Re: [ferret_users] concatenation problem

[Ansley Manke <ansley.b.manke@xxxxxxxx>]

Hi,
Ferret writes irregularly-spaced axes with upper and lower grid-cell boundaries. See the description of this in the CF standard,
http://cf-pcmdi.llnl.gov/documents/cf-conventions/1.6/cf-conventions.html#cell-boundaries

It looks as if the last upper bound from one dataset is earlier than the first lower bound from the next dataset.  Ferret does not allow for discontinuous bounds on coordinate axes, so it puts in the extra point. 

A solution might be to define the full destination axis, and regrid each 10-year section to that axis before writing. You'd need to specify the time range on each SAVE command.

Ansley

On 4/30/2013 8:31 PM, FISSEHA G. BERHANE wrote:

Hello dear all,

I have monthly precipitation data. I have seven files from 1940-2010, each covering ten years. The total number of months is 7*120=840. But when I concatenate all the files, I get L= 846 instead of 840; also, I have observed that it repeats some months and gives a missing value to one of them as shown below.

16-DEC-2000 00 / 725:   68.5
 31-DEC-2000 12 / 726:   ....
 16-JAN-2001 00 / 727:   72.1

Your help on how to solve this case is highly appreciated

Thank you,
Fisseha