Re: vort

From: Erik Kluzek (erik AT XXXXXX)
Date: Thu Sep 28 2000 - 10:21:04 MDT


On Thu, 28 Sep 2000, Benjamin Lamptey wrote:

> Hi,
> I am trying to calculate vorticity but my data conatins missing values.
> The function uv2vrG does not alway missing values. What do I do?
>

In similar situations, what I've done is to fill in the missing values with
a something (maybe weighted by inverse distance or nearest neighbor fill).
Then run "uv2vrG" and then after you are done, set the data points previously
missing and one grid point away to missing value (or possibly farther away
depending on how picky you are).

Just realize that the values you are filling in will corrupt the solution.
You might check sensitivity of the solution to what you plug in for the
missing values. If your solution doesn't change much regardless of what you
use to fill in -- your answer is stable and it can be trusted. If the solution
does change you might mask a larger area to missing value after you are done.

Erik Kluzek, (CGD at NCAR)
National Center for Atmospheric Research
Boulder CO, (off) (303)497-1326 (fax) (303)497-1324
--------- Home page and public PGP key---------------
          http://www.cgd.ucar.edu/~erik
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!



This archive was generated by hypermail 2b29 : Thu Sep 28 2000 - 10:29:30 MDT