Can not run geocat 1.0b

luthfi_a
Posts: 17
Joined: Thu May 21, 2020 5:08 pm

Re: Can not run geocat 1.0b

Post by luthfi_a »

Hi Geoff,

I ran the command and put the files in ftp named as luthfi_xstride1.tar.gz. There were no nc files created. I collected geocat_hdf in the same tar file.

I also ran the same command with only one different parameter: --xstride 10. I put all the files created in luthfi_xstride10.tar.gz.

As I have asked questions in previous post, I would like to know whether xstride 10 decrease the spatial resolution. I also would like to know the cloud type information from level 2 product.

Thank you and best regards,
Luthfi
geoffc
Posts: 34
Joined: Mon Feb 14, 2011 3:02 pm
Location: Madison, WI
Contact:

Re: Can not run geocat 1.0b

Post by geoffc »

Hi Luthfi,
yes, the "xstride" switch decreases the spatial resolution in the "x" direction, or along the scans. So if you gave "--xstride1=10" on a Mesoscale (*R301*.DAT) image (500x500), geocat samples every 10th pixel in the x direction, so the output files would be of size (500x50).

I'm not quite sure what you are asking about the level-2 cloud type, but in the level-2 output there are the datasets:
  • "enterprise_cldphase_10_11_13_14_15_cloud_phase" (with types [clear, water, supercooled, mixed, ice, unknown])
    "enterprise_cldphase_10_11_13_14_15_cloud_type" (with types [clear, spare, water, SC, Mixed, Thick_Ice, Thin_Ice, Multilay, Spare, Uncertain])
I do not have any information about the level-2 products, but I will ask around.

Geoff
Geoff P. Cureton, PhD
Cooperative Institute for Meteorological Satellite Studies
University of Wisconsin-Madison
1225 W. Dayton St.
Madison WI 53706, USA
Phone: +1 608 890 0706
luthfi_a
Posts: 17
Joined: Thu May 21, 2020 5:08 pm

Re: Can not run geocat 1.0b

Post by luthfi_a »

Thank you for your explaination about xstride.
About my question on level 2 product, I found several pixel values like 0, 1, 2, 3 in eps_cmask_ahi_cloud_mask but I don't know the meaning. In enterprise_cldphase_10_11_13_14_15_cloud_type, some of the values are 0, 2, 4, 5, 6, 7. If there is a look up table defining the meaning of each pixel value, that would be very helpful. I am planning to do experiments with cloud products.

Thank you and best regards,
Luthfi
luthfi_a
Posts: 17
Joined: Thu May 21, 2020 5:08 pm

Re: Can not run geocat 1.0b

Post by luthfi_a »

Hi Geoff, how do you calculate the coordinate of each pixel? I read some documents from jaxa and chiba, one say that the image covers 80E-160W and the other say that the coverage is 85E-155W.

About viewport_xy parameter, I am still confused with y argument. We've used viewport_xy 2000 3500 800 3000 in our command. To calculate y coordinates, shall we start from the bottom (60S) or from top (60N)?

Thank you and best regards,
Luthfi
geoffc
Posts: 34
Joined: Mon Feb 14, 2011 3:02 pm
Location: Madison, WI
Contact:

Re: Can not run geocat 1.0b

Post by geoffc »

Hi Luthfi,
if you look at the attributes of the level-2 file (but doing "ncdump -h geocatL2.HIMAWARI-8.2019085.213000.FLDK.R20.nc", or using panoply), you will see for the cloud mask and the cloud type:

Code: Select all

        byte eps_cmask_ahi_cloud_mask(lines, elements) ;
                eps_cmask_ahi_cloud_mask:_FillValue = -128b ;
                eps_cmask_ahi_cloud_mask:reference = "goes-r awg cloud team" ;
                eps_cmask_ahi_cloud_mask:algorithm_index = 9LL ;
                eps_cmask_ahi_cloud_mask:flag_meanings = "confident_cloudy probably_cloudy probably_clear confident_clear" ;
                eps_cmask_ahi_cloud_mask:algorithm_name = "enterprise ahi cloud mask" ;
                eps_cmask_ahi_cloud_mask:flag_values = 0LL, 1LL, 2LL, 3LL ;
                eps_cmask_ahi_cloud_mask:units = "1" ;
                eps_cmask_ahi_cloud_mask:algorithm_version = "none" ;

        byte enterprise_cldphase_10_11_13_14_15_cloud_type(lines, elements) ;
                enterprise_cldphase_10_11_13_14_15_cloud_type:_FillValue = -128b ;
                enterprise_cldphase_10_11_13_14_15_cloud_type:reference = "GOES-R AWG Cloud Team - Pavolonis" ;
                enterprise_cldphase_10_11_13_14_15_cloud_type:algorithm_index = 18LL ;
                enterprise_cldphase_10_11_13_14_15_cloud_type:flag_meanings = "clear spare water SC Mixed Thick_Ice Thin_Ice Multilay Spare Uncertain" ;
                enterprise_cldphase_10_11_13_14_15_cloud_type:algorithm_name = "Enterprise Cloud Phase - 7.4, 8.5, 10, 11, 12 um" ;
                enterprise_cldphase_10_11_13_14_15_cloud_type:flag_values = 0LL, 1LL, 2LL, 3LL, 4LL, 5LL, 6LL, 7LL, 8LL, 9LL ;
                enterprise_cldphase_10_11_13_14_15_cloud_type:units = "1" ;
                enterprise_cldphase_10_11_13_14_15_cloud_type:algorithm_version = "none" ;
The attributes "flag_values" and "flag_meanings" give the possible dataset values and their meanings respectively.

The coordinates used for viewport_xy are coordinates for the image, not on the Earth itself. So for the FLDK data the coordinate (0,0) would be off the full disk, in the north-western corner. Increasing "x" moves you East, and increaasing "y" moves you south.

However, I'm not sure that "xstride" will give you the results you are expecting (it might not be working as it is supposed to, I'll have to check). It would be better to not use the xstride switch, and use the viewport_xy option to run a reduced portion of the full disk for now.

Geoff
Geoff P. Cureton, PhD
Cooperative Institute for Meteorological Satellite Studies
University of Wisconsin-Madison
1225 W. Dayton St.
Madison WI 53706, USA
Phone: +1 608 890 0706
luthfi_a
Posts: 17
Joined: Thu May 21, 2020 5:08 pm

Re: Can not run geocat 1.0b

Post by luthfi_a »

Great! Thank you very much Geoff.
I will learn the information in the header.

I prefer to use viewport_xy like you said, rather than using xstride. It seems clear for me now how to use geocat.

Thank you and best regards,
Luthfi
Post Reply