Re: [ferret_users] regarding eof analysis

[Russ Fiedler <russell.fiedler@xxxxxxxx>]

Hi,

By the looks you are trying to pad the ends with zeros. The easiest way to do this is using the MISSING function setting missing values to zero.

let sst_n=missing(sst_ff,0)

or

let sst_n=missing(sst_ff,0) + sst_f   ! This will preserve missing values of the original.


should do the trick

Cheers,
Russ

On 21/09/16 17:32, saurabh rathore wrote:

Hi,

as mentioned in previous mail i list my filtered data at x=85,y=16 i found that the data is missing from 1-jan-1901:1-dec-1904 in the beginning and from 1-jul-2012:end. so i tried to plot eof between this range only but still having the same bailing out error moreover i tried to pad the missing values after filtration but didn't get any success in pading also.

i am attaching script for padding please have a look and rectify my mistake in performing eof of filtered data.

regards, saurabh 

On Wed, Sep 21, 2016 at 11:03 AM, Russ Fiedler <russell.fiedler@xxxxxxxx> wrote:

Hi,

Most likely the end points of the time series are being removed by the filter thus resulting in an incomplete time series. Have a look at the listing at a single point. The first and last (N-1)/2 (where N is the value of filter_span) should be missing, I think.

You'll need to figure out a way to pad the endpoints sensibly or only apply the EOF to the  valid points. L=@FLN only interpolates between points so it's of no use here.

Russ

On 21/09/16 14:47, saurabh rathore wrote:

Hi all,

i am having a problem regarding the eof analysis. i am using HadiSST monthly data from 1901-2016. following steps involve in my analysis.

1) making monthly anomaly

2) performing LSL_LowPASS filter of 96 months

3) using step 2 filter SST anomaly for eof analysis.

but in eof analysis i am getting an error

Bailing out of external function "eofsvd_space":
    There are no spatial locations having complete time series.
 **ERROR: error in external function

​i tried by filling the gaps also by L=@FLN also but still the error persists. Moreover its a grided data from model so there should be no discontinuity but after making the filtered anomaly this is happening. so please help me in rectifying this.

regards, saurabh

--

REGARDS

Saurabh Rathore

Research Scholar (PhD.)

Centre For Oceans, Rivers, Atmosphere & Land Science Technology

Indian Institute Of Technology, Kharagpur

contact :- 91- 8345984434

--

REGARDS

Saurabh Rathore

Research Scholar (PhD.)

Centre For Oceans, Rivers, Atmosphere & Land Science Technology

Indian Institute Of Technology, Kharagpur

contact :- 91- 8345984434