Hi Yilmaz :
On 3/11/2011 8:55 AM, Yilmaz Arslanoglu wrote:
> Hi;
>
> I was wondering the following points regarding dimensionless vertical
> coordinates:
>
> 1) Assume that we have a variable U (let's say ocean current eastward)
> and is defined on a dimensionless vertical coordinate (for example
> ocean-s coordinate).
> Do the variables indicated in "formula_terms" of this vertical
> coordinate need to share exactly the same longitude and latitude
> coordinates (possibly auxiliary) with
> the main variable U, or can they be defined over their own
> longitude and latitude coordinates independently?
I will give my opinion as someone who has written software to deal with
this: 1) the variables in the formula must be on the same coordinates as
the vertical coordinate, and 2) a vertical coordinate can only be used
by data variables on the same horiz coordinate.
this is a real problem with staggered grids, which i assume is your use
case. One needs an interpolation or regridding scheme to solve it. So
gridspec is one solution, though heavyweight. Rich SIgnell proposed
another "simple regridding" proposal which we should revisit.
> 2) Just given the "positive" attribute and the information whether the
> sigma coordinate is monotonically increasing or decreasing, is it
> possible to
> conclude that the array of z values z(sigma, t, x, y) are always
> increasing or decreasing? That is, is it possible to make the
> following conclusion
> for any dimensionless vertical coordinate ? ( sigma here is just
> an example )
>
> I was actually wondering what will the "positive" attribute
> exactly mean for a sigma coordinate.
My understanding is that positive, which means "Direction of increasing
vertical coordinate value" refers to the actual sigma coordinate value,
not any coordinates derived from it. So i think your table is correct,
since sigma = 1 is at the surface and =0 is at the top of the
atmosphere. Thus larger values of sigma go "down".
> Sigma Positive Z
> --------- ------------ --------------
> increasing down increasing
> increasing up decreasing
> decreasing down decreasing
> decreasing up increasing
>
>
>
> Best regards,
> Yilmaz
> _______________________________________________
> CF-metadata mailing list
> CF-metadata at cgd.ucar.edu
> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <
http://mailman.cgd.ucar.edu/pipermail/cf-metadata/attachments/20110311/f5cc0e77/attachment-0001.html>
Received on Fri Mar 11 2011 - 14:07:42 GMT
