[Thread Prev][Thread Next][Index]

Re: regridding issue



Hi ,
The default regridding that is used in

   let sst10=sst[G=xy10grid]

is a linear interpolation. @AVE regridding weights the
computation according to the portion of the grid cells lying in the
requested interval. Perhaps this FAQ will help to fill in the details.

http://www.ferret.noaa.gov/Ferret/FAQ/analysis/regridding_with_ave.html

Namba Takaya wrote:

>Dear Sir
>
>(very similar example in Ch. 4 of User Manual)
>If I have the SST defined on a 2x2 degree grid in long/lat
>like coads_climatology. Then if I wish to regrid to 10x10
>degree spacing in long/lat. In order to check the
>regridded result, I compared the data(sst10) regrided with
>the value(sst10aveX) averaged over area 10x10 as below.
>There is a little difference in value.
>
>Define axis/x=160:180:10/units=degrees x10deg
>Define axis/y=-35:-55:10/units=degrees y10deg
>Define grid/x=x10deg/y=x10deg xy10grid
>
>let sst10=sst[G=xy10grid]
>list sst10
>let sst10ave1=sst[x=160:170@ave,y=-35:-45@ave]
>let sst10ave2=sst[x=170:180@ave,y=-35:-45@ave]
>let sst10ave3=sst[x=160:170@ave,y=-45:-55@ave]
>let sst10ave4=sst[x=170:180@ave,y=-45:-55@ave]
>list sst10ave1,sst10ave2,sst10ave3,sst10ave4
>
>I think that  the range averaged over, e.g. 160:170,
>should be something like 160:169.9. Could you make me
>understand the reason for this difference?
>
>Best Regards,
>
>Takaya Namba
>
>
>
>
>__________________________________
>Let's Celebrate Together!
>Yahoo! JAPAN
>http://pr.mail.yahoo.co.jp/so2005/
>
>  
>

[Thread Prev][Thread Next][Index]

Dept of Commerce / NOAA / OAR / PMEL / TMAP

Contact Us | Privacy Policy | Disclaimer | Accessibility Statement