subsetting not working correctly, polar orbiting swath

[kbedka]

I have attached a couple graphics below. I am dealing with derived product data in NetCDF from polar orbiting satellite swath. You can see that I've requested data from only over polar regions. (it would be optimal to see a globe here in the sub setter, but the box across the entire top of the flat map will do). I get data from the entire globe when I create the display which shouldn't be happening.

[bobc]

Hi Kris -

I think the problem of McV displaying the full domain of the data instead of what was subsetted in the Region tab is related to Inquiry 1128. I have a few different NSS.GHRR* files that I've run tests on, and one of them allows me to subset over the North Pole correctly while the others don't. I don't have the exact file that you're working with, and none of my files have the 'NOAA Climate Data Record*' fields in the Field Selector.

Could you please post your file to our anonymous ftp:
ftp ftp.ssec.wisc.edu
anonymous (for username)
email address (for password)
cd pub/incoming
bin
put file_name

We can then take a look at the file and verify that you're seeing the problem written up in Inquiry 1128.

As for adding the functionality of changing the Region tab of the Field Selector to use a globe instead of a 2D map, I will write up a new inquiry.

Thanks -
Bob Carp

[kbedka]

I've redone some of the metadata recently, the file structure between your files and this one is the same regardless if it says NOAA Climate Data Record. It sounds like you're able to reproduce the problem that I'm seeing so if all those files work, then it is highly likely that any of them will work.

I see that this was noted to be a problem in 2011 and we're still dealing with it in 2014. Is there any way we can bring this to closure as this is still adversely impacting our work?

[jayh]

Hi Kris-

I alerted Tom to this forum and the continuing issues. He is going to go over his posts on the forum with this topic and put something together.

Thanks, Jay

[kbedka1]

Instead of working with a full orbit of the earth, we have been focused on analysis of a small segment of NetCDF data that crosses the poles. In order to probe the data with apple-left click, past forum posts have suggested that we need to draw a subset box around our region of interest. The attached screenshot shows that since the orbit crosses the poles , we have to draw a box across the whole globe. Our segment of data consists of 2 NetCDF files. One of the files displays just fine when subsetted. The other only shows up as a small box (see attached subsetted.jpg). If I click "Use default" in the lower right of the data explorer, then I can get the whole file to display fine. But then I cannot probe the data that is displayed.

It was my understanding that past challenges we faced with this data were caused by the fact that it was a full orbit around the Earth. What I'm displaying here is simply lat,lon and a 2-D data array over a small region. I'd imagine that others will run into the same type of issue when trying to display and interrogate NetCDF-based data.

[ghansham]

Hi Bob Sir
Can you post one NSS.GHRR file on ftp server?
Hi Kris Sir
Can you place the file for which you have attched
images in the last post?
I have some idea about why this is happenimg and how
it can be handled?
regards
Ghansham

[bobc]

Hi Ghansham -

I put NSS.GHRR.NN.D08290.S1045.E1235_NASA_LaRC_cloud_properties.nc in /pub/incoming:

ftp.ssec.wisc.edu/pub/incoming/NSS.GHRR.NN.D08290.S1045.E1235_NASA_LaRC_cloud_properties.nc

I ran tests with this file in 1.4 as well as our nightly and attempting to subset a region over the entire North Pole (as done in post 1 of this forum) results in data being displayed over the entire globe.

Thanks -
Bob

[ghansham]

Thanks Sir
The issue traces back to ncj but I will try to see
if it can be handled in a different way way.
regards
ghansham

[kbedka1]

I have put the following files on the ssec ftp server at pub/incoming

D08121.S1004.E1151_00653_01000.nc
D08121.S1004.E1151_00001_00652.nc

The D08121.S1004.E1151_00653_01000.nc is the one that is not subsetting properly.

[ghansham]

Thanks kris sir.
I will get back to you on this.
regards
Ghansham