[Thread Prev][Thread Next][Index]

Re: [ferret_users] *urgent help* needed: loading opendap data via ferret



Hi,

Yes there are nco operators that you could use to reduce the number of dimensions to 4. See information on the NCO homepage: http://nco.sourceforge.net/nco.html 

Without looking into the detail of the particular file, I would do something like this:

1. Extract the ensemble number that you would like using ncks (http://nco.sourceforge.net/nco.html#ncks-netCDF-Kitchen-Sink), e.g.

ncks -d ensemble,0 my_OPeNDAP_file.nc my_data.nc

2. I think this will leave a redundant ensemble dimension (of 1), which can be "averaged away" using ncwa (http://nco.sourceforge.net/nco.html#ncwa-netCDF-Weighted-Averager), e.g.

ncwa -a ensemble my_data.nc my_4D_data.nc

Good luck!

Paul

On Mar 11, 2009, at 1:15 PM, Ansley Manke wrote:

Hi,
As you say, currently Ferret has just 4 dimensions, although we are aware that datasets like this are becoming more common, and have plans to add more dimensions to Ferret to allow for more dimensions.  Unfortunately Ferret does not have a way to skip a dimension when opening a dataset. You would need something like USE/ORDER=xyz_t but we don't have such a syntax.

I had the idea that at least you could use the browser interface and download the data from  http://ensembles.ecmwf.int/thredds/dodsC/ocean/ecmwf/eg/thetao.html
I tried choosing all the variables, with for the ensemble coordinate choose only one step (0:1:0) to pick the first one. It turns out that only the coordinates are returned.  Interestingly on that form,the coordinate variable ensemble is not listed among the variables, even though the other coordinates Time Longitude, Latitude, Depth are listed.

Does anyone else have other workarounds for accessing data with more than 4 dimensions for use in Ferret?  There may be  nco operators that could be used to create a file with just the 4 dimensions you want.

Ansley

Haoping Yang wrote:
I am having trouble to load temperature data from some OPeNDAP servers such as http://ensembles.ecmwf.int/thredds/dodsC/ocean/ecmwf/eg/thetao.html. From the web browser, one can open this link and see the data structure. The variable 'thetao' has 5 dimensions: *[time = 0..575]*[ensemble = 0..4][level = 0..32][latitude = 0..178][longitude = 0..359]. The dimension 'ensemble' is useless to me. When I try to load the data via ferret, ferret uses 'ensemble=0..4' as the time dimension and no temperature data can be loaded. Ferret assumes the data has four dimensions instead of five. I don't know how to remove the  dimension 'ensemble' and let 'time' to point to the axis 'T' (index 'L'). There must be some tricks to load this data. I guess it might be something like "yes? use/ORDER=.... URL" and could be very easy for some ferret experts. Please help me. Thanks

--------------------------------------------------------------------------------------------------------------------------
yes? use "http://ensembles.ecmwf.int/thredds/dodsC/ocean/ecmwf/eg/thetao"
yes?
yes? show data
    currently SET data sets:
   1> http://ensembles.ecmwf.int/thredds/dodsC/ocean/ecmwf/eg/thetao  (default)
         name     title                                                      I             J           K          L
INSTITUTION
         Institution responsible for the                           ...            ...          ...         ...
EXPERIMENT_ID
         Experiment identifier                                        ...            ...          ...         ...
REALIZATION
         Number of the simulation in the                       1:5          ...          ...         ...
SOURCE   Method of production of the dat              ...           ...          ...         ...
THETAO   Potential temperature ref to su              1:360     1:179     1:33      1:5

yes?
yes? show grid THETAO
   GRID GRM3
name                      axis                           # pts                   start                end
LONGITUDE       LONGITUDE          360mr                 0E                   1W
latitude                   LATITUDE              179 r                  89S                  89N
level                       DEPTH (m)               33 i-                     0                    5500
ENSEMBLE         T                                5 r                       1                    5
yes?
yes?
----------------------------------------------------------------------------------------------------------------------
*** *


-----
Paul Young

Chemistry and Climate Processes
Chemical Sciences Division
NOAA Earth System Research Laboratory
325 Broadway R/CSD8
Boulder CO 80305-3328
USA

Tel:   +1 303-497-4711
Fax:   +1 303-497-5686


[Thread Prev][Thread Next][Index]

Contact Us
Dept of Commerce / NOAA / OAR / PMEL / TMAP

Privacy Policy | Disclaimer | Accessibility Statement