[Met_help] [rt.rap.ucar.edu #68168] History for pattern correlation

[John Halley Gotway via RT]

----------------------------------------------------------------
  Initial Request
----------------------------------------------------------------

hi john, 
i want to know how can i generate spatial plot of pattern correlation between the WRF forecast (9km) and the gridded observations (50km). 

will that be a simple correlation where first variable is RF (obs) and RF FCST will be 2nd variabble. 

can u guide??
geeta 		 	   		  

----------------------------------------------------------------
  Complete Ticket History
----------------------------------------------------------------

Subject: pattern correlation
From: John Halley Gotway
Time: Mon Jul 14 14:51:38 2014

Geeta,

I'm not exactly sure what you're looking for.  If you put the fcst and
obs
on a common grid, you could run the series-analysis tool and compute
any of
the traditional stats for each grid point through time, such as
Pearson
correlation.  You hand series_analysis a list of gridded forecast
files and
a list of gridded observation files.  For example, a time series of
24-hour
forecasts for 1 month would work.  Rather than computing statistics as
a
spatial average over some masking regions, each grid point is treated
as
it's own "masking region".  You can compute 1 or more statistics for
each
grid point.  The output is gridded NetCDF file showing the model
performance grid point by grid point.

Is that what you're looking for, or something different?

Thanks,
John Halley Gotway
met_help at ucar.edu


On Sat, Jul 12, 2014 at 6:49 AM, Geeta Geeta via RT
<met_help at ucar.edu>
wrote:

>
> Sat Jul 12 06:49:42 2014: Request 68168 was acted upon.
> Transaction: Ticket created by geeta124 at hotmail.com
>        Queue: met_help
>      Subject: pattern correlation
>        Owner: Nobody
>   Requestors: geeta124 at hotmail.com
>       Status: new
>  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68168 >
>
>
> hi john,
> i want to know how can i generate spatial plot of pattern
correlation
> between the WRF forecast (9km) and the gridded observations (50km).
>
> will that be a simple correlation where first variable is RF (obs)
and RF
> FCST will be 2nd variabble.
>
> can u guide??
> geeta
>

------------------------------------------------