GOES-R is GOES East Now?

[stormchasing101]

Hello,

I tried out the adde server for the real time GOES-R data, I noticed that the older GOES East server seems to use a TANC projection and the new one uses a ABIN projection. I admit I do not understand the difference but, it looks like it uses a much different angle on the data and doesn't quite extend as far north. Is this the nature of the imager or because it is not in the final location?

-- Carl

P.S I am interested in using the GLM netcdf file as well

I actually preferred the perfectly square TANC projection, I have been trying to duplicate it but the data is always canted and curved, is there a way I can import a projection from AWIPS II?

[beckys]

Hi Carl,

What server and dataset were you using for your GOES-13 data?

Becky

[stormchasing101]

The adde.ucar.edu site for the gini east dataset, uses the GOES 13 data and TANC projection, while the lead.unidata.ucar.edu and the RTGOESR dataset uses the new one. I am sure I will get used to the new angle as the quality of the GOES 16 is astounding.

[bobc]

Hi Carl -

P.S I am interested in using the GLM netcdf file as well

Here are some quick steps to working with local GLM netCDF files in McIDAS-V.

1. From the Main Display window, go to Tools>Plugins>Manage to open the Plugin Manager.
2. At the bottom of the Plugin Manager, click the green plus icon next to Data Sources>GLM NetCDF Files.
3. Restart McIDAS-V.
4. Go to the General>Files/Directories chooser and navigate to the directory with your data.
   *** Note that the files must be mission-standard format and follow the GRB naming convention, and thus look similar to OR_GLM-L2-LCFA_G16_s20172151749200_e20172151749400_c20172151749467.nc. ***
5. From the Plugin, under Data Type dropdown in the General>Files/Directories chooser, you will see three new data types dependent on what you are looking to plot:
   
   • Events: GLM Events Data files
   • Flashes: GLM Flashes Data files
   • Groups: GLM Groups Data files
     For this example, I'll use Groups.
6. Select the GLM Groups Data files Data Type.
7. At this point you can select 1 (20-second) file, or you can select multiple files at once to create a longer loop of data. If multiple files are selected, they will be merged into one data source. In my case, I'm selecting 30 files at once with Shift+left-click.
8. Click Add Source.
9. Each data point in every file technically has its own time step within the data. Because of this, if you try displaying the data as-is from the Field Selector, McIDAS-V will want to plot thousands of time steps. You can work around this by adjusting the time binning settings. To do this, from the Field Selector, right click on the Data Source on the left side of the tab (mine is called "30 files") and choose Properties.
10. At the bottom of the Properties tab of the Properties window, set the time binning as you desire. In my example here, I'll set Bin Size and Round To both to "2 minutes". This will allow for creating a loop of data where each timestep is 2 minutes, compressing all of the data points into fewer timesteps. Note that you will need to click on the little calendar icon for both Bin Size and Round To to manually enter 2 minutes for each of them. Click OK once you have set your time binning values.
11. From the Field Selector, choose the Point Data field and the Point Data Plot display.
12. If desired, subset a domain of data in the Region tab. Also, form the Layout Model tab, click the double-down blue arrow icon next to "None" and choose "GLM>Group"
13. Click Create Display
14. By default, McIDAS-V does some decluttering to most Point Data Plot displays, so not every point will be displayed until you zoom in far on the data. Optionally, you can disable the Declutter checkbox in the Layer Controls to plot all of the data points.

Here is a GLM animation I just created overlaid on a single time of ABI FD L1b Band 13 (10.3um) Temperature data.

GLM Animation

Please let us know if you have any questions, comments, or suggestions about plotting GLM data in McIDAS-V.

Thanks -
Bob Carp
McIDAS Help Desk

[stormchasing101]

Looks good, I just need a source for the data files....

[bobc]

Hi Carl -

I believe that the Unidata THREDDS group has a testing catalog that includes GLM data. I don't believe they've advertised it to the public, but you can check with them here:

https://www.unidata.ucar.edu/software/thredds/v4.5/tds/

There's a support email address on the right side of the page. I would suggest contacting them and asking for the catalog URL. Assuming they are able to give you a URL, you can paste it into the General>Catalogs chooser. From the General>Catalogs chooser, you have the Data Source Type dropdown available where you can select the GLM data source you want to work with. Aggregating files should work the same way as it does through the General>Files/Directories chooser.

If I become aware of a publicly available source of these GLM netCDF files I'll let you know.

Thanks -
Bob

[stormchasing101]

Thanks for the help I will certainly check it out!

[bobc]

Hi again Carl -

It looks like they did advertise the URL to the Python mailing list:

https://www.unidata.ucar.edu/mailing_lists/archives/python-users/2017/msg00016.html

The URL you will want to enter into the General>Catalogs chooser is:

http://thredds-test.unidata.ucar.edu/thredds/catalog/satellite/goes16/catalog.xml

After pasting this URL into the Catalogs chooser, press Enter and the Browse panel will update to show a directory listing from that URL. For GLM, expand "GOES-16 Products>GRB16>GLM>LCFA" and you'll directories from the last 10 days. You can work with this data the same way you do through the General>Files/Directories chooser as outlined in my post earlier.

Note that this catalog is maintained by Unidata. It also has the word "test" in the URL, so I don't know how stable it is (as far as downtime) or how long it will be around.

Thanks -
Bob

[stormchasing101]

Thanks for that link! It works great, I have one question, I am kind of fuzzy on RGB creations. I heard about something called the Day Cloud Phase RGB and I have a picture of it, and when I tried to match the channels I don't get the appropriate colors. The nature of it is to show glaciations with the white convective clouds turning green as they grow and glaciate. I have a screen capture can you guys attempt to replicate this rgb and supply any tips. Sorry if I am a pest...

[bobc]

Hello -

I wrote up a script to try creating this RGB display with:

Code: Select all

addeParms = dict(
    server = 'lead.unidata.ucar.edu',
    dataset = 'RTGOESR',
    descriptor = 'CONUS',
    day = '2017-12-26',
    time = '17:02:25',
    coordinateSystem = LATLON,
    location = (38, -100),
    place = CENTER,
    size = (800, 800),
    )

redData = loadADDEImage(band=13, mag=(1,1), **addeParms)
grnData = loadADDEImage(band=2, mag=(-4,-4), **addeParms)
bluData = loadADDEImage(band=5, mag=(-2,-2), **addeParms)

rgbData = mycombineRGB(redData, grnData, bluData)
layer = activeDisplay().createLayer('RGB Composite', rgbData)

The only somewhat tricky thing with this is specifying the mag values individually for each band since they are all different resolutions. The red band (band 13) is 2km resolution, the green band (band 2) is 0.5km resolution, and the blue band (band 5) is 1km resolution. It should be possible to use the Resample Grid formula/function to resample bands 2 and 5 to 2km resolution if you wish to avoid scripting.

I chose these bands by looking at this Satellite Liaison blog post:
https://satelliteliaisonblog.com/2017/11/10/using-goes-16-day-cloud-phase-distinction-rgb-to-analyze-complex-daytime-cloud-and-snow-cover-scenes/

The image generated from my script looks to be pretty similar to those on the blog linked above. I did modify the gamma value in the Layer Controls to 1.4 to make some of the colors stand out more.

Do you know the band combination used in the image you posted? Perhaps more was done to generate this display, such as subtracting bands or rescaling values to get the individual red, green, and blue components?

Thanks -
Bob