RBTOP and JSL Table in ascii format for quick use today?

Post ideas for new functionality you'd like to see in McIDAS-V or ideas for new tutorials.
Post Reply
User avatar
stormchasing101
Posts: 37
Joined: Sun Sep 06, 2015 6:24 am

RBTOP and JSL Table in ascii format for quick use today?

Post by stormchasing101 »

Hi,
I was wondering if someone better at creating color tables could take a crack at these two in ascii format. I was looking at the RBTOP and JSL images for Matthew on the NHC homepage and I noticed how nice the imagery looked. I tried finding the same color tables and couldnt find the jsl table or the RBTop one. I tried using MS paint and tryied to build the table that way, but they didn't turn out at all. I don't know what the start and end range is for the files. So I was wondering if someone better at building tables and perhps understanding the ranges could try building them for me . I will post attachments. Thank you kindly

--Carl
Attachments
JSL.PNG
RB TOP.PNG
User avatar
joleenf
Posts: 1123
Joined: Mon Jan 19, 2009 7:16 pm

Re: RBTOP and JSL Table in ascii format for quick use today?

Post by joleenf »

Hi,

I found the reference for JSL2.ET online at http://www.ospo.noaa.gov/data/FAQ/text/JSL.txt.

This table originally was stretched over a pixel brightness value from 0-255. If you display the 10.7 IR using Brightness as the display field, you should see the image you captured. However, if you are displaying the field using temperatures, modifications need to be made. If any Brightness space color table is already in McIDAS-V, it can be modified using a formula. I have done that for this color table and attached the modified color table for temperature (JSL2_TEMP.xml) that can be imported. This color table was created using a range of 163-330K. This the range which needs to be used if you want an exact reproduction of the original 0-255 brightness value color table.

Please see the attached color tables and if you are interested in an outline of the conversion code, it is posted at
http://mcidasv.ssec.wisc.edu/forums/vie ... ed3d5c3a74.

I could not find the RBTOP, but the Longwave Deep Infrared Convection table already packaged with McIDAS-V is very close (found under Satellite>Group 2>Longwave Deep Infrared Convection)

The inversion should also be stretch from 163-330K.

Joleen
Attachments
LongWaveDeepInfraredConvection_TEMP.xml
(5.99 KiB) Downloaded 532 times
JSL2_TEMP.xml
JLS2 modified for Temperature UNIT, stretch should be from 163K to 330 K.
(5.98 KiB) Downloaded 612 times
JSL2.ET
JSL2.ET from http://www.ospo.noaa.gov/data/FAQ/text/JSL.txt .

Scales and displays well when BRIT unit is used.
(3.19 KiB) Downloaded 390 times
User avatar
joleenf
Posts: 1123
Joined: Mon Jan 19, 2009 7:16 pm

Re: RBTOP and JSL Table in ascii format for quick use today?

Post by joleenf »

Sorry, I know this is late, but maybe for another storm. I found RBTOP online as well at

http://www.ospo.noaa.gov/Organization/F ... ments.html

Here is the temperature scaled version that I created:
rbtop_temperature.xml
Use on UNIT=TEMP (Temperature) Scaled from 163-330 K for traditional look.
(5.99 KiB) Downloaded 633 times


The original 0-255 BRIT space color table is
RBTOP.ET
Use on BRIT (Brightness) from 0-255.
(3.19 KiB) Downloaded 398 times


Here is an example:
rbtop_asTempmerature.jpg


Joleen
User avatar
stormchasing101
Posts: 37
Joined: Sun Sep 06, 2015 6:24 am

Re: RBTOP and JSL Table in ascii format for quick use today?

Post by stormchasing101 »

Thanks a lot for the tables, I can see I still have a lot to learn about building them :)

By the way how do I use the formula you linked to?
--Carl
User avatar
stormchasing101
Posts: 37
Joined: Sun Sep 06, 2015 6:24 am

Re: RBTOP and JSL Table in ascii format for quick use today?

Post by stormchasing101 »

I have tried to figure out how to stretch the tables like you mentioned, but I seem to be drawing a blank on that one. There are a lot of enhancements on this site. http://wx.db.erau.edu/faculty/mullerb/W ... h/enh.html They are all given in Celsius. Do you have a simple method for determining the breakpoints and values for the conversion of the temps to brightness breakpoints. I tried to stretch them to figure out a ratio, but I had some big trouble....

--Carl
User avatar
joleenf
Posts: 1123
Joined: Mon Jan 19, 2009 7:16 pm

Re: RBTOP and JSL Table in ascii format for quick use today?

