Div: Number of dimensions do not match, can't continue
>fatal:Execute: Error occurred at or near line 59 in file wspdprecip.ncl

I am not sure what you are doing here:

ccf_ssum = esccr(wspd_ssum(lat|:,lon|:,time|:) \
                ,rain_ssum(lat|:,lon|:,time|:),0) \
                /(stddevws_ssum*stddevrn_ssum)
                
The following returns a 3d array

  ccr_3d = esccr(wspd_ssum(lat|:,lon|:,time|:) \ ; (lat,lon,mxlag)
                ,rain_ssum(lat|:,lon|:,time|:),0)

The denominator is 2d ... hence the error msg.

----
I suggest using  escorc
   http://www.ncl.ucar.edu/Document/Functions/Built-in/escorc.shtml           
ccf_ssum = escorc(wspd_ssum(lat|:,lon|:,time|:)  \
                 ,rain_ssum(lat|:,lon|:,time|:)) \
                /(stddevws_ssum*stddevrn_ssum)
        
>
>   I checked the dimension of both the variable they are equal. I cannot figure
>out whats wrong. I have attached the code along with this mail.
>
>Date: Fri, 13 Oct 2006 10:17:38 -0600 (MDT)
>From: Mary Haley <haley_at_ucar.edu>
>To: araligin_at_onid.orst.edu
>Cc: Dennis Shea <shea_at_ucar.edu>, Adam Phillips <asphilli_at_ucar.edu>
>Subject: Re: Dimension problem
>MIME-Version: 1.0
>X-Virus-Scanned: amavisd-new at ucar.edu
>
>
>Hi Nilesh,
>
>Can you provide the data files that goes with your script:
>
>     ftp ftp.cgd.ucar.edu
>     <log in as "anonymous">
>     <Use email address as password>
>     cd incoming
>     mput t2.nc qscat.150130.int.nc
>     quit
>
>If they are too big to put there, then let me know and I'll suggest
>a way around this.
>
>Thanks,
>
>--Mary
>
>
>On Thu, 12 Oct 2006 araligin_at_onid.orst.edu wrote:
>
>>
>> Hi,
>>
>>  I get the following error when I am computing the correlation coefficient.
>>
>> warning:esccr: Non-fatal conditions encountered: all missing or constant 
values
>> fatal:Div: Number of dimensions do not match, can't continue
>> fatal:Execute: Error occurred at or near line 59 in file wspdprecip.ncl
>>
>>   I checked the dimension of both the variable they are equal. I cannot 
figure
>> out whats wrong. I have attached the code along with this mail.
>>
>> Thanking you,
>> Nilesh
>>
>>
_______________________________________________
ncl-talk mailing list
ncl-talk_at_ucar.edu
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Fri Oct 13 2006 - 10:36:56 MDT

This archive was generated by hypermail 2.2.0 : Fri Oct 13 2006 - 14:01:26 MDT