question

Sadegh Dalfardi sadegh.dalfardi at tvrl.lth.se
Fri Aug 13 10:50:35 CDT 2021


Hello

Thank you so much. I will give it a try

Best

________________________________
From: Latham, Robert J. <robl at mcs.anl.gov>
Sent: Friday, August 13, 2021 1:58:51 AM
To: parallel-netcdf at lists.mcs.anl.gov; Sadegh Dalfardi
Subject: Re: question

On Fri, 2021-08-13 at 12:04 +0000, Sadegh Dalfardi wrote:
> Hello
> Thanks for your answer.
> I mean i have do downloaded precipitation data for 20 years in GES
> DISK and for half hour time interval and space resolution of 0.1
> degree in netcdf format. now I need them in for instance 6 hour time
> interval and with resolution 1 km. How can i do it? please guide and
> help me to do that if it is possible.

Well, this is getting kind of off topic for a parallel-netcdf mailing
list...

It does sound like the NetCDF Operators (http://nco.sourceforge.net/)
are what you should apply here.  If you are new to NCO, there is a
lengthy user gudide http://nco.sourceforge.net/nco.html (also available
in PDF format)

I have seen some pretty elaborate NCO workflows, extracting variables,
averaging over time series, but I emphasise the "seen" part -- I do not
have a lot of first hand experience with NCO myself.

==rob


> Best
> Sadegh Dalfardi
> From: Latham, Robert J. <robl at mcs.anl.gov>
> Sent: Thursday, August 12, 2021 3:13:14 AM
> To: parallel-netcdf at lists.mcs.anl.gov; Sadegh Dalfardi
> Subject: Re: question
>
> On Thu, 2021-08-12 at 12:35 +0000, Sadegh Dalfardi wrote:
> > I have a netcdf file and want to change the spatial and temporal
> > scale whithout changing and conversion to another format. I want to
> > know is it possible? If so please help or guide me cause i am a
> > beginner.
>
> I'm not 100% clear on what you're asking, so please follow up if I'm
> missing the mark.
>
> A NetCDF dataset (which parallel-netcdf and Unidata NetCDF both can
> read and write) is fairly rigid in its structure.  If you have crated
> a
> few variables of a given dimension, the only way to change those
> variables and dimensions is to re-create the dataset.
>
> There are tools such as the "netcdf operators" (NCO) which make such
> manipulations easier, but if you have very large datasets it will
> take
> some time.
>
> If you have variables and you want them to grow over time then a
> "record variable" might suit your needs.  A typical use case would be
> a
> sensor array that is logging the environment.  Maybe the variable has
> dimenions for latitude and longidue for each sensor in the array and
> a
> third "UNLIMITED" dimension recoding the temperature every five
> minutes.
>
> ==rob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/parallel-netcdf/attachments/20210813/1be9af1f/attachment.html>


More information about the parallel-netcdf mailing list