Post by joleenf »

Hi Carl,

There is not a straightforward answer to setting breakpoints for color enhancements. What is chosen depends on the satellite channel and the important features in the image. The COMET courses covering GOES channels at MetEd may be interesting to you. There are a few that cover the response of satellite channels to various features and also RGB tutorials. They are free and very useful.

First, the color tables RBTOP and JSL were posted in a format compatible with an old (still in use) software package for satellite imagery (McIDAS-X). In this software package, brightness values were scaled to temperature, and a bi-linear enhancement was usually applied.

Reading the documentation, the OSPO color tables you posted (JSL and RBTOP) are all using this bi-linear enhancement. In my code, the equation is written from temperature to brightness value so that I can pick out the color level I want and reorganize the color table according to temperature scaling. The OSPO website lists the conversion from brightness value to temperature. However, it is the same formula (see also: http://www.goes.noaa.gov/ECWV1.html).

Therefore, the enhancement is a linear relationship from Brightness Value to Temperature for Brightness Values greater than 176:

If B > 176, T = 418 - B

The enhancement has a bi-linear relationship from Brightness Value to Temperature for Brightness Values less than or equal to 176:

If B <= 176, T = 330 - (B/2)

On the other hand, the Fog Depth Enhancement that is on Gary Ellrod's GOES Enhancement Table page was using a stretch table, which apply a linear scaling of a temperature range (-13 to >6) to a brightness value (0 to 255).

To achieve the same color table without code, you could use the interactive color table editor in McIDAS-V (find: Color Table Editor documentation in the User’s Guide
In addition, if you have not yet worked through the Satellite Imagery Tutorial, you could work through an example with the Problem Set #1 Solution to gain a better understanding of the process.

I will provide an outline of how I approached the problem:

1.) set the number of colors of the color table to 257. Do this by clicking the + arrows above the color bar in the color bar editor.

2.) Click File->Save As-> to create a new color table name.

3.) Set the range to -13, 7

4.) Set breakpoints at the breakpoints listed in the table. In the case of the fog enhancement it is -13, -3, 2, 3, 4, 5, 6, 7. (To set breakpoints, right click above color table in color table editor and select Add BreakpointAt Data Point

5.) Click on each breakpoint and set the color of those breakpoints from the table. Note: Gary lists these breakpoints as Blue, Green, Red, but are Red, Green, Blue from top to bottom in the color table editor. Work in one direction. Once the breakpoint color is set, use the right-click Edit Colors to fill either right or left to next breakpoint

6.) Continue to next breakpoint and follow the same procedure as was done at the previous breakpoint.

7.) When working from -13, 3, select interpolate rather than fill.

Give it a try. If it does not work for you, post again and let use know where you are struggling. You should be able to replicate a linear color table fairly well with the color table editor.

Here is a photo of what you completed color table would look like when it is done

FOGENH_TEMP.png


This reproduces Gary's original example fairly well

fog_enh.png


I believe once you learn this procedure, you could reproduce most if not all the enhancements on Gary's page.

Hope this helps!
Joleen
User avatar
stormchasing101
Posts: 37
Joined: Sun Sep 06, 2015 6:24 am

Re: RBTOP and JSL Table in ascii format for quick use today?

Post by stormchasing101 »

Hi,

Thanks for the continued help. I guess part of my problem with the tables on Gary's site was trying to modify the temperature scales to the 0-255 brightness range, not the other way around. I am surprised at how intricate and important color tables really are for proper data interpretation. I will work with your suggestions and let you know how they turn out.

--Carl
User avatar
stormchasing101
Posts: 37
Joined: Sun Sep 06, 2015 6:24 am

Re: RBTOP and JSL Table in ascii format for quick use today?

Post by stormchasing101 »

Hi,
Well, the brightness tables turned out real well, but I am not able to run your formula in McIDAS-V to stretch a table to the 330-163k ratio. I will post the file; maybe you could convert it with McIDAS-X for me? Its a better version of the Dvorak table. I got the brightness version perfect. It just won't look right when I use a temperature display. Thanks, also can I just use the simple a-b difference to create the fog enhancement image?

--Carl
Attachments
DVK.xml
(11.28 KiB) Downloaded 423 times
User avatar
joleenf
Posts: 1123
Joined: Mon Jan 19, 2009 7:16 pm

Re: RBTOP and JSL Table in ascii format for quick use today?

Post by joleenf »

Hi Carl,

The Basic Dvorak Infrared Color Table was already located in the Satellite>Group 1>Tropical Cyclone tree. There may be a couple other color tables there that interest you.

The converted color table is at

ftp://ftp.ssec.wisc.edu/pub/ssec/joleenf/DVK_TEMP.ascii

Joleen
Post Reply