| EOF Arctic Oscillation for beginner [message #90535] |
Fri, 06 March 2015 14:03  |
siumtesfai
Messages: 62
Registered: April 2013
|
Member |
|
|
Hello All,
I have difficulty in reproducing AO regression map
( http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao _index/ao.loading.shtml)
I am working with the code (http://www.idlcoyote.com/code_tips/eof_analysis.html)
According the website it says
Note that my latitude values go from 47.5 to 90.0 degrees of latitude, in 2.5 degree increments. I am going to subtract 1.75 degrees from each of these latitude values, in order to locate the latitudes in the center of the grid, and to avoid multiplying by zero for all the latitudes at 90 degrees.
so dlat/2. should be 1.75
dims = Size(air_temp, /Dimensions)
nlon = dims[0] & nlat = dims[1] & ntime = dims[2]
lon = Reform(lon_ncep[*,0]) & dlon = Abs(lon[1]-lon[0])
lat = Reform(lat_ncep[0,*]) & dlat = Abs(lat[1]-lat[0])
weights = Sqrt(Cos((lat_ncep - dlat/2.) * !DtoR))
FOR j=0,ntime-1 DO air_temp[*,*,j] = air_temp[*,*,j] * weights
any suggestion
Best regards
|
|
|
|
|
|
| Re: EOF Arctic Oscillation for beginner [message #90537 is a reply to message #90536] |
Fri, 06 March 2015 14:37  |
siumtesfai
Messages: 62
Registered: April 2013
|
Member |
|
|
On Friday, March 6, 2015 at 5:19:35 PM UTC-5, David Fanning wrote:
> siumtesfai@gmail.com writes:
>
>> According the website it says
>> Note that my latitude values go from 47.5 to 90.0 degrees of latitude, in 2.5 degree increments. I am going to subtract 1.75 degrees from each of these latitude values, in order to locate the latitudes in the center of the grid, and to avoid multiplying by zero for all the latitudes at 90 degrees.
>>
>> so dlat/2. should be 1.75
>
> I think 1.25 probably makes more sense. :-)
>
> Cheers,
>
> David
> --
> David Fanning, Ph.D.
> Fanning Software Consulting, Inc.
> Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
> Sepore ma de ni thue. ("Perhaps thou speakest truth.")
Thanks
I have used geopotential height at 1000hPa from NCEP/NCAR
I looked at JFM average , and other seasons.
The PC1 looks similar to their website ( http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao _index/JFM_season_ao_index.shtml)
However, when I regress the EOF with PC1 , I do not see dipole structure.
Did I get the method right or somethings is wrong which I do not understand ?
My final answer should be similar to the figure from NOAA website
( http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao _index/ao.loading.shtml)
Thanks for your help
|
|
|
|
| Re: EOF Arctic Oscillation for beginner [message #90538 is a reply to message #90537] |
Fri, 06 March 2015 14:58 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
siumtesfai@gmail.com writes:
> I have used geopotential height at 1000hPa from NCEP/NCAR
> I looked at JFM average , and other seasons.
> The PC1 looks similar to their website ( http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao _index/JFM_season_ao_index.shtml)
>
> However, when I regress the EOF with PC1 , I do not see dipole structure.
>
> Did I get the method right or somethings is wrong which I do not understand ?
>
> My final answer should be similar to the figure from NOAA website
> ( http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao _index/ao.loading.shtml)
I really couldn't tell you. I did that work a long time ago. I don't
remember much about it, except that I spent weeks working it out. It
certainly seemed to produce correct answers (at least consistent with
other methods I tested) when I was doing that particular study.
But, I'm retired now, and even thinking about EOF analysis gives me a
headache. :-(
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
Sepore ma de ni thue. ("Perhaps thou speakest truth.")
|
|
|
|
| Re: EOF Arctic Oscillation for beginner [message #90539 is a reply to message #90538] |
Fri, 06 March 2015 15:00 |
siumtesfai
Messages: 62
Registered: April 2013
|
Member |
|
|
On Friday, March 6, 2015 at 5:58:06 PM UTC-5, David Fanning wrote:
> siumtesfai@gmail.com writes:
>
>> I have used geopotential height at 1000hPa from NCEP/NCAR
>> I looked at JFM average , and other seasons.
>> The PC1 looks similar to their website ( http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao _index/JFM_season_ao_index.shtml)
>>
>> However, when I regress the EOF with PC1 , I do not see dipole structure.
>>
>> Did I get the method right or somethings is wrong which I do not understand ?
>>
>> My final answer should be similar to the figure from NOAA website
>> ( http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao _index/ao.loading.shtml)
>
> I really couldn't tell you. I did that work a long time ago. I don't
> remember much about it, except that I spent weeks working it out. It
> certainly seemed to produce correct answers (at least consistent with
> other methods I tested) when I was doing that particular study.
>
> But, I'm retired now, and even thinking about EOF analysis gives me a
> headache. :-(
>
> Cheers,
>
> David
> --
> David Fanning, Ph.D.
> Fanning Software Consulting, Inc.
> Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
> Sepore ma de ni thue. ("Perhaps thou speakest truth.")
Oh Sorry . Thank you . You did help me
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
