Re: QuikScat HDF reading problem?

Hello,

I am appending a modified script. The output looks ok to me.

The reason for so many "little holes" is that you chose
a color map ["testcmap"] that has white right in the
middle of the data. This lead to both valid values
and _FillValue [missing values] to be white [transparent].

It does look like their are more missing values at higher latitudes.

I also modified the ascii data you were outputting.

good luck
D

On Sun, 15 Jan 2006, David Atkinson wrote:

> Hi, I am trying to read an HDF4 file with QuikScat marine winds.
> I have a reader that largely works and seems to plot what is a
> reasonable result, that is, winds over the ocean and not over
> the land. However the windfield data are interspersed with little
> holes where the wind value drops to zero. When plotted, there
> is a certain pattern to the drop outs that makes it look like some
> sort of offset problem in lat/long, however the land boundaries
> are correct; these would also be messed up if there were an
> offset. I'll check with JPL as well, because it could be a Quikscat
> or HDF writing issue from their end, but in the chance it is an
> ncl issue, I am posting it here to see if anyone has an idea or if
> they have worked with this format before and have seen this.
>
> The ncl program to read and subset the data is here:
> http://www.frontier.iarc.uaf.edu/~datkinson/QuikScat_reading/read_HDF_quikscat.ncl
>
> and a test data file from JPL is here:
> http://www.frontier.iarc.uaf.edu/~datkinson/QuikScat_reading/QS_XWGRD3_2003120.20031231631.hdf
>
> Thanks for any ideas,

_______________________________________________
ncl-talk mailing list
ncl-talk_at_ucar.edu
http://mailman.ucar.edu/mailman/listinfo/ncl-talk